See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/340681981

Sport Science: Current and Future Trends for Performance Optimization

Book · January 2018

CITATIONS READS

5 5,019

3 authors:

Pedro Morouço Hideki Takagi

Polytechnic of Leiria University of Tsukuba
262 PUBLICATIONS 2,433 CITATIONS 180 PUBLICATIONS 1,504 CITATIONS

SEE PROFILE SEE PROFILE

Ricardo Fernandes
University of Porto
550 PUBLICATIONS 6,095 CITATIONS

SEE PROFILE

All content following this page was uploaded by Pedro Morouço on 08 July 2020.

The user has requested enhancement of the downloaded file.

Datasheet
Title
Sport Science: Current and Future Trends for Performance Optimization

Editors
Pedro Morouço
Hideki Takagi
Ricardo J. Fernandes

Edition
Escola Superior de Educação e Ciências Sociais — Instituto Politécnico de Leiria
Centro para o Desenvolvimento Rápido e Sustentado do Produto

Graphics and Composition

Rui Lobo

ISBN
978-989-8797-19-3

Legal Deposit
xxxx

© 2018 · ESECS/Instituto Politécnico de Leiria

Index

3 Datasheet
7 Preface

10 Lost in translation – getting your sport science research message across

David B. Pyne and Naroa Etxebarria

24 Periodization in sport training: traditional, blocks and polarized

Jose Maria González Ravé

51 Development of talent of adolescents in Australian sports high schools

Ross Sanders, Rhonda Orr, Chuang-Yuan Chiu, Angus Chard, Claire Tompsett,
Jennifer Fleeton, Timothy Davies, Alycia Fong Yan

77 Research into optimising performance from a multidisciplinary team: the case

of the IGOID research group
Leonor Gallardo, Enrique Colino, José Luis Felipe, Jorge García-Unanue, Manuel
León, Jorge López-Fernández, Javier Sánchez-Sánchez and Esther Ubago-Guisado

100 Propulsive forces and their efficient application during swimming

performance
Vassilios Gourgoulis

121 Hyperthermia and dehydration in competitive swimmers: current aspects and

recommendations
Orlando Laitano, Lara Elena Gomes, Alex J. Mattingly and Flávio Antônio de Souza Castro

136 Biomechanics of relay swimming starts: state of the art

K. de Jesus, K. de Jesus, L. Mourão, L. S. Coelho, L. Machado, R.J. Fernandes, J.P.
Vilas-Boas

146 Competition analysis in swimming: inter- and intra-individual variability of

speed and stroking parameters management
David Simbaña-Escobar, Philippe Hellard, Marta Avalos, Ludovic Seifert

168 Acute effects of various post-activation potentiation protocols on 15-m and

25-m freestyle swimming
Daniel Juárez Santos-García, Enrique Alonso Plaza and José María González Ravé

176 Altitude training for sea level performance: a systematic review

Ferran A. Rodríguez and Sergi Àvila

196 Dietary nitrate effects in human athletic performance

Maria Silva and Mónica Sousa

213 Exercise periodization and Taleb’s antifragility

José Afonso, Ivo Hilvoorde, Niek Pot, Alexandre Medeiros and Len Almond

3
225 Effects of different rest intervals between potentiation exercises on sprint
performance in trained soccer players
Vasiliki Manou, Vasilis Minos, Athanasios A. Dalamitros

237 Swedish tennis test: a field test to estimate maximal oxygen uptake in tennis
players
Pierre-Marie Leprêtre, Philippe Lopes and Pascal Sirvent

250 How Down syndrome impacts on the athletes’ sports performance?

Tânia Bastos, Ana Querido, Ariel Villagra, José Irineu Gorla, Ricardo J. Fernandes, Rui
Corredeira

271 Aerobic Interval Versus Continuous Training at Low Volume and Vigorous
Intensity
Laura Castro De Garay, Tony Meireles Dos Santos, Matthew Rhea, Victor Machado Reis

285 Comparison of energy expenditure during self-selected walking in field tracks

and in treadmill at various grades
João Paulo Brito, Ana Conceição, Félix Romero, Adenilson Targino de Araújo Junior,
Victor Machado Reis

295 Optimization as adaptability: how movement variability supports

performance in striking actions used in combat sports
Dominic Orth, John van der Kamp and Robert Rein

317 The use of power output in road cycling to optimize training and establish
performance
Alejandro Javaloyes, José Manuel Sarabia and Manuel Moya-Ramón

329 Characteristics of open water lifesaving

J. Arturo Abraldes and Ricardo J. Fernandes

347 The importance of perceptual-cognitive skills in water polo players

Filipe Casanova, Fernando Tavares, Sofia Canossa, Maickel Bach Padilha and Ricardo
J. Fernandes

4
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Preface

To improve athletes’ performance, a massive amount of effort by

themselves and coaches is obviously essential. However, one cannot
ignore the impact of that science has nowadays on their performance
improvement. In swimming, for example, swimsuits developed based
on state-of-the-art knowledge in fluid dynamics between 2007 and 2009
(which were all banned due to changes in FINA rules and regulations in
2009) allowed many swimmers to break world records, some of which
are still unbroken in 2018.

Methods used to assess athletes’ physical and technical perfor-

mances are developing every single day, and new training theories are
provided with remarkable speed. Ideally, it would be of great help for
athletes if coaches knew all updated methods and knowledge to pro-
vide the most effective training processes. However, this is quite difficult
in reality and there are always gaps between theories and practices in
sports. Athletes/coaches and scientists have been trying to fill the gaps
by working together, but just conducting collaborative work would not
solve the issue because the goals of scientists and athletes/coaches are,
in many cases, quite different. For scientists, the goal is to answer ‘why’
good athletes can achieve great performances with objective evidence.
On the other hand, athletes and coaches want to know ‘how’ they can
enhance the performance and, ultimately, objectiveness and evidence
are not essential for them.

Then, how can we fill the gaps? One possible answer is to have ‘inter-
preters’ or ‘translators’ between scientists and athletes/coaches, such as
performance analysts or scientists who also have long coaching experi-
ence. It is also important for scientists not to only focus on studies with
statistical data, but also to publish many scientific case reports for read-
ers to obtain ideas they can directly apply to the training.

This book is full of exciting research topics provided by leading ‘in-

terpreters’ in sports and we hope it will be a bridge between theories
and practices. It includes studies on individual, closed and cyclic sports,
but also on team sports and individual sports highly dependent on the
opponent behaviour (as tennis and combat sports). This is very relevant

7
 sp or t science

as, nevertheless (e.g.) running and cycling have a relevant and pioneer
historical background in research, other sports are also growing in quan-
tity and quality in terms of studies. We also aimed to achieve a balance
between hard and soft sciences approach in this book, providing both
experimental and conceptual studies.

We want to acknowledge the authors that made contribute with their

intellectual property to this book and the reviewers for giving their ex-
pert comments. We hope that it be useful for the Sports Science related
community, helping consolidating existent knowledge but also encour-
aging innovation and creativity, pleasing the readers with its contents.

Pedro Morouço, Hideki Takagi, Ricardo J. Fernandes

8
 sp or t science

Lost in translation – getting your sport science

research message across
David B. Pyne1,2 and Naroa Etxebarria1

Introduction
The sport science research process
A common view in sport science is that the research process reaches
its end at acceptance of an article for publication. In parallel with this
view, authors sometimes overlook the notion that the most important
outcomes are dissemination and implementation of research results.
The priority should be answering relevant questions that affect real
people in a given clinical, sporting or community context [1]. Therefore,
careful crafting of research questions and scientific rigour to conduct
projects are merely a means of obtaining trustworthy and evidence-
based outcomes to advance sports performance. However, the concep-
tually simple translation between quality research and the subsequent
implementing of the practical applications is often disrupted along the
way. This breakdown in the process is often a consequence of ineffective
communication between sport scientists and the end beneficiaries of
the research - the coaches and athletes. Assuming sport scientists con-
duct research that addresses relevant questions, maximising the gains
of new knowledge for coaches and athletes centres on effective commu-
nication and a translation strategy. The challenge for sport scientists is
to translate their research outcomes into practical strategies that benefit
sport performance or exercise outcomes.

Sport science is the discipline that identifies, validates, develops and

refines various protocols and strategies to improve sports performance.
Contemporary sport science includes various disciplines such as sports
nutrition, physiology, strength and conditioning, biomechanics, sports
medicine, performance analysis and skill acquisition. Specialists focus-
ing on each of these disciplines work with individual athletes and their
coach, in a team sport setting, or in a general coordinator role for agen-
cies and sporting organisations. With the involvement of so many differ-
ent third parties, sport science can turn into a complex system of mul-

1
Research Institute for Sport and Exercise, University of Canberra, Australia.
2
Australian Institute of Sport, Canberra, Australia.

10
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

tiple simultaneous interventions that might interfere with one another

rather than work synergistically towards the same goal [2]. Therefore,
professionals usually work together to maximise training adaptations
and help minimise the negative effects of training, such as an increased
risk of fatigue, illness and/or injury. This team effort does not come with-
out challenges, but integration of all these different disciplines is needed
to enhance an athlete’s or team’s performance in minor or major com-
petitions.

Sport science continues to evolve from evidence-based scientific re-

search in related areas. It is arguable that science is generating knowl-
edge at a faster pace than the coaching and sport science community
can implement and benefit from. This is particularly true in the current
digital age where gathering, creating and accessing vast information is
an everyday practice, but best use of all the information challenges most
individuals and organisations. As science and technology advance at a
rapid pace, modern sport science is about translating knowledge from
fields that might seem foreign to sport, but could be as useful as creating
new knowledge. Contemporary research outcomes that might aid sports
performance could derive from non-intuitive research related to new
light materials used in racing cars (transferred to cycling technology),
the military/NASA (for example, GPS tracking technology), and keeping
athletes healthy (for example, haematological, microbiological and im-
munological monitoring). Wearable technology is among the latest ad-
vances in training monitoring, however, they also pose the challenge of
adequately interpreting the data recorded as well as ensuring validity
and reliability [3].

The benefit of translating this knowledge and research to sport could

provide a competitive advantage if implemented correctly. Finding
more effective ways of translating both sport and non-sport research
outcomes is worth exploring for the sport scientist [4]. This process can
be aided by coaching clinics and association/club meetings where more
experienced coaches can share their experiences in implementing new
ways and ideas derived from research. New coaches are more likely to
take notice of what their mentor coaches advise as they have more affin-
ity with them than sport scientists, especially if the new coach has had
limited exposure to research. This approach would work well with expe-
rienced coaches who have the skill to interact with other professionals
for their own development and his or her athletes. Clubs/national or-

11
 sp or t science

ganisations can benefit from inviting external coaches as a professional

development exercise for less experienced coaching staff.

The literature on best practice for how to apply sport science research
is scarce. Aside a few examples looking into coach perceptions on the
importance and the application of sports science [5] and how to best
apply sport science to triathlon [6], the research body on the sports sci-
entist’s role is slim. A more authoritative and comprehensive overview
of strategies for translating research in sport is needed. Other research
areas, especially within the clinical and medical fields, have developed
strategies to implement research outcomes for the benefit of patients
[1, 7]. Some journals now feature a section where scientists summarise
the practical applications of research manuscripts submitted for publi-
cation. This approach should encourage a growing body of easily appli-
cable research in sport.

The modern trend of increasing numbers of support staff in sport has

added a degree of complexity to the coach - athlete interaction. Coaches
in contemporary sport often have to manage a large group of profes-
sionals all advising on different areas and sometimes with competing
interests. As these support staff help maximise recovery as well as adap-
tations from training, while avoiding injury/illness, prioritising each area
at any given time becomes crucial, hence communication and planning
by the group is essential to realise the benefits of research. Sports have
obtained a significant benefit from applying many of the latest evidence-
based outcomes that sport science and related disciplines offer. Howev-
er, unless there is harmonious interaction between these disciplines and
research outcomes, this ‘support’ can prove distracting to the athlete’s
performance. Implementing research outcomes in a simple, carefully
planned and effective manner is a challenge but also an opportunity.

Research outcomes worth implementing by definition are based on a

very high likelihood of being clinically or practically significant. This gen-
eralisation somewhat contradicts the reality of a coach and an athlete,
whose careers depend on the quality of performance on the day of a spe-
cific sporting event. Researchers need to generalise the observations ob-
tained from a sample of the population (study participants) to the rest of
the population of athletes with similar characteristics. For example, out-
comes from a sample cohort that are male, over 18 years old and moder-
ately active and healthy will be most relevant to the population of a simi-

12
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

lar description. Given that coaches and athletes are judged constantly on
their performance, and their funding and scholarships schemes demand
consistency within and between seasons, the sport scientist can strug-
gle to implement a new idea that athletes and coaches are unsure how
it might affect them directly. Of course, unless an athlete and/or coach
tries something new, they will not know how they might benefit, despite
an informed opinion based on the evidence from rigorous research. It is
important to keep in mind that high performing athletes are a rarity in
the general population given their genetic talent, extensive training and
expert coaching. Most research is based on highly trained athletes, but
few sports scientists manage to persuade elite athletes to participate in
research. Consequently, the application and success of a new research
idea needs to be ‘tested’ in training or minor competition by athletes to
see its likely effect on major competition.

Sports medicine includes physicians and physiotherapists among

other practitioners who have the responsibility of looking after the clini-
cal side of athlete well-being. Although the majority of sports medicine
research is based on improving and optimising clinical treatment of
illness and/or injury, a significant effort is made in preventative strate-
gies and for the athletes’ health and safety [8]. Similar to sport science,
sports medicine seeks to apply the latest medical research knowledge
to a sporting context, after understanding the reality of sport and the
intricacies of physiological adaptations in well-trained individuals. So
many of the issues of translating scientific research also apply to sports
medicine research.

Facilitating the transfer of knowledge from scientist to coach/athlete

is the responsibility of researchers and sport science practitioners who
have the capacity to understand, conduct and critically analyse rigorous
research data and apply it to a sporting context. However, both sport sci-
ence and sports medicine face similar challenges in first identifying the
underlying needs, integrating the practical applications of the latest re-
search into the busy training and competition schedules of athletes, and
disseminating research outputs to the wider sporting community. More-
over, the sport scientist has to develop the skill set to interact with peers
from his or her own and other areas of research to continue their growth
and development as a scientist. A growing opportunity to achieve this
latter professional development is in academia-industry collaborations,
where collaborative funding and expertise can benefit sport [9].

13
 sp or t science

Adding to the already long list of skills required for the sport scientist
is the obtaining of funds to develop research. Often sports scientists ful-
fil a role of a translator between researchers and coaches, athletes and
sporting organisation, however, sport scientists can also develop new
research of their own add high value to the team. Instead of having to
revert to a third party, sport scientist and coach/athlete teams can iden-
tify and develop their own customised research projects that best suit
their needs at any given time. Gaining funding grants to realise such re-
search projects often falls on the scientist as sports run on an already
tight budget. Academic allies that are eager to work with industry often
seek matched funding opportunities where universities and sports con-
tribute to the cost of running the research. Seeking collaborations with
universities, government and private entities is another opportunity the
sport scientist should develop. With limited funding opportunities di-
rected to sport-related research, collaborations seem a smart approach,
but researchers should disclose any conflict of interest when obtaining
funding from commercial companies.

Application of research outcomes

As the demands of modern sport continue to evolve, scientists need
to be adaptive and intuitive. The priority for the sport scientist should be
answering relevant questions that affect real people in a given sporting,
clinical or community context. Research outcomes from sport science
studies are crucial for developing guidelines for high performance sport,
exercise and physical activity. However, these applications can only be
realised if the research process is shared, and results contextualised and
translated for specific cohorts and other populations. For example, re-
covery strategies might differ between a contact sport and non-contact
sports as would the nutritional interventions between a female gymnast
and a male rower. Chronological and biological age of the athlete, train-
ing regime, competition demands, environmental challenges to over-
come, and individuality among other factors will influence the acute re-
sponses and long-term adaptations to training. It is, therefore, the sport
scientist that will fine tune concepts and research theory to find the best
fit between evidence-based concepts and the individual athlete(s).

The main beneficiaries of sport science research are the coaches and
athletes whom benefit from acquiring the evidence-based knowledge
that has been translated and customised to their needs. Athletes and
coaches do not have time, or often the specific understanding in complex

14
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

areas or disciplines, to understand complex or highly technical research

outcomes or their importance. In these situations, athletes and coaches
will depend on a ‘translator’ to do so. Furthermore, implementation of
strategies has to be practical, cost effective and time efficient, or other-
wise the knowledge remains a theory and benefits go unrealised. To in-
crease the ‘utility’ of research outcomes, dissemination, translation and
practical applications that are viable need to be planned, developed,
and evaluated. In the absence of any clear candidate who can take on
this role in many cases, it is typically the responsibility of the researcher
to manage this process in partnership with the other stakeholders: the
sporting organisation, coach, athlete, support staff, related disciplines,
and other members of the research team.

Development
Dissemination of research outcomes
Timely discussion and dissemination of research outcomes require
effective strategies and appropriate technical and meaningful language
for different audiences via traditional scientific publication, conference
proceedings and social media strategies. Disseminating and communi-
cating research knowledge should span scientific, medical and allied
health disciplines, local, national and international contexts, and trans-
lation into non-scientific communities. However, the application of this
knowledge has been variable and often inappropriately applied world-
wide. The transferability of results obtained from highly trained athletes
might not apply directly to the wider population, and vice versa. There-
fore, researchers need to identify, develop and implement new ways of
putting highly scientific information and outcomes into context using
real world narratives, examples and case study approaches. This com-
munication could be maximised by the use of online resources (social
media, blogs/pages, emails and visual material such as posters and in-
fographics) and jargon-free presentations or workshops to key industry
and stakeholders. Of course, the language used in any of these formats
should be appropriate to the type of communication used, and always
in plain language. The research outcomes should reach the cohorts or
population that would most benefit from it, and this responsibility lies
partly with the scientist.

15
 sp or t science

Case study – Translation of warm-up research in high performance

swimming
It is generally acknowledged in the swimming community that warm-
ing up prior to competition is a worthwhile practice. However, warm ups
have been poorly described in the scientific and coaching literature, and
the extent to which different elements and formats are effective in en-
hancing performance is unclear. Furthermore, whether coaches actually
implement evidence-based recommendations emanating from relevant
research studies is unknown. We conducted a series of projects with the
aim of enhancing warm ups and ultimately competitive performance in
high level swimmers (see Figure 1). The challenge now is to how to trans-
late each of the separate research study outcomes into a coherent mes-
sage for the swimming community. Here we present a case study illus-
trating the challenges and opportunities for translation of sport science
research using a research program examining the effects of warm-up to
enhance swimming performance.

After identifying that maximising the warm-up for swimmers is vital,

the first step in the research program was to undertake a comprehen-
sive review of the physiology and practice of warm-up [10]. Although the
requirement of detailed knowledge of the underlying literature is not
paramount for the actual swimmer or sometimes even the coach, it is
fundamental for the scientist. This review process was particularly im-
portant given the inconclusive nature of warm up (and swimming-spe-
cific warm-up studies) in the sport science literature and the timespan
from previous reviews on the subject. The combination of a physiologi-
cal (for scientists) and performance (for coaches) focus in the literature
was deemed important during this review. While both swimmers and
coaches are focussed on performance outcomes, sport scientists should
understand the mechanisms behind the various strategies, practices and
approaches (such as the pre-event warm-up) they promote.

The second step of the program involved a survey of coaches (both

domestic and international) to characterise contemporary approaches
and attitudes to warm-up practices in high performance swimming [11].
Engagement of end-users of the research (in this case the coaches) at
the start of the program was deemed an important element in trans-
lation and implementation of the outcomes. Coaches identified four
key objectives of the pre-competition warm-up: physiological (elevate
body temperature and increase muscle activation), kinaesthetic (tactile

16
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

preparation, increase “feel” of the water), tactical (race-pace rehearsal),

and mental (improve focus, reduce anxiety). The pool warm up volume
ranged from ~1300 to 2100 m, combining low-intensity swimming, drill
work, with 3-4 race or near race-pace efforts (25-100 m; ~90-100% ef-
fort) finishing with 100-400 m easy swimming. Dryland-based warm-up
exercises, involving stretch cords and skipping, were also commonly
prescribed to prepare athletes for events between 50-200 m. Coaches
preferred swimmers complete their warm-up 20-30 minutes before
race start. Lengthy marshalling periods (15-20+ minutes) and the time
required to don racing suits (>10 minutes) were identified as compli-
cating issues. The combination of dryland-based activation exercises
followed by pool-based warm-up routines seems to be the preferred
approach taken by elite swimming coaches preparing their athletes for
competition. While this anecdotal information is interesting in a descrip-
tive sense, the primary outcome of this survey was to inform the design
and methodology of subsequent experimental work, and importantly,
the content and language of coaching resource material underpinning
translation of the experimental outcomes.

Development of the experimental studies centred on a combination

of pool-based and dry-land activities [12]. Initially, sixteen junior com-
petitive swimmers completed a standardised pool warm-up followed by
a 30 min transition and 100 m freestyle time-trial. Swimmers completed
four different warm up interventions during transition: remaining seated
wearing a conventional tracksuit top and pants, wearing an insulated
top with integrated heating elements, performing a 5 min dryland-based
exercise circuit, or a combination of passive and dryland. Swimming
time-trial performance, core and skin temperature and perceptual vari-
ables were monitored. A dryland-based exercise circuit completed alone
and in combination with a heated tracksuit jacket during transition can
substantially improve sprint swimming performance. Attenuation in the
decline of core temperature between the warm up and competition, and
a reduction in start time, appear as likely mechanisms for this outcome.
While few swimmers will have access to a heating jacket during transi-
tion, they can all perform simple exercises and wear a conventional
tracksuit during the transition phase.

The initial development of the warm-up model was, as is often the

case in swimming and many other studies, with junior or age-group
swimmers/athletes. While benefits of the pool-based and dry-land ex-

17
 sp or t science

ercises were established with the juniors, the critical question for the
broader swimmer community was how these effects translate to senior
swimmers [13]. Twenty-five senior swimmers completed a standardised
pool warm-up followed by a 30-min transition phase and a 100 m free-
style time trial. During the transition phase, swimmers wore a tracksuit
jacket with integrated heating elements and performed the dry land-
based exercise routine (Combo), or a conventional tracksuit and re-
mained seated (Control). A traditional pool warm- up coupled with pas-
sive heating via heated jackets, and dry-land exercises in the transition
phase improved elite sprint swimming performance by ~0.8%. This mag-
nitude of improvement is worthwhile and formed the basis of evidence-
based recommendations to coaches and swimmers.

Transferability of effects is important in swimming, where there are

several different event distances and strokes (freestyle, butterfly, back-
stroke, breaststroke, medley) [14]. Performance in the 100 m time trial
was not substantially different in 10 breaststroke swimmers between
the different forms of warm up. It appears that targeted passive heating,
and completion of dryland-based activation exercises in the transition
phase, does not enhance sprint breaststroke performance despite elicit-
ing elevated skin temperature immediately before time trial commence-
ment. So the translation message was that ‘one size does not necessarily
fit all’ – other strategies are needed to enhance breaststroke swimming
performance. We also encouraged coaches and athletes to become fa-
miliar with the heated jackets and activation exercises in training, or a
minor competition, before using them for the first time at a major na-
tional or international competition.

A final research study investigated the timing of the warm up and

magnitude of the benefits of combined warm up activities in the con-
text of training and competition schedules [15]. There are thousands
of sport science publications examining the effects of various exercise
interventions and strategies, although not all are contextualised to the
high level sporting environment. A substantial proportion of elite ath-
letes train (and often compete) twice or three times per day prompting
questions on the timing and likelihood of carry-over effects from one
session to the next. The effect of a morning swim or priming session on
subsequent swim performance has been a perennial question in the high
performance swimming community. Thirteen competitive swimmers
completed: a) a morning session of 1200 m of variable intensity swim-

18
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

ming, b) a combination of varied intensity swimming with a resistance

exercise routine or c) no morning exercise. Following a six hour break,
swimmers completed a 100 m time-trial. We concluded that completion
of a morning swimming session alone, or together with resistance exer-
cise, can substantially enhance sprint swimming performance complet-
ed later the same day. Coaches need to consider how a morning priming
or potentiation swim (competition or training session) might enhance
subsequent performance later in the day or evening.

Transfer of research outcomes for the broader high performance

swimming community involved briefing sessions for swimming scien-
tists, coaches and athletes; distribution of scientific manuscripts to sport
scientists and plain language summary sheets as well as a one page info-
graphic summary for coaches [16]. Infographics involving visual display
of data and content are proving to be a popular form of presenting re-
search outcomes for athletes, coaches, and the general community.

Figure 1. A seven step model for the development, transfer and implementation
of research examining the use of warm-ups in high performance swimming. Step
1 Identifying key questions. Steps 2 and 3 Comprehensive review of both relevant
scientific studies and contemporary practices and approaches of leading swimming
coaches. Step 4 A series of pool-based experimental studies using both junior and
senior swimmers to develop an effective warm-up combining both pool swimming,
dry-land exercises and wearing of a battery-powered heated jacket. Step 5 Extension of
the warm-up protocol to senior swimmers across different events. Step 6 Preparation
of easy-to-read plain language education materials and briefing sessions for swimmers
and coaches, and provision of heated jackets and supervised dryland warm-up
exercises on the pool deck prior to competition. Step 7 Implementation of research
outcomes and subsequent evaluation to inform new or recurring key questions.

19
 sp or t science

Case Study - Lessons learned

Although the program of swimming warm-up studies yielded several
research publications and generated interest in the swimming com-
munity, some shortcomings were noted. A challenging issue facing re-
searchers is the time needed to complete the research and report back
to coaches, subjects and relevant stakeholders such as the national fed-
eration. Athletes and coaches understandably have a short-term focus,
and seek rapid feedback and expect worthwhile recommendations be
implemented immediately. In contrast, research studies typically take
months to years from the time of conception, through the formal experi-
mentation and data collection, to report preparation and final dissemi-
nation of outcomes. Distribution of preliminary outcomes, at least to the
stakeholders involved (privately), can complement the lengthy process
of research. There are now many ways to disseminate data online in a
more time-effective manner, however, this might compromise the ability
to publish in a journal afterwards, so care is needed.

The steps in the information or outcome dissemination process

should be built into the study planning and design, and revisited regu-
larly to ensure these milestones are achieved on time. Individual mem-
bers of the research team could be assigned this task but as per industry
standards, all researchers should approve and endorse the content of
scientific publications, conference abstracts and digital media output.

Automated processing of reports will aid the preparation and distri-

bution of individualised reports. These reports can be presented in two
versions. The first version immediately after the data collection could
feature raw data, descriptive data, previously established reference
ranges, and figures and tables. This material can be prepared before the
final data analysis and interpretation is completed. A second distribu-
tion to subjects with the final study outcomes would come at the end of
the research project. This material would have been through the data
analysis and interpretation process, but more importantly, with clearly
identified conclusions, take home messages and practical applications.
This information would need to be in plain language, or an infographic,
for non-specialist readers such as coaches, athletes and officials. An ex-
ecutive summary should be provided as the first page of a fully detailed
report, and prominent use of dot or bullet points to summarise the main
outcomes would be welcomed by all.

20
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Another identified issue in the research process was shortcomings as-

sociated with separate distribution of results to athletes, coaches and
scientists. Experience shows that the better option is for sharing and in-
teractive feedback involving the coach, athlete and scientist in a three-
way discussion. Having all parties hear the same information at the
same time is preferred. This approach avoids the problem of the sport
scientist delivering different information advice to the swimmer that
might run counter to the coach’s requirements, position or opinion. The
ideal situation is where all invested parties sit together at the same time
to review the athlete’s individual results, the group results/outcomes,
and their likely significance. At this point corrective actions can be im-
plemented to ensure the athlete continues to improve.

The life of the sport scientist in the 21st century is not an easy one
[17]. The elite sports community can be difficult for sport scientists and
their research outcomes to be heard above the media throng, social me-
dia and a competitive high profile environment. Nonetheless, the sport
scientist can take confidence that high quality evidence-based research
outcomes with practical application will be heard, provided the right
message is presented at the right time in a format suitable for the target
audience.

Recommendations
Here is a brief list to promote the translation and increase the impact
of sport science research:

ғ
Seek evidence-based knowledge that comes from sound and rigorous

research studies; although application and translation to sports is key,
robust research is a critical factor in developing effective coaching pro-
grams and training interventions.
ғ
Where available, make use of communication, media and public rela-
tions experts in your organisation, or collaborating organisations, who
have expertise to assist the translation and dissemination of research
outcomes.
ғ
Using posters in targeted areas to let people know about research activi-
ties and outcomes – instead of using them just to recruit participants.
ғ
Think laterally about who the research beneficiaries, collaborators and
other stakeholders might be. Start with a list of everyone who might be
impacted by the research outcomes (e.g. elite athletes, recreational ath-
letes, sports clubs, sporting organisations, etc.).

21
 sp or t science

ғ
Find professionals and expertise outside the primary scientific disci-

pline that might contribute to the research from a different perspective
and add value or valuable insights.
ғ
Plan for internal as well as external workshop/presentations and invite
the community to engage with the research team.
ғ
Use social media to disseminate research updates and outcomes by
using colourful charts of the most interesting findings to obtain the at-
tention of the audience that might not have access or tendency to read
detailed journal articles or attend conferences.
ғ
Prepare an ‘infographics’ slide for your research. Infographics are an ef-
fective way to simplify and disseminate your research.

Conclusions
The communication skills of the modern sport scientist now expand
beyond their specific field of expertise. The scientist needs to develop
the skills, expertise and experience for effective translation of research
outcomes. The final research outcomes need to be presented in plain
language for non-expert users of scientific and/or technical information.
Collaborations, nationally and internationally, matching research
funds, and applied research outcomes are fast becoming priorities for
many successful sport science projects. Funding agencies now expect
real world applications and a substantial practical return on the invest-
ment by stakeholders, participants and researchers.
Although researchers are mostly judged on publications and research
income, research teams should commit to the social, community, and
ethical responsibilities in their chosen field of study, and the impact of
their research in guiding clinical and sport training practices.

References
1. Assael, L.A., Translating science: from idea, to research, to clinical practice.
J Oral Maxillofac Surg, 2005. 63(6): p. 729-31.
2. Balague, N., et al., Sport science integration: An evolutionary synthesis. Eur
J Sport Sci, 2017. 17(1): p. 51-62.
3. Cardinale, M. and M.C. Varley, Wearable Training-Monitoring Technology:
Applications, Challenges, and Opportunities. Int J Sports Physiol Perform,
2017. 12(Suppl 2): p. S255-S262.
4. Eisenmann, J., Translational Gap between Laboratory and Playing Field:
New Era to Solve Old Problems in Sports Science. Translational Journal of
the American College of Sports Medicine, 2017. 2(8): p. 37-43.
5. Martindale, R. and C. Nash, Sport science relevance and application: per-
ceptions of UK coaches. J Sports Sci, 2013. 31(8): p. 807-19.
6. Millet, G.P., D.J. Bentley, and V.E. Vleck, The relationships between science

22
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

and sport: application in triathlon. Int J Sports Physiol Perform, 2007. 2(3):
p. 315-22.
7. Scott, I., The evolving science of translating research evidence into clinical
practice. Evid Based Med, 2007. 12(1): p. 4-7.
8. Pyne, D.B., et al., Probiotics supplementation for athletes - clinical and
physiological effects. Eur J Sport Sci, 2015. 15(1): p. 63-72.
9. Asadullah, K., et al., Industry-academia collaborations for biomarkers. Nat
Rev Drug Discov, 2015. 14(12): p. 805-6.
10. McGowan, C.J., et al., Warm-Up Strategies for Sport and Exercise: Mecha-
nisms and Applications. Sports Med, 2015. 45(11): p. 1523-1546.
11. McGowan, C.J., et al., Current Warm-Up Practices and Contemporary Is-
sues Faced by Elite Swimming Coaches. J Strength Cond Res, 2016. 30(12):
p. 3471-3480.
12. McGowan, C.J., et al., Heated jackets and dryland-based activation exer-
cises used as additional warm-ups during transition enhance sprint swim-
ming performance. J Sci Med Sport, 2016. 19(4): p. 354-358.
13. McGowan, C.J., et al., Elite sprint swimming performance is enhanced by
completion of additional warm-up activities. J Sports Sci, 2016.
14. McGowan, C.J., et al., Evaluating Warm-Up Strategies for Elite Sprint
Breaststroke Swimming Performance. Int J Sports Physiol Perform, 2016.
11: p. 975-978.
15. McGowan, C.J., et al., Morning Exercise Enhances Afternoon Sprint Swim-
ming Performance. Int J Sports Physiol Perform, 2016. 6: p. 1-23.
16. Le Meur, Y., Infographics, in https://ylmsportscience.com/2015/09/25/ba-
sics-warm-up-strategies-for-sport-performance-by-ylmsportscience/, Y.L.
Meur, Editor. 2015.
17. Burke, L.M., Communicating sports science in the age of the twittersphere.
Int J Sports Nutr Exer Metab, 2017. 27(1): p. 1-5.

23
 sp or t science

Periodization in sport training: traditional,

blocks and polarized
Jose Maria González Ravé1

Background
Variation in training is known as one of the key factors for achieving
success in the training schedule to optimise sports performance (Fleck,
1999). Training loads and their proper organisation constitute one of
the trainer’s most important tasks for helping athletes properly adapt
their bodies, thereby optimising their maximum sports performance.
Although an exhaustive analysis of the training loads has already been
conducted in Chapter 3, our proposed investigation emphasises the in-
terconnection and distribution of these loads. The distribution of loads
alludes to their dynamics and organisation in the various cycles of train-
ing and is the way the various loads are placed in a session, microcycle,
mesocycle or macrocycle (Verkhoshansky, 1990; Siff & Verkhoshansky,
2000; Navarro, 2000). The concept of load interconnection indicates the
relationship among the loads of different orientation when combined
over time (Siff & Verkhoshansky, 2000:430). These concepts represent an
interesting approach for research in our field, given that the transition of
one training load to another should represent the transition to perfor-
mance levels other than the current level. Research in this field therefore
represents progress in the control of these training directions.
The conceptual development of this chapter will highlight the pro-
cess of training and the variants in its periodization as a noteworthy ele-
ment within the distribution and interconnection of training loads and
will show the potential gaps in research that serve as a basis for present-
ing potential future research lines.
Training Periodization
Periodization is one of the most important concepts in training. The
term has its origin in the period, which is a portion or division of time
within which we find smaller and easier to manage divisions called train-
ing periods (Bompa, 1999).
The concept of periodization originated in Eastern Europe and was
adapted to the planning of modern training in the 1960s by sports sci-
entists such as Matveyev in the Soviet setting and Bompa in the Czecho-
slovakian setting (Norris & Smith, 2002; Wathen, 1994). Subsequently,
1
Faculty of Sport Sciences. University of Castilla la Mancha (Spain)

24
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

American scientists such as Stone and O’Briant modified Matveyev’s

study, adding phases to the preparatory period (Norris & Smith, 2002;
Wathen, 1994; Wathen & Role, 1994).
Periodization is the cyclic and gradual ordering of training exercis-
es, following principles of specificity, volume and intensity to achieve
high levels of sports performance in the most important competitions
(Wathen & Role, 1994; Fleck, 1999). Plisk and Stone (2003) defined pe-
riodization as a planned distribution or a variation in training methods
by means of time cycles. In this respect, it seems that periodization is a
fundamental concept for achieving sports performance. Hoffman (2002)
considered 2 types of periodization to achieve performance: linear mod-
els, considered as the classical form of periodization design, and non-
linear models (undulatory), in which the training volume and intensity
varies from session to session, as shown in Table 1.

Series Repetitions Rest between series

Day 1 3-4 8-10 RM 2 min
Day 2 4-5 3-5 RM 3-4 min
Day 3 3-4 12-15 RM 1 min

Table 1. Example of Nonlinear Period Training

The rational, deliberate and studied organisation of the training con-

tent is a key factor for good periodization. The distribution of training
content within the various training structures is a seldom studied aspect
of planning, which trainers conduct by recommendations or by practical
experience gained from day-to-day practice.
The objective in the periodized models is to attempt to maximise the
principle of overloading and ensure the correct relationship between
stimulation and recovery. The principle of overload in sports training is
the process by which the neuromuscular or cardiorespiratory system is
stimulated by larger loads, causing an adaptation. The periodized pro-
grams attempt to maximise this adaptation by changing the magnitude
of the loads and thereby increasing the neuromuscular and cardiores-
piratory system’s response capabilities (Howley & Franks, 1986; Stone,
O’Bryant & Garhammer, 1981).

Periodized versus Nonperiodized Models

Periodized strength training programs appear to produce greater
gains than nonperiodized programs of one or multiple series (Fleck,

25
 sp or t science

1999, Stone, O’Bryant & Garhammer, 1981, Rhea et al., 2003). The first
of the classical studies in this field dates from the start of the 1980s. The
study compared a periodized programme with other nonperiodized pro-
grammes for 6 weeks, performing 3 weekly sessions, improving 1 RM in
squats and power in the vertical leap (Stone, O’Bryant & Garhammer,
1981). Other studies have compared undulatory and linear periodized
programs and have showed that the inverse linear programmes in which
the volume is increased, and the intensity is reduced are more effec-
tive than the undulatory or linear incremental programmes (Rhea et al.,
2003). In the study by Willoughby (1993), there were no significant differ-
ences during the first 8 weeks in the total volume of work between the
periodized model and the 2 training programmes of multiple series. After
8 weeks, the volume of training was decreased compared with the other
programmes. From there to week 16, there were significant differences
in 1RM in bench press and squats between the periodized model and
the other models, even though performance was also improved in these
models. Baker et al. (1994) found no significant differences in maximum
strength in squats, bench press or vertical leap between the undulatory
and linear periodized models, during 12 weeks in which the total volume
and intensity (%) were similar in both programs, which allowed for the
variations between one and the other to be attributed to the difference
between the programmes. Despite finding no significant differences in
maximum strength between the periodized and nonperiodized models,
other studies found improvements exclusively in the group that exercis-
es according to the periodized model (Plisk & Stone, 2003).
González-Ravé et al. (2007) compared the efficacy of periodized pro-
grams against that of nonperiodized programs during 5 weeks of train-
ing in physically active participants (Sports Science students) with the
intention of improving maximum power in performing the bench press
exercise. Twenty-two individuals participated in the research study and
were divided into 2 groups: a nonperiodized group (9 participants) and a
periodized group (13 participants). The nonperiodized group performed
5 series of 6 repetitions 2 times a week for 5 weeks, reassessing maxi-
mum power at 2.5 weeks. The periodized group performed 5 series of
6 repetitions the first week, 4 series of 5 the second week, 5 series of 7
the third week, 5 series of 8 the fourth week and 4 series of 6 the fifth
week, reassessing the power halfway through the study. The total vol-
ume and intensity were evenly matched. The results of the study showed
that the power increased significantly in the intermediate measurement
(13.26%, p<0.05) compared with the pretest measurement in the perio-

26
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

dized model, while in the nonperiodized model, the significant percent-

age increase was 19.83%. However, in the final measurement compared
with the intermediate measurement, the power in the periodized model
only increased 2.86% (although not significantly), while in the nonpe-
riodized model, the power decreased significantly (10.21%, p<0.05)
compared with the intermediate measurement. We therefore conclude
that there are no significant differences in the gains in maximum pow-
er between the 2 groups. The results in the untrained participants also
showed limitations in the applicability of these models to highly trained
athletes.
A more recent study by Kell (2011) determined the influence of 12
weeks of periodized training on changes in strength in young male
and female sports enthusiasts with prior experience in nonperiodized
strength training. The results showed that 12 weeks of strength training
following traditional periodization induced significant strength gains in
women (>30%) and men (>25%) with approximately 11 months of prior
experience in nonperiodized strength training.
One of the most important contributions in the research on the effec-
tiveness of training periodization was by Rhea and Alderman (2004), who
performed a meta-analysis of the results of periodized versus nonperio-
dized programmes based on the publication of scientific studies from
1962 to 2000 for the selection of studies to be included in the statistical
analysis. The keywords for the search were periodization, strength train-
ing, anaerobic exercise, resistance training and weightlifting.
Rhea and Alderman (2004) concluded that periodized training is
more effective than nonperiodized training for men and women and for
individuals of varying levels of preparation and ages and that when one
considers the variables of volume intensity and frequency in the train-
ing program; this is manifested in adaptations of better physical perfor-
mance.

The Classical Theory of Periodization

The theory of periodization was originally defined by LP Matveyev
in 1965 as an update to the study published in 1962. The annual train-
ing cycle is conventionally divided into 3 main phases: the preparatory
phase, the competitive phase and the transition phase (Bompa, 1994).
Wathen (1994) indicated that each mesocycle has a different duration,
lasting from weeks to months, depending on the type of sport, the ath-
letes’ sports performance, competitions, intended objectives and time
between competitions. Each mesocycle contains several microcycles,

27
 sp or t science

which are generally 1-week periods, although they can be fewer train-
ing days. The microcycles are defined as a series of training sessions,
organised rationally in a short period of time (García, Navarro & Ruiz,
1996). They are not finished processes but rather represent organisa-
tional structures that will complete the processes with the mesocycles.
During the 1950s and with the success of the Soviet delegation in the
Helsinki Olympic Games, the scientist Lev Matveyev presented and pop-
ularised his general theory of sports training based on biological laws
and in particular the general syndrome of stress adaptation. The defini-
tion of sports fitness disseminated by Matveyev (1977) is defined as the
condition of optimal predisposition for achieving sports goals, which is
possible to achieve through a detailed organisation of the training pro-
gram in periodic cycles. Matveyev defined these cycles as the prepara-
tory period, the competitive period and transitory period (Fig. 1). He also
established that sports conditioning goes through 3 states: acquisition,
maintenance and temporary loss of this sports fitness (Matveyev, 1977;
Navarro, 2000).
The fundamental pathway for the acquisition and maintenance of
sports fitness is based on managing and modulating the volume and in-
tensity of the stimuli, tasks and exercises, which, along with the complex-
ity of the exercise, constitute the training load. The programmes include
modulation of the volume and intensity in an inversely proportional
manner. The preparatory period is started with increasingly greater vol-
umes and moderate intensity. Subsequently, during the maintenance
period, this situation is inverted to reduce the volume of training and
increase the intensity to prepare the body for competition. A supreme
state of performance is thereby achieved, which is manifested in the
previously mentioned sports fitness (Matveyev, 1977; Baker, et al. 1994;
Bompa, 1994, 2000)

Figure 1. Example of the Classical Model exposed by Matveyev.

28
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Foundations of periodization
The original basis of periodization was the general adaptation syn-
drome proposed by Hans Selye (1956), who described the body’s gener-
ic response to stress (Wathen, Baechle & Earle, 2000). According to this
model, the initial response phase to any stress is characterised by shock
or alarm (Selye, 1956; Wathen, Baechle & Earle, 2000). After this, there is
a supercompensation phase in which the body adapts to increase the
specific capabilities affected by the stress (Selye, 1956; Wathen, Baechle
& Earle, 2000).
The homeostasis process is caused by the training load the body is
subjected to, which produces a stimulation that leads to an imbalance
(or heterostasis) in the anabolic and catabolic processes, which the body
once again tries to rebalance. This righting is manifested in an increase
in anabolic processes to protect the structure from excessive exhaustion
(Fry et al., 1991). This means that the regenerative processes attempt
not only to recover the athlete’s initial level but also to exceed it. This
phenomenon is understood to be a protective mechanism to prevent a
future emptying of reserves in the event of greater repeated loads. This
phenomenon is known as supercompensation (Grosser, Brüggemann &
Zintl, 1989; Fry et al., 1991). In principle, supercompensation can only be
related experimentally to the processes of glycogen metabolism. Trans-
ferring the supercompensation model to other metabolism settings
does not appear possible (Martin Carl & Lehnertz, 2001). Figure 2 shows
a graphical representation of these processes. In the most current litera-
ture, the term supercompensation may be replaced by that of adaptive
reconstruction, given that adaptation to training loads is not produced
equally in all the body’s systems (Siff & Verkhoshansky, 2000).

29
 sp or t science

Figure 2. Explanatory diagram of the adaptation processes in the body.

(Yacovlev, 1977, cited by Grosser, Brüggemann and Zintl, 1989).

This process of adaptation subsequently defines a theoretical model

of sports training known as the Unifactorial Training Model (Siff & Verk-
hoshansky, 2000). Based on the principle of supercompensation, this
model establishes how the body hypothetically adapts to a higher per-
formance level in response to a repeated training load. Matveyev (1980:
48) presents this process as the overload and rest.

The systems theory has had considerable influence on sports train-

ing. A system is an entity characterised by at least one input and one
output related to the input through a mathematical function (Busso &
Thomas, 2006). The output corresponds to the system’s response to a
stimulus represented by the input. The function shows the system’s be-
haviour using the determinants of sports performance in a specific sport.
Figure 3A reflects the schematic representation of the system in a general
manner, according to the characteristics established by Busso & Thomas
(2006). Figure 3B illustrates the systemic application to sports training
in such a manner that the input is represented by the dose of training
performed, the system is represented by the athletes and the output is
reflected by the subject’s performance. This model is highly simplistic in
its formulation because the accumulation of training loads produces a

30
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

variation in the type of fatigue it causes and, in the body’s, subsequent

adaptation to a performance level that is expected to be higher. Moreo-
ver, the model omits other factors outside the training itself that affect
performance such as psychological, biological and tactical preparation.

Figure 3. Schematic representation of the systemic model applied to sports

training.

It has been assumed that the preparation of athletes is not an ana-

lytical and unilateral process but rather a complex system, composed of
several subsystems related to the athlete’s various spheres involved in
improving their sports achievements. For Matveyev (2001), the prepara-
tion is determined by the training system, the competition system and
the use of “complementary” factors of various types, such as general life
regimens and specialised diets. Without the competition of these fac-
tors, the sports preparation system would be deficient.

Thus, the sports training system considers the individual as a whole,

given that the individual only performs their vital functions in this man-
ner, thereby opposing the reductionist concept of understanding the in-
dividual from just their biological side.

Relationship between Sports Fitness and Training Periods

Sports fitness is defined by Matveyev (1993) as the state of optimal
predisposition (the best) for achieving sports goals, which is acquired
by athletes due to the corresponding preparation in each new step of
the sports improvement. Sports fitness should cover all aspects in their
entirety, both physiological and psychological.

31
 sp or t science

Thus, the process of developing sports fitness comprises 3 phases

(Matveyev, 1983):
ғ
Stabilisation
ғ
Maintenance
ғ
Temporary loss of form

Figure 4 shows a representation of these phases and their characteristics.

Development phases Training Main objectives

of sports fitness periods
1 Acquisition Preparatory Formation of pre-requisites for sports
fitness.
Accumulation of motor and
multilateral coordination capabilities.
General motor development.

2 Stabilisation Competitive Gradual improvement of the state of

preparedness.

Achievement of stable preparation,

raising the results with a certain
degree of variation
3 Temporary loss Transition To interrupt the training with heavy
load
To facilitate active recovery
Renewing the athlete’s adaptation
reserves

Figure 4. Phases of the development process for sports fitness and their cor-
respondence

The distribution of training loads during the training process.

For Verkhoshansky (1988), there are 2 types of variants for generating
the adaptation in athletes. The first (A) consists of applying loads of no-
table volume, which cause considerable mobilisation of energy sources
and that produce prolonged and deep changes in homeostasis. This type
of training is temporarily organised into 3 to 4-week mesocycles, after
which a rehabilitation break of 7 to 10 days is needed. These mesocycles
are used for 18 to 22 weeks. This type of strategy is mainly used by high-
level athletes and for rapid strength. This variant is related to concentrat-
ed loads (Issurin & Shkijar, 2002; Issurin & Lustig, 2004) and began to gain
strength in the 1980s as a criticism and alternative review of the gradual
load model around Matveyev periodization (Issurin & Shkijar, 2002). The

32
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

concentrated loads models are closely related to the particularity of

their effects, known as residual training effects (Issurin & Lustig, 2004).

The second variant (B) consists of the gradual increase in functional

indices and is expressed in the case of a moderate and continuous vol-
ume of training loads. In this variant, periodic and short-term changes
are caused. Thus, the body’s homeostasis and energy reserves are com-
pensated during the training cycle. This type of training is temporarily
organised into longer cycles compared with the concentrated load sys-
tems (specifically 5 to 6 weeks), after which a rehabilitation break of the
same length as the first variant is needed. This form of adaptation cor-
responds to the traditional training periodization idea of Harre (1987),
Matveyev (1977) and Stone, O’Bryant & Garhammer (1981).

Each variant presents completely different adaptation rhythms in

terms of the elapsed time to reach the sports performance and are rep-
resented graphically in Figure 5.

Figure 5. General diagram for the various types of body adaptation reactions
to loads according to their distribution.

The Classical Periodization Models: General Preparation vs. Spe-

cial Preparation. Workload Dynamics.
The workload can be divided uniformly in the cycle (regular or distrib-
uted workloads), which produces a uniform distribution of the means

33
 sp or t science

during the annual training cycle, or concentrated in defined phases of

the annual cycle (concentrated loads). This distribution of regular loads
(seen in topic 2) are those that configure the periodization structure of
the classical model.
Matveyev (1980) based his periodization model on a general work
paradigm of high volume/low intensity during the first part of the devel-
opment of sports fitness, known as the preparatory period. The model
then proceeds to a specific work order of the specialty in a low-volume/
high-intensity regimen, which is described as the competitive period.
The main components to consider in any sports training program are
volume, intensity and frequency or number of training sessions, which
were analysed in topic 3. Most experts in training theory and method-
ology agree that these components determine the load magnitude and
therefore the response and adaptation to training (Matveyev, 1977;
Navarro, 1999, 2010). The training content corresponding to the basic,
specific and competitive levels of the sports discipline is conducted in
a distributed and regular manner over the course of the various periods
and phases of the cycle. Meanwhile, the total workload increases gradu-
ally over the course of the cycle, with a greater emphasis on volume dur-
ing the general phase of the preparatory period and on intensity during
the specific phase of the preparatory period and during the competitive
period.

The Structure of the Classical Matveyev Model

Matveyev (1977) structured his macrocycle into 3 long periods during
the season (Fig. 6):

1) Preparatory Period
2) Competitive Period
3) Transition Period

34
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Figure 6. Classical Matveyev model (1977). PPG: General; PPE: Specific

During the preparatory period, the premises and conditions neces-

sary for the acquisition of sports fitness are created.
The competitive period ensures the maintenance of sports fitness
and is applied to achieve sports success.
The transition period arises from the need to protect athletes from
potential injuries, providing active rest to protect from overtraining.
Each period’s duration depends on the sports specialty and on the
athletes’ condition. In principle, the preparatory period cannot be short-
er than required in the specific circumstances for acquiring sports fit-
ness. The competitive period cannot be longer than that permitted by
the possibilities of sports fitness maintenance without putting the sub-
sequent process in jeopardy. Typically, these processes are determined
by the standard year (Matveyev, 1993). The time recommendations pro-
vided by Matveyev (1993) are as follows:

ғ
Preparatory period: 3-(4/5) months.

ғ
Competitive period: 1-(2/5) months.

The preparatory period is subdivided into phases of general and spe-

35
 sp or t science

cial preparation. The main purpose of general preparation is to establish

the functional basis for training, thereby improving the physical condi-
tion (Navarro, 2000). For Matveyev (2001), the term general preparation
means something that cannot be reduced to sports specialisation, but
rather creates and optimises the context of the athlete’s progress in the
chosen sports modality by means of the contribution to the general in-
crease in the level of the athletes’ functional possibilities, their multifac-
eted development and their enrichment with the knowledge, skills and
habits, which in one way or another contribute to sports improvement.
When applied to sports with technical components, we should (in addi-
tion to raising the body’s functional capabilities) apply a range of motor
foundations and skills to the technique. In the strategy-tactics of team
sports, we need to support all the systems involved as well as the basic
theoretical aspects of the game (Hernandez Moreno, 1988).

Matveyev (2001) stated that special preparation starts with general

preparation, which progressively moves towards the objective of spe-
cialisation that, with growing depth, adapts the athlete to the chosen
competitive activity. In the special preparation, the objectives for physi-
cal preparation are directed towards the development of specific physi-
cal qualities. The bases for the motor capabilities and techniques need
to be created according to the specific requirements of each sport. In
terms of the technique and tactics in team sports, we start to introduce
these activities in their actual context. The theoretical aspects follow the
same direction (Hernandez Moreno, 1988).

The competitive period needs to be subdivided into 3 parts:

An early competition phase in which we seek to develop the specific
motor and technical capabilities, an increase in the athletes’ potential so
that they can take part in a series of competitions but without achieving
the maximum level in the competitions but rather as preparation for the
next period, which contain the main competitions.
The main competition phase is the most important part of the sea-
son. Training during this period is characterised by maximum, specific
intensity similar to that of the competition. The main part of the exer-
cises consists of modelled training. There may be competitions prior to
the important ones.
The final competition phase is not a macrocycle component but can
appear because the schedule of competitions can continue after the
main competition. This involves maintaining the form insofar as still

36
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

competing but considering fatigue in the main phase.

The transition period is a rest interval for establishing the means for
proper recovery, to start the next season.

Traditional periodization can be established based on 1, 2 or 3 peaks

of form (Fig. 7), considering particular competitive periods in which we
wish to establish appropriate tuning. However, the premises on which
this design are based allow for only one peak in form to be established
(Bompa, 1999; Bompa & Haff, 2007).

Figure 7. 1-peak, 2-peak or 3-peak Periodization.

However, the regular loads model still appears to be in use, as shown

by the studies of Gorostiaga et al. (2006) and Granados et al. (2008).
These scientific studies support the benefits of the classical periodiza-
tion model given that both handball (Gorostiaga, 2006, Granados, 2008)
and volleyball (González Ravé, Arija & Clemente, 2011) show significant
improvements in physical condition during a season after the use of this
organisation and the interconnection of training loads.

Criticism of the Classical Theory

In recent years, sports training has undergone several changes com-
pared with classical periodization, which used to be employed for sev-

37
 sp or t science

eral reasons. One of the most important of these reasons has been a
series of elements that were used to develop a theory of training by clas-
sical authors such as Matveyev and Zatziorski. According to Selujanov
(1999), these do not consider the studies by Vorobiev on weightlifting or
by Bondarchuck on launchings, which contain descriptions of training
structures based on theoretical principles aimed towards adaptation. In
Selujanov’s opinion (1999), this was the big mistake by Metveyev. More-
over, Matveyev training periodization has been criticised by authors of
other training methodologies from the Soviet school and Eastern coun-
tries, such as Verkhoshansky, who criticised the periodization of train-
ing. According to Verkhoshansky, the technique used to structure sports
training is very rudimentary, “… knowing the typical microcycles, which
are formed like children’s construction blocks, the following larger part
(mesocycles) that in turn configure the large microcycles…” (Platonov,
2000:88). However, the author later softens this criticism declaring that
the process, although seemly analytical and separated among the exist-
ing elements, and the final result tells us how much more significant are
the hidden connections of these elements that materialise in the con-
tinuity and the interrelationships of the effects obtained. According to
Platonov (2000), these criticisms by Verkhoshansky are not a response
to an updated reflection of the knowledge of current periodization, but
rather a criticism of the postulates performed in the 1950s and that have
been updated by Matveyev (2001).

In one of their more recent publications, Issurin (2008; 2010), men-

tioned the need to change the old theory of training and accept a new
implementation of the modern demands for competitive swimmers.
The same author highlighted 4 weak points in traditional periodiza-
tion:

ғ
Considerable reduction in training volumes.

ғ
The conflict resulting from the mixture of multiple skills trained simulta-
neously (mixed training programs).
ғ
Insufficient training stimulus to help highly qualified athletes to contin-
ue with the progress (results of mixed or concurrent training).
ғ
The inability to maintain frequent peaks of form during the season.

These same limitations of classical periodization had already been

identified by Verkhoshansky (1996; 1999). Continuing with the proposal
by Issurin of including of the blocks method (Block Periodization, also

38
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

known as ATR) in the preparation programs for elite swimmers, Issurin

(2010) established that this method presented 3 clear advantages com-
pared with the traditional periodization method:
ғ
Frequent performance peaks within the same competition year.
ғ
Deep and specialised focus on the effects of training in a relatively short
time;
ғ
The recognition that the swimmers have a continuous state of general
physical fitness in which a specialised block overlaps the effects of spe-
cialised training.

The New Periodization Models (Blocks or ATR Model)

The structure of concentrated loads is subsequent to that of regular
or distributed charges. This structure is in response to the new demands
of the current structures of high-level sports. We begin to question the
antiquated and immovable theories of training based on the presump-
tions of classical periodization of Matveyev (1965) and Harre (1957 and
1969) (Selujanov, 1999; Tschiene, 2001; Issurin & Shkijar, 2002). The con-
centration of loads allows us to use highly concentrated training instead
of a complex development of numerous capabilities. This training allows
for more selective, immediate and cumulative effects than through the
conventional model.

For Navarro (2000), the idea of the contemporary concept is based on

the following:
1) The concentration of training loads on specific capabilities or specific
training objectives (capabilities/objectives).
2) The consecutive development of certain capabilities/objectives in spe-
cialised training blocks or mesocycles.

Figures 8 and 9 show an example of periodization based on a struc-

ture training model called blocks. This model is based on the use of
concentrated training loads. The concentrated loads are based on the
principle of residual effect that the stimulus leaves on the athletes’ body.
The organisation of these loads seeks to delay those that have a greater
residual effect. The lower the residual effect these loads have, the closer
they are to the competition, as shown by the example of Issurin & Shkijar
(2002) in Figure 9. The concept of residual effect, although intuited in
some manner in the classical model, does not appear in sports training
literature as such until mentioned by Counsilman & Counsilman (1991).

39
 sp or t science

Figure 8. Residual effect of concentrated training loads following a blocks

model.

As can be observed in Figure 9, those capabilities that have a longer

lasting residual effect are placed further from the competition. Moreo-
ver, this organisation places generic basic capabilities such as aerobic
resistance and strength in the first phase. More specific capabilities such
as anaerobic resistance and strength resistance, which have a smaller
residual effect than those in the previous phase, are placed in the second
phase. Finally, those capabilities more closely related to the competition
situation, which have a smaller residual effect, are situated in the phase
closest to the competition. To promote the residual effect of the basic
and specific capabilities, memory miniblocks of 2-3 days are employed
to prevent a reduction in the previously trained capacity. Each selective
load has a specific temporary effect (residual) on the body. The charges
are heterogeneous depending on the type of metabolism requested. Is-
surin & Shkijar (2002) determined the duration of the residual effects of
load training in the following time references:

ғ
Alactic anaerobic capacity: 5 days.

ғ
Muscular endurance: 14 days
ғ
Anaerobic endurance: 18 days
ғ
Maximum strength: 30 days
ғ
Aerobic endurance: 30 days

40
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Figure 9. Prolonging the effects of training using training load miniblocks

(Issurin & Lustig, 2004).

The factors derived from the residual effect of concentrated loads

provide the following practical consequences on high-level sports
training (Issurin & Lustig, 2004):

ғ
The duration of the mesocycles varies from 2 to 6 weeks, although for

Verkhoshansky (1988) the ideal period would be 3 to 4 weeks. This vari-
ability among authors could be due to a lack of scientific foundation for
many of these approaches (Gorostiaga, 2000). For Issurin & Lustig (2004)
and Gorostiaga (2000), a 4-week mesocycle is more advantageous than a
longer cycle because the residual training effects are better taken advan-
tage of and the athletes’ biological response is more appropriate.
ғ
In the competitive period, the competition mesocycles are shorter than
in the preparatory period, because stress, the product of competition,
decreases the residual effect of training in successive mesocycles. Moreo-
ver, physiological and emotional stress increases the catabolic processes,
causing a reduction in muscle mass and reducing the residual effect of
training in the maximum and explosive strength. Just as a highly intense
activity with a marked anaerobic metabolism, the stress causes a loss in

41
 sp or t science

aerobic capacity accompanied by a reduction in the anaerobic threshold.

ғ
The use of 2 to 3-day memory miniblocks prevents a reduction in ini-
tially trained capacity and with it the short-term residual effect.
ғ
Athletes with more years of training can perform longer transition pe-
riods and lower training volumes (15-30% less) than younger athletes.
ғ
Thus, the training cycle of veteran athletes is shorter due to a longer
transition period. The longer duration of the residual effect of training
offers the possibility of performing a lower load volume.

Structure of the ATR (block) model

In the concentrated training model, the number of trainable capa-
bilities in a mesocycle should be decreased to 2 motor capabilities and
1 capability that is characteristic of the technique. The capabilities to be
worked on should be established so that a simultaneous task is set up
between each mesocycle, considering the effects produced by the task
through the interaction of loads (see the mesocycles section).
There are 3 mesocycles within the macrocycle:

ғ
The accumulation mesocycle attempts to raise the athlete’s technical

and physical potential.
ғ
The transformation mesocycle attempts to transform the potential of
the physical and technical capabilities by specific preparation.
ғ
The implementation mesocycle attempts to achieve maximum results
within the workout performed.

This ordering of mesocycles, and as a consequence of its microcycles,

is based on the cumulative residual effect produced by the earlier work-
out. In other words, the accumulation work should be maintained longer
than the transformation workout, which should be maintained longer
than the implementation workout in relation to its training content and
the duration of the effects it causes. Thus, each mesocycle is based on a
workout that develops the following capabilities based on the premise
by Issurin & Lustig (2004):

MESOCYCLE ACCUMULATION:
ғ
Aerobic training.
ғ
Maximum strength.
ғ
Greater residual effect.

42
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

MESOCYCLE TRANSFORMATION
ғ
Strength resistance.
ғ
Anaerobic glycolytic capacity.
ғ
Medium residual effect.

MESOCYCLE REALIZATION
ғ
Alactic anaerobic training.
ғ
Training in competitive conditions
ғ
Specific tactics.

Smaller residual effect.

This distribution can be observed in Figure 10.

Block Periodization
Navarro (2000)
M esocycles

A cumulation T r ansfor mation R ealization

Training contents

Basic Specific Competitive

Endurance Endurance endurance

Basic Specific Speed Capabilities

Strength Strength

Specific Competitive
Basic Skills Skills Skills

Thus, the mesocycles run during the season in several macrocycles,

which constitute the ATR model, as can be seen in Figure 11 below:

43
 sp or t science

Figure 11. Establishment of several macrocycles per season.

This figure shows how an annual cycle has several macrocycles that
run during the year. Moreover, we can obtain longer macrocycles by
consecutively placing 2 accumulation mesocycles and 2 transformation
macrocycles, achieving precise and longer training combinations.
In this case, we need to include a capacity in each mesocycle, thereby
ensuring a longer residual effect. The predominant training types, if we
consecutively employ several mesocycles of the same type, could be
those proposed by Navarro (2000:155) as established in Table 2:

Mesocycles Main type of training Additional objectives

Accumulation 1 Aerobic endurance Maximum strength
General conditioning
Accumulation 2 Maximum strength Aerobic endurance
General conditioning
Transformation 1 Muscular endurance Aerobic-anaerobic
endurance
Perfecting the skills
Transformation 2 Aerobic-anaerobic endurance Strength-resistance
Anaerobic endurance
Perfecting the skills
Realization Modelling the competition. Anaerobic endurance
Speed capabilities Competitive skills
Tactics

44
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Table 2. Training options with 2 mesocycles of accumulation and transformation.

Finally, the competitions in this model are preferentially located in

the mesocycle of implementation and always at the end, hence the situ-
ation of competitions affecting the planning design. However, for com-
petitions with less preparation, these can be distributed in the imple-
mentation or transformation mesocycles.
Of course, the number of competitions will depend on the type of
sport and its level of implementation and popularity and the athletes’
current level. Thus, elite athletes must attend more competitions and
should therefore have more microcycles.

The ATR model is not exempt from criticism, as stated by Kiely (2010);
however, the reply by Issurin (2010) to his critics merely fuels a most in-
teresting discussion on the scientific validity of the ATR model. There are
studies that supports its validity, although they are few. The study by
Breil et al. (2010) used an ATR model for 11 days (15 intensive interval
training sessions) with elite skiers. The skiers who used concentrated
loads improved their VO2max by 6.0% (p<0.01; men, 7.5%; women,
2.1%), relative peak power by 5.5% (p<0.01) and second ventilatory
threshold by 9.6% (p<0.01). The control group had no changes. Another
study that showed improvements in the study parameters was carried
out by Nuñez et al. (2008) on a sample of football players. The study em-
ployed training blocks with only resistance content or with only strength
contents. The authors concluded that aerobic resistance improved in the
initial phase of the season because of the training. To increase strength,
the number of training session of this type needs to be increased. It is
advisable to separate aerobic resistance and strength training to create
wider blocks during the last 2 macrocycles. García-Pallarés et al. (2010)
performed a comparative study with elite kayaking athletes between
an ATR model (13 weeks) and a classical model (23 weeks). The authors
concluded that performance was improved in the 2 models. However,
the blocks model was more efficient in parameters such as stroke speed
and maximum power manifested in the VT2.

Polarized training periodization

This periodization model is conceptualised as the integration and
distribution of various intensities of training focused on low and high
(thereby polarising the distributions), as established by Stoggl & Sper-
lich (2014). Based on the concept of distribution of training intensity,

45
 sp or t science

whose most representative studies are those by Seiler (2010) and Seiler
& Tonnessen (2009), which show how (in the analysis performed during
a season) most of the work volume is performed at low intensity (ap-
proximately 80%), while the rest is performed at high intensity. Based
on this procedure, this periodization model has been articulated around
resistance capacity. The review conducted by Stoggl & Sperlich (2015)
concluded that polarised periodization was an effective strategy for
some elite athletes during certain phases of the season. However, ex-
perimental studies that lasted 6 weeks to 5 months showed superior re-
sponses with polarised training, especially when comparing models that
emphasised the anaerobic threshold with contrasts of high volume-low
intensity. This combination can improve resistance performance, with
potentially less nerve and hormonal stress, and prevent the typical mo-
notony. Mujika et al. (1995) measured the distribution of training inten-
sity for international swimmers during a whole season based on 5 areas
of intensity, which in turn were based on blood lactate levels. Although
the swimmers specialised in the 100 m and 200-m events that required
approximately 60 s to 120 s, these swimmers swam 77% of the 1150
km completed during a season at an intensity below 2 mM of lactate.
The rest was performed at high intensity. The study by Yu et al. (2012)
showed positive results using this periodization model in professional
skaters. Muñoz et al. (2014) assessed the impact of a polar training distri-
bution compared with a low-intensity training program and a traditional
programme in recreational athletes. The study showed improvements
in all groups, although the polar training programme provided greater
improvements than the low-intensity training. Other studies, such as
those by Hydren & Cohen (2015) and Varela et al. (2016), have shown the
advantages of this training programme.
The season distribution, according to Stoggi and Sperlich (2014), is
performed as follows:
1) Preparation Period.
2) Precompetitive Period.
3) Competitive Period.

Each of these periods distributes the training intensity in very differ-

ent ways, as shown in Figure 12.

46
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Figure 12. Training intensity distribution in the various periods (Stoggi & Sper-
lich, 2014).

References
Baker, D. (1994) Periodization the effect on strength of manipulating volume
and intensity. J. Strength Cond. Res.; 8: 235-242.
Bompa, T (1999) Periodization. Theory and methodology of training. Human
kinetics. Champaign, IL.
Breil FA, Weber SN, Koller S, Hoppeler H, Vogt M.(2010) Block training periodiza-
tion in alpine skiing: effects of 11-day HIT on VO2max and performance. Eur
J Appl Physiol. 2010 Aug;109 (6):1077-86.
Busso, T.; Thomas, L. (2006) Using mathematical modelling in training plan-
ning. International J. of Sport Physiol and Performance. 1: 400-406.
Counsilman BE, Counsilman J. (1991) The residual effects of training. J Swim
Res 1991; 7: 5-12.
Fleck, S. J. (1999) Periodized strength training: A critical review. J. Strength
Cond. Res. 13(1):82–89.

47
 sp or t science

Fry, R.W., Morton, A.R.; Keast, D. (1991) Overtraining in athletes. Sports Medicine
12, 32-65.
García, J.; Navarro, M. y Ruiz, J. (2000). Planificación del entrenamiento depor-
tivo. Madrid. Gymmos.
García-Pallarés J, García-Fernández M, Sánchez-Medina L, Izquierdo M. (2010)
Performance changes in world-class kayakers following two different
training periodization models. Eur J Appl Physiol. 110(1):99-107.
González Ravé, JM, Arija, A, and Clemente-Suarez, V. (2011) Seasonal changes in
jump performance and body composition in women volleyball players. J
Strength Cond Res 25(6): 1492–1501.
González Ravé, JM; Juárez, D; Garcia, JM; Navarro, F (2007) Eficacia de la perio-
dización del entrenamiento sobre la potencia máxima. Archivos Medicina
del Deporte, 119, 179-186.
Gorostiaga, E. M., C. Granados, j. Ibañez, j. J. Gonzalez-Badillo, and M. Izqui-
erdo.(2006) Effects of an Entire Season on Physical Fitness Changes in Elite
Male Handball Players. Med. Sci. Sports Exerc., 38 (2) 357–366.
Granados, C., M. Izquierdo, J. Ibañez, M. Ruesta, E. M. Gorostiaga. Effects of an
Entire Season on Physical Fitness in Elite Female Handball Players. Med.
Sci. Sports Exerc., Vol. 40, No. 2, pp. 351–361, 2008.
Grosser, M. Brüggemann, P; Zintl, F. (1989) Alto Rendimiento Deportivo: Planifi-
cación y Desarrollo. Martínez Roca. Barcelona.
Hernandez Moreno, J (1988) Baloncesto: iniciación y entrenamiento. Paidotri-
bo. Barcelona.
Hoffman, J. (2002) Physiological aspects of sport training and performance.
Champaign (IL): Human Kinetics.
Howley ET; Franks, BD (1986) Health/fitness instructor handbook. Humankinet-
ics. Champaign. IL.
Hydren JR, Cohen BS. (2015) Current Scientific Evidence for a Polarized Cardio-
vascular Endurance Training Model. J Strength Cond Res. Dec; 29(12):3523-
30. doi: 10.1519/JSC.0000000000001197.
Issurin, V. (2008) Block Periodization versus traditional training theory: a re-
view. J Sports Med Phys Fitness; 48 (1): 65-75.
Issurin, V; Lustig, G (2004) L´effetto residuo di allenamento. Classificacione, du-
rata e componenti pratiche dell´effetto residuo di allenamento. Sds. Riv-
ista di cultura Sportiva, 60-61, 25-29.
Issurin,V (2010) New horizons for the methodology and physiology of training
periodization. Sports Med 2010; 40 (3): 189-206.
Issurin,V; Shkijar, V (2002). la struttura a blocchi dell´allenamento. La concezi-
one della struttura a blocchi dell´allenamento degli atleti de alto livello.
Sds. Rivista di cultura Sportiva 56:6-9.
Kell, JR (2011) the influence of periodized resisted training on strength changes
in men and women. J Strength Cond Res. 25(3): 735–744, 2011.
Kiely, J (2010) New Horizons for the Methodology and Physiology of Training Pe-
riodization Block Periodization: New Horizon or a False Dawn Sports Med
2010; 40 (9): 803-807.
Martin, D; Carl, K; Lehner, K (2001) Manual de metodología del entrenamiento
deportivo. Paidotribo. Barcelona.
Matveyev, L.P. (1977) Periodización del entrenamiento deportivo. Instituto Na-

48
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

cional de Educación Física y Deportes. Madrid.

Matveyev, LP (2001) Teoría general del entrenamiento deportivo. Paidotribo.
Barcelona.
Mujika I, Chatard JC, Busso T, Geyssant A, Barale F, Lacoste L (1995). Effects of
training on performance in competitive swimming. Canadian Journal of
Applied Physiology 20, 395-406.
Muñoz I, Seiler S, Bautista J, España J, Larumbe E, Esteve-Lanao J. (2013) Does
polarized training improve performance in recreational runners? Int J
Sports Physiol Perform. 2014 Mar;9(2):265-72. doi: 10.1123/ijspp.2012-0350.
Navarro, F (2000) La controversia actual sobre la planificación del entrenami-
ento. INFOCOES, 5(2), 55-69.
Navarro, F. (2010) Planificación del entrenamiento y su control. Ed: Cultiva Li-
bros. RFEN. Madrid.
Navarro, F. (2000) Principios del entrenamiento y estructuras de la planificación
deportiva. C.O.E. Madrid.
Norris, SR; Smith, DJ (2002) Planning, periodization and sequencing of train-
ing and competition: the rationale for a competently planned, optimally
executed training and competition program, supported by a multidiscipli-
nary team. In M. Kellmann (ed.) Enhancing recovery. Preventing underper-
formance in athletes. Champaign: IL. Pp 121-141.
Norris, SR; Smith, DJ (2002) Planning, periodization and sequencing of train-
ing and competition: the rationale for a competently planned, optimally
executed training and competition program, supported by a multidiscipli-
nary team. In M. Kellmann (ed.) Enhancing recovery. Preventing underper-
formance in athletes. Champaign: IL. Pp 121-141.
Núñez VM, Da Silva-Grigoletto ME, Castillo EF, Poblador MS, Lancho JL. (2008)
Effects of training exercises for the development of strength and endur-
ance in soccer. J Strength Cond Res. 22(2):518-24.
Platonov, VN (2000) El concepto de la periodización del entrenamiento y el de-
sarrollo de una teoría del entrenamiento. INFOCOES, 5(1), 87-93. Traduc-
ción de Leitungsport 1/1999.
Plisk, SS; Stone, MH (2003) Periodization strategies. Strength and Conditioning
J., 25 (6), 19-37.
Rhea, MR. and Aldeman, BL. (2004) A meta-analysis of periodized versus non-
periodized strength and power training programs. Reserch Quarterly for
exersice and sport. 75:413-422.
Seiler, K. S., Tonnessen, E. (2009). Intervals, thresholds, and long slow distance:
the role of intensity and duration in endurance training. Sportscience 13,
32–53.
Seiler, K. S. (2010). What is best practice for training intensity and duration dis-
tribution in endurance athletes? Int. J. Sports Physiol. 5, 276–291.
Selujanov, VN (1999) La clasificación científica-histórica del concepto de la pe-
riodización del entrenamiento y su crítica. INFOCOES, 4(1), 89-91. Traduc-
ción de Leitungsport 2/1999.
Seyle, H. (1956)The Stress of Life. New York: McGraw-Hill. 1956.
Stone, MH; O´bryant, HS; Garhammer, J. (1981) A hypotherical model for
strength training. J. Sports Med. Phys. Fitness, 21: 342-351.
Stöggl T, Sperlich B(2014) Polarized training has greater impact on key endur-

49
 sp or t science

ance variables than threshold, high intensity, or high volume training.

Front Physiol. 4;5:33. doi: 10.3389/fphys.2014.00033.
Stöggl TL, Sperlich B (2015) The training intensity distribution among well-
trained and elite endurance athletes. Front. Physiol. 6:295. doi: 10.3389/
fphys.2015.00295
Verchoshanskij, Y; Siff, M (2000) Superentrenamiento. Paidotribo. Barcelona.
Verkhoshanski, Y (1990) Entrenamiento deportivo. Planificación y progra-
mación. Martínez Roca. Barcelona.
Wathen, D (1994) Periodization: concepts and applications. In Baeche, TR (eds)
Essentials of strength training and conditioning. NSCA. Humankinetics.
Champaign. IL. pp 459-472.
Wathen, D., T. R. Baechle, and R. W. Earle.(2000) Training variation: Periodiza-
tion. In: Essentials of Strength Training & Conditioning (2nd. ed.). Baechle
and Earle (eds.). Champaign, IL: Human Kinetics. PP. 513–527.
Wathen, D; Roll, F (1994) Training methods and modes. In Baeche, TR (eds) Es-
sentials of strength training and conditioning. NSCA. Human kinetics. Pp
403-415.
Willoughby, DS. (1993) The effects of meso-cycle-length weight training pro-
grams involving periodization and partially equated volumes on upper and
lower body strength. J. Strength Cond. Res. 7: 2-8.
Yu H(1), Chen X, Zhu W, Cao C (2012) A quasi-experimental study of Chinese top-
level speed skaters’ training load: threshold versus polarized model. Int J
Sports Physiol Perform. 7(2):103-12.

50
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Development of talent of adolescents in

Australian sports high schools
Ross Sanders, Rhonda Orr, Chuang-Yuan Chiu, Angus Chard, Claire
Tompsett, Jennifer Fleeton, Timothy Davies, Alycia Fong Yan 1

Introduction
There are seven high schools in NSW designated as ‘sports’ high
schools. They offer a special curriculum to enable students with high
potential in specific sports to have additional coaching and talent devel-
opment opportunities. Entry to the ‘sport stream’ of these schools is se-
lective and competitive. The students are selected based on their sport-
ing background and interview. The program has been developed by the
Discipline of Exercise and Sports Science (ESS) in the Faculty of Health
Sciences, The University of Sydney, in response to an approach by the
Principals of the sports high schools. Ensuing discussions between two
Principals and the sport scientists of ESS identified the following issues
to be addressed to achieve the schools’ mission of enabling students to
fulfil their academic and sporting potential: (i) students sporting careers
being affected by injuries which may be prevented through attention to
predisposing factors early in their development; (ii) limitations to ongo-
ing success and improvement not being identified and rectified; (iii) stu-
dents with potential not being recognized and nurtured due to chrono-
logical and biological age differences among students in the same age
division; (iv) students not selecting, or being streamed into, sports and
positions that make the best use of their attributes.

Development of the Program

The program is designed to be longitudinal in nature, addressing the
needs of individual students over a five-year period commencing from
each student’s first year of high school (Yr 7 - age ~13 years) to their pe-
nultimate high school year (Year 11 – age ~17). Assessment of their cur-
rent status in various attributes related to performance is conducted
three times per year. This provides a basis for selection of supplemen-
tary activities during their training sessions to address their develop-
mental needs.

1
Discipline of Exercise and Sport Science, Faculty of Health Sciences, The Uni-
versity of Sydney, Sydney, Australia

51
 sp or t science

1. Assessment of Injuries and Injury Risk

Injury in adolescent years may negatively impact longer-term health
and wellbeing as well as potential representative and/or professional
careers (1, 2). Assessment of injuries and injury risk comprises ongoing
injury surveillance and measurement of joint characteristics that are
known to be associated with injury. The injury surveillance includes a
questionnaire of injury history prior to the time of entry to the program
in Year 7 and injury status at that time. Demographic information (age,
ethnicity), medical history, medications and supplements and athlete
sport-specific characteristics (playing history, playing position, protec-
tive gear) was also collected. Ongoing surveillance is achieved through a
standardised monitoring process.
Injury was defined as ‘….any physical or medical condition that oc-
curs during participation in match or training activities that results in a
missed match or training participation’ (3). Match and training injuries
(including concussions) are recorded at each game and training session
in a standardised format using the Athlete Injury Surveillance Profile (4).
Factors pertaining to injury include:

1) Characteristics
2) Onset
3) Diagnosing/Treating Person
4) Treatment given
5) Injury Follow-up
6) Modified training required and training time and/or match time lost

Presentations are conducted to familiarise sports convenors and

coaches to the study objectives and data collection procedures. Sport
convenors or coaches complete the injury recording form which records
the: body location; type; mechanism; occurrence (date, match vs. train-
ing; first vs. second half); environmental conditions (weather, ground,
indoor); initial diagnosis; initial treatment (e.g. ice, strapping); treating
person; and referrals pertaining to the injury (e.g. hospital, physiothera-
pist, doctor). To maximise consistency and reduce reporting bias, forms
are to be completed as soon as possible after the injury and by the same
individual for each sport, where possible.
An injury follow-up form (subsequently completed by the physiother-
apist or medical practitioner and sport convenor) documents the: injury
diagnosis; further referral (e.g. physiotherapist, diagnostic scans, sur-
geon); number of modified and/or missed training sessions; and number

52
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

of missed matches. Injury severity is classified according to the number

of matches missed (0 = transient; 1 = mild; 2-4 = moderate; ≥5 = major).
Injury incidence calculations (per 1000 player hours) are in accordance
with King et al. (5).
The surveillance data are assessed in conjunction with the measure-
ment of joint characteristics including stiffness, mobility, configuration
and posture. These have been identified as important in the literature
and are described below:

Foot Posture Index (FPI)

The FPI is the gold standard for differentiating and quantifying foot
posture. This easily implemented in-field tool considers six different as-
pects of standing foot posture as seen in Table 1, allowing the clinician to
quantifiably differentiate and categorise individual foot postures, as seen
in Table 2, from one another. This method can distinguish at risk groups
considering biomechanical risk factors that may allow for the prevention
and early intervention of musculoskeletal conditions (6-8) and can help
to predict dynamic foot function (9). Features the FPI included reliability,
implementation simplicity, quantitative scoring related to the complex-
ity of foot function, minimal subjectivity and easy field implementation
without needing sophisticated measuring equipment (10). As foot pos-
ture measures taken during weight-bearing are more representative of
dynamic foot function than non-weight bearing measures (11), the as-
sessment was devised to take place during quiet, double-limb-supported
standing. The resulting system, the foot posture index (FPI) considers
rearfoot sagittal, frontal and transverse plane positions, as well as mid-
foot sagittal and transverse plane foot positions (Table 1).

Joint Plane Test

Rearfoot transverse talar head palpation
frontal/ transverse curvature at the lateral malleoli
frontal inversion/eversion of the calcaneus
Forefoot transverse talonavicular bulging
sagittal congruence of the medial longitudinal arch
transverse abduction/adduction of the forefoot on the
rearfoot

Table 1. The six-item foot posture index (FPI) tests and their relationships to
joint and plane of motion

53
 sp or t science

Keenan, Redmond (12) transformed the FPI scores to interval logit

scores in accordance with the Rasch model and normal ranges of foot
posture were defined for 619 healthy individuals ranging in age from
3—96 years. From these data, a slightly pronated foot posture of +4 was
considered healthy, with normal foot posture achieved by the age of
eight. Scores either side of the neutral point indicated potentially abnor-
mal and truly abnormal ranges (13) Table 2).

Truly Potentially Potentially Truly

Normal range
pathological abnormal abnormal pathological
Highly Highly
Supinated Healthy Pronated
supinated pronated
-1 +1
< -2 SD -2 SD Mean +2 SD > +2 SD
SD SD

FPI raw
< -3 -3 +1 +4 +7 +10 > +10
score

Table 2. FPI scores and their relationship to healthy normal foot posture

The FPI has been subjected to thorough validation. It proved to be an

adequately reliable assessment of adults in various clinical settings, with
intra-class correlation coefficients (ICC) of 0.62—0.91 (14-17). In children
and teenagers aged 5—16 years, inter-rater reliability was almost perfect
(Kw = 0.88; (18)). More recently, in children aged 7—15 years, the FPI was
found to have good intra-rater reliability using the Portney and Watkins
criteria (19) with ICC = 0.93—0.94 (20). The finalised FPI instrument pre-
dicted 64 % of the variance in stationary standing foot posture and 41 %
of the dynamic walking midstance foot posture (10).

FPI as a Pathology Predictor

The FPI is a valid, reliable, easily-implemented, clinically-relevant
descriptive foot posture tool that can distinguish at-risk groups. It can
distinguish foot related biomechanical risk factors, which may allow
prevention and early intervention of lower limb musculoskeletal condi-
tions. The FPI is sensitive to distinguishing musculoskeletal pathologies
with evidence from several populations:

ғ
Naval recruits with a pronated foot postures (FPI ≥ +6) were significantly
(P = 0.002) more likely to develop medial tibial stress syndrome (MTSS)
than recruits with normal (+1 to +5) or supinated (0 to -16) foot postures,
with a risk ratio of 1.70 (21).

54
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

ғ
Adult participants with highly supinated foot postures reported greater

incidence of foot pain (60%) compared to participants with normal foot
postures (23 %, FPI 0 to +6; P = 0.009). Pressure–time integrals under
the whole foot, rearfoot and forefoot were higher than in normal feet
(P < 0.01). Pressure–time integrals in subjects reporting foot pain were
higher than for pain-free subjects (P < 0.001). There was a significant
correlation between pressure–time integral and foot pain (r = 0.49, P <
0.001) (6).
ғ
Triathletes with highly supinated foot postures (FPI ≤ -2) had a signifi-
cantly greater likelihood of overuse injuries during the competition sea-
son. Risk of injury occurring in this foot type was 4.3 times greater; P =
0.013, than those without (7).
ғ
Adults with chronic plantar heel pain were found to be more likely to
have a pronated (FPI ≥ 4) foot posture (OR = 3.7, 95% CI = 1.6 – 8.7, P <
0.01) (22).

Function Predictor
ғ
Foot motions of healthy adults aged 18—47 can be predicted during
normal walking with the static FPI. In the case of pronated and highly
pronated participants (FPI > +6), reduced midfoot frontal plane range of
motion was seen in late stance. In supinated and highly supinated par-
ticipants, less midfoot motion was seen during initial contact and mid-
stance, and altered frontal and transverse plane motions were observed
when compared to control participants with normal FPIs (23). Adults
ranging from 18—71 years of age displayed a greater lateral centre-of-
pressure excursion, correlating with greater supinated foot postures
and, conversely, the more pronated the foot posture, the smaller the
area of lateral centre-of-pressure excursion. In addition to this, the supi-
nated foot type had a larger centre-of-pressure total excursion area, and
the pronated foot type had a smaller centre-of-pressure total excursion
area.

Ist MPT Joint Stiffness

First metatarsal phalangeal (1st MPT) joint extension is necessary for
healthy foot function (24, 25). Restricted sagittal plane motion, that is,
less than 64 ° (26) can be an important predictor of injury (27-29) and foot
function (30-32). Using a hand-held goniometer, the sagittal plane axis of
the 1st metatarsal and proximal phalanx are determined and measured.
Restricted extension (<64 °) or signs of joint derangement (26), including;
‘hard end point’ at maximum extension, dorsal joint lipping with bony

55
 sp or t science

growth (exostosis) formation, pain with dorsomedial palpation or valgus

deviation greater than 15 ° would be recorded (33, 34). Intra-rater reli-
ability has been determined as 0.99 (35). Inter-rater reliability has with
an inter correlation coefficient of 0.95 has been determined (26, 33).

Lunge Test
Restricted ankle dorsiflexion has been identified as a contributing
factor in overuse injuries of the lower limb and foot (36, 37) and strong
predictor of injury (p = 0.03) (38). Normative values have been reported
(38). This test has been demonstrated to have an intra-tester reliability of
ICC = .98 (SEM = 1.1 °) and an inter-tester reliability of ICC = .99 (SEM = 1.4
°) (39) for adults and for children aged 7-15 years of age ICC = 0.85-0.95.
Prior to determining the ankle lunge measurement, the tibia must
be marked for the standardised measuring location. This is done with
the participant lying supine. The tibial tuberosity is first isolated and
marked. Using a 10 cm ruler a point is marked distal to the tibial tuber-
osity on the most lateral bony ridge of the anterior tibia. The participant
aligns their 2nd toe to a line on the floor to the anterior and heel bisec-
tion to the posterior, then places the contralateral leg to a line on the
floor and assumes comfortable easy rest position. Placing the test legs
corresponding hand on a chair back, for balance, the participant steps
forward flexing the test leg in the sagittal plane. The examiner takes care
to ensure femoral abduction/adduction is negated as the knee cap flexes
over the 2nd toe. The participant flexes and straightens their knee three
times before the measure is taken with knee maximally flexed will main-
tain full heel contact. The examiner aligns the long edge of a digital incli-
nometer (iPhone 6 or equivalent using the application “iHandy-Level™”)
to the most lateral ridge of the anterior tibia with the midpoint of the
inclinometer aligned with the existing mark already determined. The an-
gle is measured to the supporting surface and recorded. (22, 39).

Beighton Scale
The Beighton scale is a commonly used valid and reliable clinical
measure of generalised joint hypermobility or ligamentous laxity (40, 41)
and normative data exists for athletes. Generalised joint laxity was not
shown to be significantly elevated in adolescent male athletes (42) but
females were found to be significantly more hypermobile than males
during puberty (43). Aside from a greater risk of knee injuries in contact
sports among those with hypermobility, there is currently insufficient
evidence associating hypermobility with injury risk (44). The Beighton

56
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

scale (41) is used to ascertain the presence of joint hypermobility at the

wrist, fifth metacarpal phalangeal joint, elbow, knee (all bilateral and
non-weight-bearing) and the lumbo-sacral spine (forward flexion, in
stance). The Beighton scale yields a score related to static postures of
the upper limbs, and hips, whereby the arbitrary cut-off of 5/9 or greater
for females and 4/9 for males indicates joint hypermobility (41).

Lower Limb Assessment Scale

Since the Beighton score is largely focused on the upper limbs, the
Lower Limb Assessment scale was developed to assess lower limb hy-
permobility (45). This scale includes 12 items, with separate scores for
the right and left legs (45). A majority of the items require passive ma-
nipulation of the joint to assess the joint laxity. The scale is valid and
reliable and was able to predict generalised hypermobility in a greater
percentage of people than the Beighton Scale (45). Clinically, both LLAS
and Beighton Scale are used together. In children, there is no significant
difference between right and left legs prior to puberty (45), therefore, it
is possible to test one leg if expediency is required.

2. Identifying and Improving Attributes

Related to Success
The school coaches work to develop the sport specific skills through
their various skills practices. However, development of these skills and
their execution in game situations can be affected by various limitations.
To address the issue of ‘limitations to ongoing success and improvement
not being identified and rectified’ we have structured an assessment
and remediation program around several categories of potential limita-
tions including ‘basic human movements’, ‘strength and conditioning’,
‘balance’, and ‘flexibility’.

Basic Human Movements

Research has revealed that despite years of practice and develop-
ment of sport specific skills, ongoing performance levels and skill de-
velopment may be affected by poor development of basic human move-
ments (BHM) in both primary and high school athletes (46). These are
considered the pre-requisite essentials underpinning control of more
complex sports skills such as running, jumping, catching, striking and
throwing skills (47, 48). Deficits in BHMs are often unnoticed and unex-

57
 sp or t science

pected among those who, to all appearances, are ‘sporty’ and well-co-
ordinated.
Sport training with appropriate conditioning reduces risk of injury in
young athletes (49). Documented intrinsic risk factors have suggested
neuromuscular control (50), core instability (51), and contralateral mus-
cular imbalances (52) are related to injury risk. Compensatory move-
ment patterns utilized to achieve complex skill performance are often
overlooked in the young competitive population. However, these ineffi-
cient movement systems may reinforce poor biomechanical movement
patterns during typical activities, resulting in injury (53). The BHM as-
sessment tool is used to detect compensatory movement patterns and
core instability in young athletes and can be used for early intervention
to reduce injury risk. To assess the development of BHMs, participants
demonstrate a squat, lunge, push up, pull up, hinge (bend), rotation and
brace (plank) which are scored according to the criteria illustrated in Fig-
ure 1 (54).

SQUAT Weight evenly distributed between the legs during ascent and
descent.
Gaze remain forwards or up throughout.
Heels remain in contact with the ground throughout.
Spine remains in a parallel position with the shins.
Knees reach at least parallel with the hips.
LUNGE Foot lands directly in front of the hip.
Knee tracks over the foot while the shin remains vertical.
Hips remain in line horizontally.
Smooth and balanced transition between steps.
Back foot remains facing the front.
Trunk remains upright throughout the lunge.
HINGE Weight distributed evenly between sides throughout the
movement.
Spine maintains a natural curve during the ascent and descent.
Shoulder blades are retracted to maintain natural curve of upper
back.
Head remains in neutral alignment with the spine.
Knees remain at a fixed angle, i.e. do not bend/straighten
throughout the movement.
ROTATION The feet and knees remain facing the front.
Body remains vertically aligned.
Shoulders remain in horizontal alignment throughout rotation.
Rotates at least 90° each side.
Spine maintains neutral alignment.

58
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

PULL UP Weight evenly distributed between the hands during the ascent
and descent.
Shoulders remain square during the ascent and descent.
Trunk maintains a rigid position throughout the movement.
Trunk maintains a rigid position throughout the movement.
Chest is pulled until body is horizontal to the ground.
PLANK Shoulders are horizontally aligned.
Weight is evenly distributed between sides.
Shoulders align with the trunk through to the toes.
Head is held in line with the spine.
Correct position is held for at least 45 s.

PUSHUP Weight evenly distributed between the hands during the ascent
and descent.
Shoulders remain square during the ascent and descent.
The elbows flex to at least 90° to the ground at the bottom of the
movement.
Spinal column maintains straight alignment from the base of the
neck through to the knees.

Figure 1. Description of Basic Human Movements.

Strength and Conditioning

Counter Movement Jump
The counter movement jump (CMJ) is a vertical jump for maximum
height in which a counter movement (flexion of the hips and knees
immediately prior to the jump) is permitted. The CMJ is a widely used
field-based test to assess muscular power of the lower body (55). Stud-
ies have shown that performance in the vertical jump test is able to dis-
tinguish between athletes that compete at various levels in a variety of
sports such as Australian Rules football (56, 57) and rugby league (58,
59). The performance of CMJ can be assessed using a contact platform
(e.g. SmartJump; Fusion Sport, Coopers Plains, Australia) to determine
the time between take-off and landing (i.e. flight time) from which the
height jumped is estimated using the formula for projectile motion. An
alternate method of assessing jump height is the Vertec (Questek) meas-
uring device, which assesses vertical jump height by measuring the dif-
ference between the athlete’s standing reach height and their maximal
jump and reach height.
Portable contact platforms such as SmartJump (Fusion Sport, Coop-
ers Plains, Australia) have been shown to produce reliable measure-
ments with errors of no more than ~ 3.0 % for countermovement jumps
(60). To improve reliability of CMJ height assessment it is recommended
that all athletes are provided with thorough familiarisation and complete

59
 sp or t science

a minimum of three trials, with the maximal CMJ height reported. In the
case of the Vertec measuring system, a separate familiarisation session
is recommended (55). Intrasession reliability of the countermovement
jump height measured with the Vertec device has been found in females
ICC = 0.89, SEM = 2.1 cm, CV 6.9 %; and in males ICC = 0.94, SEM = 2.2cm,
CV = 5.5 % (55) Intersession reliability for both males and females com-
bined ICC = 0.80, SEM = 2.7 cm, CV = 8.6 % (55).

Medicine Ball Chest Throw

The medicine ball chest throw is a simple field test to measure up-
per body power (61). Upper body power is an important aspect in many
sports, from throwing based field sports such as javelin to fending tack-
les in rugby league. The student throws a 3 kg medicine ball while sit-
ting on an upright bench with feet flat on the floor, knees at 90 º flexion,
while their head, shoulders, and lower back remains against the back-
rest of the bench. Each student has two practice attempts for familiarisa-
tion and warm up prior to performing three trials with 15 s of recovery
between trials. The score is the greatest distance thrown of the three
throws measured to the nearest 0.1 m.
Garrido, Marinho (62) found that this test had a strong correlation ex-
plaining 25-49 % of the variance in swim times of national level youth
swimmers. The medicine ball chest throw has very high test-retest reli-
ability with an ICC of 0.92 (62).

Sprint Speed and Acceleration

Sprint speeds are used alongside other measures such as leg power
and agility to distinguish athletic talent (63). Sprint testing is common-
ly included in the test battery for athletes involved in team field-based
sports such as rugby league, rugby union, Australian Rules Football and
basketball (64-67). It is assumed that a distance of 10 m can be used to
assess an individual’s ability to accelerate, whereas maximal velocities
are achieved at 20-40 m (64, 65, 67).
Subjects perform maximal effort sprints over 20 m and are timed at
5 m, 10 m and 20 m. The timings can be obtained using timing gait sys-
tems such as ‘SmartSpeed’ (Fusion Sport, Coopers Plains, Australia). Af-
ter a standardized warm-up (including jogging, dynamic range of motion
exercises, and submaximal sprints), subjects perform three maximal
sprints with a 3 min rest between attempts. Subjects commence each
sprint from a 2-point standing start (dominant foot placed anteriorly of
the non-dominant foot, ~ 30 cm apart, the arms do not make contact

60
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

with the ground) 30 cm behind the first timing gate. Subjects commence
in their own time and sprint at their maximum pace through the 20m
timing gate and are told to continue running through markers situated
~ 2 m beyond the 20 m timing gate before decelerating. The best of the
three times is used as their total 20 m sprint score, timed to the nearest
0.01 s, with split times from 0-5m and 0-10m used to determine accelera-
tion ability (65). Good reliability has been shown with sprint times for
10, 20, and 40 m in semi-professional rugby league players (typical error
ranging from 1-2 %) (68).

505 Change of Direction Speed Test

The ability to repeatedly sprint and change direction while sprinting
is a determining factor of performance in many field and court sports
including rugby, field hockey, tennis, basketball and soccer (69). Scan-
lan, Humphries (70) described agility performance as dependent on not
only the physical aspects of an athlete (technique, linear speed, muscu-
lar qualities and morphology), but also perceptual and decision making
components (visual scanning, situational knowledge, pattern recogni-
tion and anticipation). Thus to be considered a true assessment of agil-
ity performance a field based test must be an open skill, requiring an
unrehearsed reaction to a stimulus, as well as a change in speed or direc-
tion (69). Within the time constraints of the current program, where as-
sessments were carried out within normal class times necessitating the
assessment of up to 30 athletes within as little as 36 min, an assessment
involving the perceptual and decision making aspects of agility was not
possible. Thus the 505 change of direction speed (CODS) test was used
as it is a time-efficient, valid and reliable test to assess the neuromuscu-
lar and biomechanical aspects of an athlete’s ability to rapidly change
direction (71, 72).
Following a standardised warm up as described above, athletes start
in a 2-point standing start position and sprint forward through timing
gates placed at 10m to a marked turning line at 15 m. Athletes plant their
foot on the turning line (using their preferred foot), turn 1800 and sprint
the 5 m back through the timing gates. Participants should be allowed
three practice trials, followed by three timed trials with 3 min rest be-
tween each trial. The fastest of the three trials, recorded to the nearest
0.01 s, is used for analysis. Male and female combined ICC = 0.88, SEM =
0.06 s, CV = 2.40 % (72).

61
 sp or t science

Multistage 20 m Shuttle Run Test

The maximal multistage 20 m shuttle run test (MST), commonly re-
ferred to as the ‘beep’ test, is a field test commonly used to estimate
maximal oxygen uptake (VO2max). The MST is comprised of 23 levels, with
each level lasting ~ 1 min (73). The starting speed is 8.5 km/hr, increasing
by 0.5 km/hr at each level thereafter. A participant’s score is the last level
attained prior to failing to reach the marker before the ‘beep’ on two suc-
cessive occasions, or when participants reach volitional fatigue.
Strong correlations have been found (r=0.92) between MST perfor-
mance and VO2max as measured in a laboratory. Flouris, Metsios (74) de-
veloped an equation to predict V02max from MST scores, incorporating
data collected via indirect calorimetry while subjects performed the MST
thus improving the efficacy of this test compared to earlier prediction
models. Standard error of the estimate using this equation was found to
be 1.9 ml.kg-1.min-1.
EQTT VO2max = (MAS x 6.65 – 35.8) x 0.95 + 0.182
(EQTT = equation to predict treadmill test VO2max; MAS = maximal at-
tained speed in the MST (i.e. final shuttle speed)) (74).
VO2max is an important measure in sports performance, as many sports
require a minimum threshold of aerobic capacity for performance at an
elite level (59, 75, 76). Strong correlations have been found between MST
performance and match performance parameters (e.g. sprinting, total
distance covered etc.) in national soccer players (77). The MST has been
found to be a reliable test for both children (r = 0.89) and adults (r = 0.95)
(78).

Yo-Yo Intermittent Recovery Test level 1 (Yo-Yo IR1)

The Yo-Yo Intermittent Recovery test level 1 (Yo-Yo IR-1) is a measure
of sport-specific aerobic endurance and is related to an athlete’s ability
to perform prolonged, high intensity intermittent running as required for
many field-based team sports such as Rugby League (79) and Rugby Sev-
ens (80). It is more sensitive in discriminating between different levels
of athletes and in evaluating the results of training interventions than
V02max and is a more accurate predictor of on-field performance (80). Fol-
lowing a standardised warm up, athletes perform two 20m runs back
and forth at a progressively increasing speed, controlled by a series of
audible signals played over a portable sound-system. Between each run-
ning bout players have a 10-second active rest period consisting of 2 x 5
m of jogging ‘out and back’ around a marker. The total distance covered
by each athlete is recorded as their score (79).

62
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Using the Yo-Yo IR-1, Deprez, Coutts (81) were able to discriminate
between elite, sub-elite, and recreational youth soccer players aged
between 11-17 years. Intersession reliability for U13, U15 and U17 age
groups were ICC = 0.82, 0.85 and 0.94 with CV = 17.3 %, 16.7 % and 7.9
%, respectively. Age-related normative data are available for elite and
non-elite youths (81).

Flexibility
Sit and Reach Test
The sit and reach test is used to determine hamstring and lower back
flexibility (82) and is a common field-based assessment (83). Lack of
hamstring muscle extensibility has been shown to reduce the mobility
of the pelvis (84), which can increase the susceptibility of individuals to
musculoskeletal injuries (85, 86).

Figure 2. Demonstration of the Sit and Reach Test using the Flex-Tester® (Novel
Products Inc., USA)

The testing apparatus used was the Flex-Tester® (Sit and reach) (Nov-
el Products, Inc., USA). Scale three (AAHPERD) was used to measure the
reach distance. This sets the feet nine inches away from the zero on the
scale. High reliability has been found for the sit and reach test (4.48 %
typical error; 0.84 % change in the mean, 0.95 ICC) by Ayala et al., 2011.
Following a set dynamic warm up, the participant sits on the floor with
knees straight and feet pressed firmly against the box and medial foot
alignment vertical with ankles at 90°. With the arms reaching forward,
one hand placed on top the other, and palms faced down the participant
reaches forward in one smooth motion, holding the final reach position
for 2-3 s while the distance is recorded (87) (Figure 2). Two trials are per-
formed with 10-15 s between trials. The average of the two trials is re-
corded.

63
 sp or t science

Balance
Foot Lift Test
The foot lift test is a measure of static balance, and has good test-
retest reliability (Pearson’s R=0.78; ICC2, 1 = 0.73, 95 % CI 0.40-0.89) (88).
Results of the foot lift test have been associated with chronic ankle insta-
bility and thus a risk for ankle injuries like ankle sprains (88, 89). The par-
ticipant stands on one leg with the contralateral foot touching the calf,
hands on hips and then closes the eyes. As soon as the eyes close the
participant must maintain their single leg stance for 30 s. The number
of times any part of the foot lifts of the ground, the person takes a step,
and for each second the non-supporting leg is in contact with an error is
recorded. If the participant opens their eyes then the trial does not count
(88). Two trials are performed and the score is the number of errors in
the trial with the smaller number of errors (89).

Anthropometry
Anthropometric tests provide information regarding a player’s body
dimensions, shape, and body composition as well as an estimation of
his or her adult height. The players are tested in close-fitting shorts and
sports tops (for female subjects).

Stature
Stature is one of the characteristics related to suitability for different
sports and positions and is essential in adult height prediction (90) and
in understanding body size and shape (e.g. somatotype scoring (91)).
Body size and shape has been found to be related to sports performance
(92). For example, being tall and slender is important in swimming as it
effects the hydrodynamic resistance encountered by the swimmer (93)
while short stature is an advantage in gymnastics to minimise rotational
inertia (94). Thus, stature is an important factor in talent identification,
athlete selection and sports development (95). Further, stature and so-
matotype are associated with suitability for different sports and posi-
tions (91, 96) and are strong determinants of the best playing position in
sports such as AFL, basketball, netball, soccer, field hockey, rugby league
and rugby union.
Stature is measured using the ‘self-stretch stature method’ with a
height scale or stadiometer which is adapted from the traditional stretch
stature method (97). The participant stands against a vertical wall or
stadiometer, touching the wall with the heels, buttocks, and back with
heels together. The head is aligned so that the upper border of the ear

64
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

opening and the lower border of the eye socket are on a horizontal line.
The participant is instructed to stretch upward and to take and hold a
full breath. The headboard of the stadiometer is moved downwards un-
til it firmly touches the vertex of the head. The measurement of stature
is taken in mm at least two times for each subject. Reliability of stature
should be within 1 % (98).

Body Mass
Body mass is part of the assessment of characteristics related to suit-
ability for different sports and positions. It is used in combination with
stature and girths to determine body composition. It is also used in com-
bination with age, stature and sitting height to predict adult height. Par-
ticipants step onto clinical scales graduated to 0.1 kg. At least two trials
are recorded. Reliability of body mass should be less than 1 % (98).

Girths
Girths of the neck, waist, and hips are measured as part of the as-
sessment of characteristics related to suitability for different sports and
positions. Girths are used in combination with stature and body mass
to determine body composition. Anatomical landmarks are identified by
palpation by a certificated anthropometrist. A tape measurement is tak-
en by following the rules of International Society for the Advancement
of Kinanthropometry (97). Each girth measurement is taken in mm at
least two times. Reliability of girth measurements should be less than 1
% (98).

Body Composition
Body composition, the quantity of fat as a percentage of body mass,
is estimated by a formula (99, 100) that includes stature, body mass,
and girths collected from anthropometric tests. Body composition is re-
lated to sport performance (101). From a biomechanical aspect, body
fat masses cause extra energy consumption and has a negative impact
on sport performance (102, 103). Given two players with the same body
mass, the one with lower body fat percentage should generally be able
to sustain movement at a higher level and for a longer period than the
one with higher body fat percentage. In addition, Bahr and Krosshaug
(104) indicated that body composition may be a factor which influences
the risk of sports injury.

65
 sp or t science

Nutrition Assessment
Nutrition assessment is conducted in this cohort to assess differ-
ent categories of food intake in adolescent athletes in order to moni-
tor the dietary consumption patterns that may impact performance in
their sport. Adequate energy intake is essential for adolescent athletes
to support proper growth, maturation and development with possible
increased intake to match exercise training and competition require-
ments. Dietary patterns and hydration strategies should reflect their
daily exercise demands. Adolescence is a period in the lifespan that can
present challenges to appropriate eating patterns and food selection.
In particular, poor dietary habits can limit performance by affecting en-
ergy reserves, muscle and bone development, and body composition.
Depending on the sport and playing position, players can have too low
or too high percentage of fat. For example, some body fat is desirable for
swimmers as it assists buoyancy and body alignment when swimming.
Rugby props need bulk and strength and benefit from having a higher
percentage of fat than players in other positions. However, too much fat
reduces their speed and endurance in open play.
The Students As Lifestyle Activists (SALSA) Questionnaire is a short,
straightforward questionnaire for use in adolescents where individual
items have been validated (105-108). The SALSA Questionnaire estimates
daily consumption of breakfast, fruit, vegetables, water, sugary drink in-
take (i.e. fruit juice, soft drink, sport drinks, cordial, and energy drinks),
and intentions of future consumption of these items. In addition there
are simple measures of daily physical and sedentary activities and rec-
reational screen time. The SALSA Questionnaire has been modified from
the Short Food Frequency Questionnaire (SFFQ) (109). The SFFQ was re-
ported to have fair to moderate validity and moderate to good reproduc-
ibility in primary school children aged 10-12 years. Validity and reliability
studies of the SALSA questionnaire are currently being conducted.

3. Students with potential not being

recognized and nurtured due to chronological
and biological age differences among students
in the same age division.
Age
Age is calculated in days from dates of birth obtained in a question-
naire completed by parents. This chronological age is then used in com-

66
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

bination with stature and sitting height to predict adult stature. Age is
essential in adult height prediction (90).

Biological Age and Adult Height Prediction

Adult height prediction (AHP) provides an estimate of the final stat-
ure and may be related to young athletes’ potential in particular sports.
Thus, AHP is widely used for talent identification, athlete selection and
sports development (95).
It is also important for coaches to understand the effects of early and
delayed maturation on team selection and playing position selection.
For example, players who mature early, being taller than peers, may be
trained as forwards in basketball or lineout jumpers in rugby but may
not be appropriate for those positions as adults due to reaching their
final stature being less than later maturing peers. Similarly, those who
mature later than peers miss opportunities due to their smaller stature,
mass, and muscle development. This is particularly the case where stat-
ure, mass, and muscularity are important, for example in rugby league
and rugby union. It is well known that swimmers who mature early have
an advantage over later maturing peers due to resistance being related
to body length and propulsion also being aided by long limbs. Trainabil-
ity, for example muscle development, aerobic and anaerobic capacity,
increases with biological age into adulthood.
Additionally, team selection is biased towards the ‘big kids’, par-
ticularly in sports where size confers an advantage. Indeed, this disad-
vantages those who mature late as well as those whose birthday falls
close to the cut-off of the age division. Thus, those who are biologically
or chronologically disadvantaged may miss the opportunity to play at
a higher level and to enter the ‘development pathway’ despite having
good potential and skill development. A further adverse possibility is
that athletes with potential who cannot maintain performance relative
to peers due to late maturation become discouraged. Indeed, this has
been recognised as a major cause of ‘dropout’ (110).
In this talent development program, the prediction equations of
Sherar, Mirwald (90) are applied to estimate AHP and also to classify in-
dividuals as ‘early’, ‘average’ or ‘late’ in terms of their stage of maturation
relative to chronological age. The estimates are based on the ratio of sit-
ting height to standing height which changes with maturation due to the
difference in timing of lower limb and trunk growth (lower limb growth
precedes trunk growth during adolescence).

67
 sp or t science

Sitting Height
The approach for measuring sitting height is adapted from the tradi-
tional method (97). To determine sitting height, players are seated on a
measuring box or level platform with their hands resting on their thighs.
The player is then instructed to take and hold a deep breath and while
keeping the head in the Frankfort plane the measurer applies gentle up-
ward lift through the mastoid processes. The head board of the stadiom-
eter is placed firmly down on the vertex, compressing the hair as much
as possible. The measurement of sitting height is taken in mm at least
two times. Reliability of sitting height should be less than 1 % (98).

4. Students not selecting, or being streamed into, sports and po-

sitions that make the best use of their attributes.
In addition to the effects of biological age and anthropometric vari-
ables there are many other factors that affect an individual’s choice of
sport and playing position. In selecting sports, children are influenced
by parents, peers, teachers, and others. Many will take up the sport that
their parents played or that their friends play. Knight (111) summarised
the influence of parents, siblings, and coaches, on ‘starting sport’, ‘pro-
gressing in sport’, and ‘excelling in sport’ indicating the relative influ-
ences of each of these by the thicknesses of the lines in a diagram. These
influences commonly dominate the selection of a sport despite, in many
cases, the individual having attributes that would be better suited to a
different sport.
The measurement of attributes of individuals in this program, and
the consequent comparisons with norms of elite performers in particu-
lar sports, may enable identification of potential in a sport other than
the one in which they are currently engaged. For example, a gymnast
who is destined to become taller than those in the range of successful
performers may be encouraged to take up a sport which uses many of
the attributes already developed, for example, dance. Similarly, within
a sport, an athlete might be identified as having potential to play in a
position that is more suited to their set of attributes. For example, a large
and muscular rugby player who is fast and has explosive power may be
better suited to being a ‘blockbusting’ winger than a forward where the
muscles, although strong for short bursts of activity, are not suited to
enduring long periods of repeated scrummaging.

68
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

How the Program will Contribute to Optimising Sports Perfor-

mance in the Future
The program is yielding a large amount of data that are both age
specific for chronological age as well as biological age. From these data,
norms are determined which provide potential sports persons with real-
istic targets to which they can aspire and a means of assessing their pro-
gress towards elite level. The norms can be used as generic to all sports
and also for specific sports and playing positions. Having chronological
age and biological age references norms will help coaches and selectors
to recognise, and continue to develop, potential despite current per-
formances being below those of early maturing peers. This will help to
reduce ‘dropout’ and those with potential ‘slipping through the cracks’.
The data will also support initiatives to organise competition categories
based on biological, rather than chronological age or on the basis of an-
thropometric compatibility, for example, weight division for youth rugby
league and rugby union.
Ongoing analysis of the results will help identify correlations between
variables that could be associated with sport performance. Predictors
of injury can also be identified through ongoing injury surveillance and
repeated testing of these suggested variables. Currently the database is
being built from the seven sports high schools in NSW. However, there is
potential to combine these with data from other sources and from other
researchers. A large global data base that is available to athletes and sci-
entists internationally is an ultimate goal and realistic vision given the
modern technological capacity for information storage and access.

Conclusion
The program described in this paper is being applied successfully in
NSW Sports High Schools. The long term goal is to grow the program to
involve other schools and to build a large data base that provides moti-
vational targets for adolescents aspiring to fulfil their sporting potential.
Normalisation by biological age may reduce the incidence of ‘drop-out’
and lost talent. Replication of the program by others may enable merg-
ing of data to provide a larger picture of adolescent sport development
across broad geographical, cultural, and international regions.

Acknowledgments
The authors would like to acknowledge the support and intellectual
input of the Principals of Westfields Sports High School, Roger Davis,
and Endeavour Sports High Schools, James Kozlowski, and their sports

69
 sp or t science

directors Alex Larkin and Dave Davids respectively. We also acknowl-

edge the cooperation and enthusiasm of the staff and coaches of both
schools. These are essential for the success of the program.

References
1. Gabbe BJ, Bailey M, Cook JL, Makdissi M, Scase E, Ames N, et al. The asso-
ciation between hip and groin injuries in the elite junior football years and
injuries sustained during elite senior competition. British Journal of Sports
Medicine. 2010;44(11):799-802.
2. Grimmer KA, Jones D, Williams J. Prevalence of adolescent injury from rec-
reational exercise: an australian perspective. Journal of Adolescent Health.
2000;27(4):266-72.
3. Hodgson L, Gissane C, Gabbett TJ, King DA. For debate: consensus injury
definitions in team sports should focus on encompassing all injuries. Clini-
cal Journal of Sport Medicine. 2007;17(3):188-91.
4. Orr R, Cheng HL. Incidence and characteristics of injuries in elite Austral-
ian junior rugby league players. Journal of science and medicine in sport.
2016;19(3):212-7.
5. King DA, Gabbett TJ, Gissane C, Hodgson L. Epidemiological studies of in-
juries in rugby league: Suggestions for definitions, data collection and re-
porting methods. Journal of Science and Medicine in Sport. 2009;12(1):12-9.
6. Burns J, Crosbie J, Hunt A, Ouvrier R. The effect of pes cavus on foot pain
and plantar pressure. Clinical Biomechanics. 2005;20(9):877-82.
7. Burns J, Keenan AM, Redmond A. Foot type and overuse injury in triath-
letes. J Am Podiatr Med Assoc. 2005;95.
8. Yates B, White S. The incidence and risk factors in the development of me-
dial tibial stress syndrome among naval recruits. The American Journal of
Sports Medicine. 2004;32(3):772-80.
9. Buldt AK, Levinger P, Murley GS, Menz HB, Nester CJ, Landorf KB. Foot pos-
ture is associated with kinematics of the foot during gait: A comparison of
normal, planus and cavus feet. Gait & Posture. 2015;42(1):42-8.
10. Redmond AC, Crosbie J, Ouvrier RA. Development and validation of a novel
rating system for scoring standing foot posture: The Foot Posture Index.
Clinical Biomechanics. 2006;21(1):89-98.
11. McClay I. Report of ‘‘Static and dynamic classification of the foot’’meeting,
Annapolis, Maryland, May 2000. Journal of Orthopedic and Sports Physical
Therapy. 2001;31(3):158.
12. Keenan A-M, Redmond AC, Horton M, Conaghan PG, Tennant A. The Foot
Posture Index: Rasch analysis of a novel, foot-specific outcome measure.
Arch Phys Med Rehabil. 2007;88(1):88-93.
13. Redmond AC, Crane YZ, Menz HB. Normative values for the Foot Posture
Index. J Foot Ankle Res. 2008;1:6.
14. Evans AM, Copper AW, Scharfbillig RW, Scutter SD, Williams MT. Reliability
of the foot posture index and traditional measures of foot position. J Am
Podiatr Med Assoc. 2003;93(3):203-13.
15. Noakes H, Payne C. The reliability of the manual supination resistance test.
J Am Podiatr Med Assoc. 2003;93(3):185-9.

70
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

16. Payne C, Oates M, Noakes H. Static stance response to different types of

foot orthoses. J Am Podiatr Med Assoc. 2003;93(6):492-8.
17. Yates B, White S. The incidence and risk factors in the development of
medial tibial stress syndrome among naval recruits. Am J Sports Med.
2004;32:772 - 80.
18. Morrison S, Ferrari J. Inter-rater reliability of the Foot Posture Index (FPI-
6) in the assessment of the paediatric foot. Journal of Foot and Ankle Re-
search. 2009;2(1):26.
19. Portney L. G.; Watkins, M., P.:“Foundations of Clinical Research “. Prentice-
Hall, New Jersey; 2000.
20. Evans A, Rome K, Peet L. The foot posture index, ankle lunge test, Beighton
scale and the lower limb assessment score in healthy children: a reliability
study. Journal of Foot and Ankle Research. 2012;5(1):1.
21. Yates B, White S. The Incidence and Risk Factors in the Development of
Medial Tibial Stress Syndrome among Naval Recruits. American Journal of
Sports Medicine. 2004;32(3):772-80.
22. Irving DB, Cook JL, Young MA, Menz HB. Obesity and pronated foot type
may increase the risk of chronic plantar heel pain: a matched case-control
study. BMC Musculoskeletal Disorders. 2007;8(1):41.
23. Buldt AK, Levinger P, Murley GS, Menz HB, Nester CJ, Landorf KB. Foot pos-
ture is associated with kinematics of the foot during gait: A comparison of
normal, planus and cavus feet. Gait and Posture. 2015;42(1):42-8.
24. Bojsen-Møller F. Calcaneocuboid joint and stability of the longitudinal arch
of the foot at high and low gear push off. Journal of Anatomy. 1979;129(Pt
1):165-76.
25. Hicks JH. The mechanics of the foot: II. The plantar aponeurosis and the
arch. Journal of Anatomy. 1954;88(Pt 1):25-30.1.
26. Zammit GV, Munteanu SE, Menz HB. Development of a diagnostic rule
for identifying radiographic osteoarthritis in people with first metatar-
sophalangeal joint pain. Osteoarthritis and Cartilage. 2011;19(8):939-45.
27. Clough JG. Functional Hallux Limitus and Lesser-Metatarsal Overload.
Journal of the American Podiatric Medical Association. 2005;95(6):593-601.
28. Dananberg H. Gait style as an etiology to chronic postural pain. Part II. Pos-
tural compensatory process. Journal of the American Podiatric Medical As-
sociation. 1993;83(11):615-24.
29. Menz HB, Morris ME, Lord SR. Foot and ankle risk factors for falls in older
people: a prospective study. J Gerontol A Biol Sci Med Sci. 2006;61.
30. Bevan L, Harradine P, Durrant B. The effect of temporary immobilisation of
the 1st metatarsophalangeal joint upon in-shoe gait analysis parameters: a
preliminary study. Br J Podiatr. 2004;7:54.
31. Dubbeldam R, Nester C, Nene AV, Hermens HJ, Buurke JH. Kinematic cou-
pling relationships exist between non-adjacent segments of the foot and
ankle of healthy subjects. Gait & Posture. 2013;37(2):159-64.
32. Lafuente G, Munuera PV, Dominguez G, Reina M, Lafuente B. Hallux Limitus
and its relationship with the internal rotational pattern of the lower limb.
Journal of the American Podiatric Medical Association. 2011;101(6):467-74.
33. Hopson M, McPoil T, Cornwall M. Motion of the first metatarsophalangeal
joint. Reliability and validity of four measurement techniques. Journal of

71
 sp or t science

the American Podiatric Medical Association. 1995;85(4):198-204.

34. Grady JF, Axe TM, Zager EJ, Sheldon LA. A retrospective analysis of 772 pa-
tients with Hallux Limitus. Journal of the American Podiatric Medical As-
sociation. 2002;92(2):102-8.
35. McPoil TG, Cornwall MW. Relationship Between Three Static Angles of the
Rearfoot and the Pattern of Rearfoot Motion During Walking. Journal of Or-
thopaedic & Sports Physical Therapy. 1996;23(6):370-5.
36. Hughes LY. Biomechanical analysis of the foot and ankle for predisposition
to developing stress fractures. Journal of Orthopaedic & Sports Physical
Therapy. 1985;7(3):96-101.
37. Warren BL, Jones CJ. Predicting plantar fasciitis in runners. Med Sci Sports
Exerc. 1987;19(1):71-3.
38. Baumbach SF, Braunstein M, Seeliger F, Borgmann L, Böcker W, Polzer H.
Ankle dorsiflexion: what is normal? Development of a decision pathway for
diagnosing impaired ankle dorsiflexion and M. gastrocnemius tightness.
Archives of Orthopaedic and Trauma Surgery. 2016;136(9):1203-11.
39. Bennell K, Talbot R, Wajsweiner H, Techovanich W, Kelly DH, Hall AJ. Intra-
rater and Inter-rater reliability of a weight-bearing lunge measure of ankle
dorsiflexion. Aust J Physiother. 1998;44.
40. Boyle KL, Witt P, Riegger-Krugh C. Intrarater and interrater reliability of
the Beighton and Horan joint mobility index. Journal of Athletic Training.
2003;38(4):281-5.
41. van der Geissen LJ, Liekens D, Rutgers KJ, Hartman A, Mulder PG, Oranje
AP. Validation of Beighton score and prevalence of connective tissue signs
in 773 Dutch children. J Rheumatol. 2001;28.
42. Rajeb A, Whiteley R. Beighton scoring of joint laxity and injury inci-
dence in athletic adolescent males. British Journal of Sports Medicine.
2014;48(7):653-.
43. Quatman CE, Ford KR, Myer GD, Paterno MV, Hewett TE. The effects of gen-
der and pubertal status on generalized joint laxity in young athletes. Jour-
nal of Science and Medicine in Sport. 2008;11(3):257-63.
44. Pacey V, Nicholson LL, Adams RD, Munn J, Munns CF. Generalized joint hy-
permobility and risk of lower limb joint injury during sport. The American
Journal of Sports Medicine. 2010;38(7):1487-97.
45. Ferrari J, Parslow C, Lim E, Hayward A. Pediatric rheumatology joint hyper-
mobility: The use of a new assessment tool to measure lower limb hyper-
mobility. Clin Exp Rheumatol. 2005;23(413):20.
46. Tompsett C, Burkett B, McKean M. Comparing performances of fundamen-
tal movement skills and basic human movements: a pilot study. Journal of
Fitness Research. 2015;4.
47. Giles K, Penfold L, Giorgi A. A guide to developing physical qualities in
young athletes. Queensland, Austrailia: Movement Dynamics Pty Ltd. 2005.
48. Tompsett C, Burkett B, McKean M. Development of physical literacy and
movement competency: A literature review. Journal of Fitness Research.
2014;3(2):53-74.
49. Myer GD, Faigenbaum AD, Ford KR, Best TM, Bergeron MF, Hewett TE. When
to initiate integrative neuromuscular training to reduce sports-related in-
juries in youth? Current sports medicine reports. 2011;10(3):155-66.

72
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

50. Myer GD, Ford KR, Palumbo JP, Hewett TE. Neuromuscular training im-
proves performance and lower-extremity biomechanics in female athletes.
Journal of Strength and conditioning Research. 2005;19(1):51.
51. Leetun DT, Ireland ML, Willson JD, Ballantyne BT, Davis IM. Core stability
measures as risk factors for lower extremity injury in athletes. Medicine &
Science in Sports & Exercise. 2004;36(6):926-34.
52. Nadler SF, Malanga GA, Feinberg JH, Prybicien M, Stitik TP, DePrince M. Re-
lationship between hip muscle imbalance and occurrence of low back pain
in collegiate athletes: a prospective study. American journal of physical
medicine & rehabilitation. 2001;80(8):572-7.
53. Chorba RS, Chorba DJ, Bouillon LE, Overmyer CA, Landis JA. Use of a func-
tional movement screening tool to determine injury risk in female colle-
giate athletes. North American Journal of Sports Physical Therapy : NA-
JSPT. 2010;5(2):47-54.
54. Tompsett C, Cobley S. Basic Human Movement Assessment: User Guide-
lines. Sydney, Australia: The University of Sydney; 2016.
55. Nuzzo JL, Anning JH, Scharfenberg JM. The Reliability of Three Devices
Used for Measuring Vertical Jump Height. The Journal of Strength & Condi-
tioning Research. 2011;25(9):2580-90.
56. Bilsborough JC, Greenway KG, Opar DA, Livingstone SG, Cordy JT, Bird SR,
et al. Comparison of anthropometry, upper-body strength, and lower-body
power characteristics in different levels of australian football players. The
Journal of Strength & Conditioning Research. 2015;29(3):826-34.
57. Woods CTE, Raynor AJ, Bruce L, McDonald Z, Collier N. Predicting playing
status in junior Australian Football using physical and anthropometric pa-
rameters. Journal of Science and Medicine in Sport. 2015;18(2):225-9.
58. Gabbett TJ, Jenkins DG, Abernethy B. Relative importance of physiological,
anthropometric, and skill qualities to team selection in professional rugby
league. Journal of Sports Sciences. 2011;29(13):1453-61.
59. Till K, Cobley S, O’Hara J, Chapman C, Cooke C. A longitudinal evaluation
of anthropometric and fitness characteristics in junior rugby league play-
ers considering playing position and selection level. Journal of Science and
Medicine in Sport. 2013;16(5):438-43.
60. Enoksen E, Tønnessen E, Shalfawi S. Validity and reliability of the New-
test Powertimer 300-series® testing system. Journal of Sports Sciences.
2009;27(1):77-84.
61. Foran B. High-performance sports conditioning: Human Kinetics; 2001.
62. Garrido N, Marinho DA, Barbosa TM, Costa AM, Silva AJ, Pérez-Turpin JA,
et al. Relationships between dry land strength, power variables and short
sprint performance in young competitive swimmers. Journal of human
sport and exercise. 2010(II).
63. McGee KJ, Burkett LN. The national football league combine: a reliable
predictor of draft status? The Journal of Strength & Conditioning Research.
2003;17(1):6-11.
64. Baker D, Nance S. The relation between running speed and measures of
strength and power in professional rugby league players. The Journal of
Strength & Conditioning Research. 1999;13(3):230-5.
65. Barr MJ, Sheppard JM, Gabbett TJ, Newton RU. Long-term training-induced

73
 sp or t science

changes in sprinting speed and sprint momentum in elite rugby union play-
ers. The Journal of Strength & Conditioning Research. 2014;28(10):2724-31.
66. Hoffman JR, Fry AC, Howard R, Maresh CM, Kraemer WJ. Strength, speed
and endurance changes during the course of a division I basketball season.
J Appl Sport Sci Res. 1991;5(3):144-9.
67. Young WB, Newton RU, Doyle TLA, Chapman D, Cormack S, Stewart C, et
al. Physiological and anthropometric characteristics of starters and non-
starters and playing positions in elite Australian Rules football: a case
study. Journal of Science and Medicine in Sport. 2005;8(3):333-45.
68. Gabbett TJ. Influence of physiological characteristics on selection in a
semi-professional first grade rugby league team: a case study. Journal of
Sports Sciences. 2002;20(5):399-405.
69. Sheppard JM, Young WB. Agility literature review: Classifications, training
and testing. Journal of Sports Sciences. 2006;24(9):919-32.
70. Scanlan A, Humphries B, Tucker PS, Dalbo V. The influence of physical and
cognitive factors on reactive agility performance in men basketball play-
ers. Journal of Sports Sciences. 2014;32(4):367-74.
71. Spiteri T, Nimphius S, Hart NH, Specos C, Sheppard JM, Newton RU. Contri-
bution of strength characteristics to change of direction and agility perfor-
mance in female basketball athletes. The Journal of Strength & Condition-
ing Research. 2014;28(9):2415-23.
72. Stewart PF, Turner AN, Miller SC. Reliability, factorial validity, and interre-
lationships of five commonly used change of direction speed tests. Scandi-
navian Journal of Medicine & Science in Sports. 2014;24(3):500-6.
73. Léger LA, Mercier D, Gadoury C, Lambert J. The multistage 20 metre shuttle
run test for aerobic fitness. Journal of Sports Sciences. 1988;6(2):93-101.
74. Flouris AD, Metsios GS, Koutedakis Y. Enhancing the efficacy of the 20 m mul-
tistage shuttle run test. British Journal of Sports Medicine. 2005;39(3):166-
70.
75. Burgess D, Naughton G, Hopkins W. Draft-camp predictors of subsequent
career success in the Australian Football League. Journal of Science and
Medicine in Sport. 2012;15(6):561-7.
76. Currell K, Jeukendrup AE. Validity, Reliability and Sensitivity of Measures of
Sporting Performance. Sports Medicine. 2008;38(4):297-316.
77. Davis JA, Brewer J, Atkin D. Pre season physiological characteristics of
English first and second division soccer players. Journal of Sports Scienc-
es. 1992;10(6):541-7.
78. Castagna C, Manzi V, Impellizzeri F, Weston M, Barbero Alvarez JC. Re-
lationship between endurance field tests and match performance in
young soccer players. The Journal of Strength & Conditioning Research.
2010;24(12):3227-33.
79. Gabbett TJ, Seibold AJ. Relationship between tests of physical qualities,
team selection, and physical match performance in semiprofessional
rugby league players. The Journal of Strength & Conditioning Research.
2013;27(12):3259-65.
80. Higham DG, Pyne DB, Anson JM, Eddy A. Physiological, anthropometric,
and performance characteristics of rugby sevens players. International
Journal of Sports Physiology and Performance. 2013;8(1):19-27.

74
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

81. Deprez D, Coutts AJ, Lenoir M, Fransen J, Pion J, Philippaerts R, et al. Reli-
ability and validity of the Yo-Yo intermittent recovery test level 1 in young
soccer players. Journal of Sports Sciences. 2014;32(10):903-10.
82. Wells KF, Dillon EK. The sit and reach—a test of back and leg flexibility. Re-
search Quarterly American Association for Health, Physical Education and
Recreation. 1952;23(1):115-8.
83. Hui SS, Yuen PY. Validity of the modified back-saver sit-and-reach test: a
comparison with other protocols. Med Sci Sports Exerc. 2000;32(9):1655-9.
84. Kendall FP. Muscles: testing and function with posture and pain. 5th ed.
Baltimore, MD: Lippincott Williams & Wilkins; 2005.
85. Erkula G, Demirkan F, Alper Kılıç B, Kıter E. Hamstring shortening in healthy
adults. Journal of Back & Musculoskeletal Rehabilitation. 2002;16(2/3):77.
86. Jones CJ, Rikli RE, Max J, Noffal G. The reliability and validity of a chair
sit-and-reach test as a measure of hamstring flexibility in older adults. Re-
search Quarterly for Exercise and Sport. 1998;69(4):338-43.
87. Ayala F, Sainz de Baranda P, De Ste Croix M, Santonja F. Absolute reliability
of five clinical tests for assessing hamstring flexibility in professional futsal
players. Journal of Science and Medicine in Sport. 2012;15(2):142-7.
88. Hiller CE, Refshauge KM, Herbert RD, Kilbreath SL. Balance and recovery
from a perturbation are impaired in people with functional ankle instabil-
ity. Clinical Journal of Sport Medicine. 2007;17(4):269-75.
89. Pourkazemi F, Hiller C, Raymond J, Black D, Nightingale E, Refshauge K.
Using balance tests to discriminate between participants with a recent in-
dex lateral ankle sprain and healthy control participants: a cross-sectional
study. Journal of Athletic Training. 2016;51(3):213-22.
90. Sherar LB, Mirwald RL, Baxter-Jones ADG, Thomis M. Prediction of adult
height using maturity-based cumulative height velocity curves. The Jour-
nal of Pediatrics. 2005;147(4):508-14.
91. Carter JL, Heath BH. Somatotyping: development and applications: Cam-
bridge University Press; 1990.
92. Perroni F, Vetrano M, Camolese G, Guidetti L, Baldari C. Anthropomet-
ric and somatotype characteristics of young soccer players: Differences
among categories, subcategories, and playing position. The Journal of
Strength & Conditioning Research. 2015;29(8):2097-104.
93. Arellano R, Pardillo S, Gavilán A, editors. Underwater undulatory swim-
ming: Kinematic characteristics, vortex generation and application during
the start, turn and swimming strokes. Proceedings of the XXth Internation-
al Symposium on Biomechanics in Sports, Universidad de Granada; 2002.
94. Georgopoulos NA, Markou KB, Theodoropoulou A, Benardot D, Leglise M,
Vagenakis AG. Growth retardation in artistic compared with rhythmic elite
female gymnasts. The Journal of Clinical Endocrinology & Metabolism.
2002;87(7):3169-73.
95. Malina RM, Rogol AD, Cumming SP, Coelho e Silva MJ, Figueiredo AJ. Bio-
logical maturation of youth athletes: assessment and implications. British
Journal of Sports Medicine. 2015;49(13):852-9.
96. Gualdi-Russo E, Zaccagni L. Somatotype, role and performance in elite
volleyball players. Journal of Sports Medicine and Physical Fitness.
2001;41(2):256.

75
 sp or t science

97. Stewart A, Marfell-Jones M, International Society for Advancement of K.

International standards for anthropometric assessment. Lower Hutt, New
Zealand: International Society for the Advancement of Kinanthropometry;
2011.
98. Perini TA, Oliveira GLd, Ornellas JdS, Oliveira FPd. Technical error of meas-
urement in anthropometry. Revista Brasileira de Medicina do Esporte.
2005;11(1):81-5.
99. Hodgdon JA, Beckett MB. Prediction of percent body fat for US Navy men
from body circumferences and height. NAVAL HEALTH RESEARCH CENTER
SAN DIEGO CA; 1984.
100. Hodgdon JA, Beckett MB. Prediction of percent body fat for US Navy wom-
en from body circumferences and height1984. 30 p.
101. Milsom J, Naughton R, O’Boyle A, Iqbal Z, Morgans R, Drust B, et al. Body
composition assessment of English Premier League soccer players: a com-
parative DXA analysis of first team, U21 and U18 squads. Journal of Sports
Sciences. 2015;33(17):1799-806.
102. Ackland TR, Lohman TG, Sundgot-Borgen J, Maughan RJ, Meyer NL, Stew-
art AD, et al. Current status of body composition assessment in sport.
Sports Medicine. 2012;42(3):227-49.
103. Duthie G, Pyne D, Hooper S. Applied physiology and game analysis of rugby
union. Sports Medicine. 2003;33(13):973-91.
104. Bahr R, Krosshaug T. Understanding injury mechanisms: a key compo-
nent of preventing injuries in sport. British Journal of Sports Medicine.
2005;39(6):324-9.
105. Dewar DL, Lubans DR, Plotnikoff RC, Morgan PJ. Development and evalu-
ation of social cognitive measures related to adolescent dietary behav-
iors. International Journal of Behavioral Nutrition and Physical Activity.
2012;9(1):36.
106. Lubans DR, Lonsdale C, Plotnikoff RC, Smith J, Dally K, Morgan PJ. Develop-
ment and evaluation of the motivation to limit screen-time questionnaire
(MLSQ) for adolescents. Preventive Medicine. 2013;57(5):561-6.
107. Niven P, Scully M, Morley B, Crawford D, Baur LA, Wakefield M. Socio-
economic disparities in Australian adolescents’ eating behaviours. Public
Health Nutrition. 2014;17(12):2753-8.
108. Scott JJ, Morgan PJ, Plotnikoff RC, Lubans DR. Reliability and validity of a
single-item physical activity measure for adolescents. Journal of Paediat-
rics and Child Health. 2015;51(8):787-93.
109. Gwynn JD, Flood VM, D’Este CA, Attia JR, Turner N, Cochrane J, et al. The re-
liability and validity of a short FFQ among Australian Aboriginal and Torres
Strait Islander and non-Indigenous rural children. Public Health Nutrition.
2011;14(3):388-401.
110. Larsen CH, Alfermann D. Understanding dropout in the athlete develop-
ment process. In: Baker J, Cobley S, Schorer J, Wattie N, editors. Routledge
Handbook of Talent Development in Sport2017. p. 325-35.
111. Knight CJ. Family influences of talent development in sport. In: Baker J,
Cobley S, Schorer J, Wattie N, editors. Routledge Handbook of Talent De-
velopment in Sport2017. p. 181-91.

76
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Research into optimising performance from a

multidisciplinary team: the case of the IGOID
research group
Leonor Gallardo1, Enrique Colino1, José Luis Felipe2,1, Jorge
García-Unanue2,1, Manuel León1, Jorge López-Fernández1, Javier
Sánchez-Sánchez2,1 and Esther Ubago-Guisado1

1. Introduction
The IGOID Group (Grupo IGOID: Investigación en la Gestión de Organi-
zaciones e Instalaciones Deportivas) is a research group specialising in
research in management of organisations and sports facilities. However,
the world of sport has evolved with great speed and with a marked in-
terdisciplinarity. Therefore, the analysis of performance in the different
manifestations of sport has become a subject of relevant importance
in the world of research in the sports sciences. This way, in the IGOID
Group (www.igoid.es) we have been working for years in this sector. Our
research is related with sports facilities and the quality of life that we
want to give to our society. Therefore, we are going to present our re-
search, which is divided in different areas of study: sports surfaces and
their influence of performance; new technologies in the control and pre-
vention of injuries, growth and bone development; and optimisation of
management in sports companies. The rest of the chapters are divided
in one part for each area of study.

2. Sports surfaces and their influence on performance

Professional football is commonly played on a natural grass surface
as it is considered the most suitable surface for this sport. However,
semi-professional and amateur football are increasingly opting more
for artificial turf systems of third generation due to the social, environ-
mental and economic return of these surfaces.1,2 The fast growth in the
use of artificial turf pitches in Spain, as well as the diverse use of materi-
als for their construction, led the IGOID Group to raise two fundamental
questions about these surfaces: i) how adequate are these surfaces for
football practice? ii) what do the main agents involved in the game think
about them?

1. Grupo IGOID. Universidad de Castilla-La Mancha

2. Universidad Europea de Madrid. Facultad Ciencias de la Actividad Física y el
Deporte

77
 sp or t science

Regarding the first question, the quality reached by third-generation

systems of artificial turf made international organisations, like the Eu-
ropean Committee for Standardization or the Fédération Internationale
de Football Association, accept them for playing football.1 However, as
any manufactured product it requires fulfilling a minimum standard to
be used for playing a sport on it. For that reason, both these entities de-
veloped standards of certification for artificial surfaces which are con-
stantly updated and reviewed.3

After studying the mechanical properties of 20 artificial turf pitches

across Spain, Burillo et al.4 concluded that the mechanical properties
of artificial turf systems of third generation are influenced by different
parameters such as the type of fibre, type of rubber infill, elastic base,
years of use and maintenance system. Therefore, when implementing
a new surface of artificial turf, all these parameters must be considered.
Likewise, this study also reveals that despite the existence of a stand-
ardised norm for these surfaces, they were not fulfilled by the artificial
turf fields of Spain, increasing the risk of suffering an injury. These con-
clusions were also supported by Sánchez-Sánchez, et al.5 after analysing
four different systems of artificial turf. Despite this work only analysing
four surfaces, they found that the choice of materials of support can
increase the lifetime of the artificial turf system considerably. Further-
more, a bad election of these materials together with the lack of main-
tenance can make the surface unusable in a few years. Therefore, when
implementing a new field of artificial turf, the manager must consider
the structural components of the system and the maintenance that is
going to be applied and the number of uses per week. Finally, to provide
the best experience possible and avoid unnecessary risks, the manager
should check that the new field of artificial turf has been built according
to the standardised norms and that the system maintains these proper-
ties over time.

Concerning the second question, several authors have reported

greater footballer satisfaction with natural surfaces compared to syn-
thetic ones,6,7 even though synthetic systems seem to provide similar
performance to natural grass fields.8 The work from IGOID Group reveals
that the satisfaction that artificial turf is positively valued by players,
coaches and referees.1,2 However, satisfaction level is related with previ-
ous experience, as footballers who used to play on dust pitches reported

78
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

higher satisfaction than those who used to do it on grass surfaces. Within

this topic, Andersson et al.6. identified that the style of play is modified
by the type of surface and by the perception of the footballers. For in-
stance, players tend to avoid tackles and increase the number of short
passes when playing on a field of artificial turf. Thus, this can explain
the difference of satisfaction level according to the previous experience.

With the development of tracking technologies like Global Position-

ing System (GPS), it has been possible to study how the game surface
influences the performance of footballers.7-9 Thus, comparative studies
between artificial and turf pitches indicate similar players’ responses
(physical and physiological) regardless of the surface used.7,8,10 However,
these studies only focused on standardised tests, so tactical and tech-
nical actions were not studied. The main conclusion of these studies is
that the new third-generation pitches do not cause greater fatigue, do
not delay recovery or reduce sprint performance. On the other hand,
longitudinal studies about injury rates on artificial turf and natural grass
prove that playing football on the first surface is not riskier than on natu-
ral grass.11 Nonetheless, these authors did not analyse the mechanical
properties of the fields used in their studies, so they could not relate
these results with the mechanical properties of the newest fields of ar-
tificial turf.

To increase the understanding about the role of mechanical prop-

erties on players’ responses when playing on both surfaces, the IGOID
Group has developed two studies (not published by August 2017). Both
studies used the same two surfaces whose mechanical properties were
analysed (Table 1)

Artificial Natural p value

FR (%) 43.59±5.09* 49.69±6.33 p = 0.07

StV (mm) 6.77±0.60 5.29±0.83 p < 0.01

ER (%) 53.84±2.23 39.22±6.76 p < 0.01

Note: * Non-compliance with the specifications of the standard EN 15330-

1:2014. FR = force reduction; StV = standard vertical deformation; ER = en-
ergy restitution

Table 1. Mechanical Properties of Both Selected Surfaces

79
 sp or t science

The first work analyses the accumulated fatigue of sixteen players

after performing the first three bouts of a soccer simulation protocol
(three blocks of 16 min with 3 min recovery between blocks,10). The re-
sults showed similar muscle response on both surfaces right after finish-
ing the test, as players got similar performance in a countermovement
jump. Likewise, through a tensiomyography it was analyse the response
of both biceps femoris and rectus femoris to an electric stimulus not
finding differences according to the surface. Therefore, despite the exist-
ing differences of mechanical properties of both surfaces, they were not
high enough to cause different impact on muscle fatigue.

The second work also used the first three bouts of a soccer simulation
protocol, but in this case, we analysed the physiological responses and
performance of players on the test. Each bout of the soccer simulation
protocol is composed of eight cycles and one repeated sprint (RS: 6 ×
15 m sprints departing every 18 s) block between the fourth and fifth
cycles structured as follows: i) 3 × 20 m at a walking pace of 1.43 m/s;
ii) 1 × sprint-agility run (S-AR) at maximal intensity (20 s for sprint and
recovery); iii) 3 × 20 m at a running speed of 2.5 m/s; iv) 3 × 20 m at a
running speed of 4.0 m/s. After analysing the performance of players on
each sprint-agility run and the repeated sprint of the three bouts, play-
ers presented lower time sprint in some part of agility test on artificial
turf. Therefore, the higher energy restitution of the artificial turf seems
to cause faster performance in non-linear sprints. Nevertheless, these
findings suggest that players should not present lower performance in
linear and non-linear sprints when playing football on artificial turf than
on natural grass.

While these results strengthen the previous findings,7,8,10 care must be

taken when evaluating them, because as mentioned previously, the me-
chanical properties of an artificial turf pitch can differ greatly between
systems.1,5,12,13 In fact, through different investigations,14-17 the IGOID
Group has proved a significant variation in footballers’ response in func-
tion of the type of artificial turf used in both real-life football situations
and in standardised match tests. Therefore, a greater physiological value
of peak heart rate has been found on artificial turf systems with greater
damping capacity, while harder surfaces are related with higher depend-
ence of creatine phosphate.9,17 Furthermore, the main differences found
between artificial systems is in actions like change of direction and

80
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

turns, where factors like rotational traction and energy restitution of the
ground seem to have a great influence.5,17

Conclusions
In conclusion, the structural components of the system together with
other factors such as number of hours of use per week or maintenance
plan as they play a key role on the mechanical properties of artificial turf
systems. Moreover, even though natural grass is still perceived as the
most suitable surface for playing football, the findings of previous stud-
ies suggest that artificial turf systems can provide similar performance
to natural grass fields, not increasing the risk of suffering an injury.
However, it is required that artificial turf systems fulfil the international
standards as those that do not meet them could negatively affect play-
ers’ performance or security.

3. New technologies in the control and prevention of injuries

Our research is related with sports facilities and the quality of life that
we want to give to our society. Therefore, we can focus on topics of re-
search rather than in the equipment. These materials can be then men-
tioned as recent technology that supports the advanced research made
by the group (e.g. GPS technology, tracking technology and positioning
of sports facilities, thermography, tensiomyography to different studies
on surfaces and bone mass, etc.). Thereby, the main research topic is the
importance of guaranteeing the correct use of artificial turf, with all its
qualities in performance and its optimisation by sports companies.

Global Positioning System (GPS)

GPS devices register the activity and physiological profiles of play-
ers with a sampling frequency of 10 Hz. The devices are located at the
top part of the player’s back, inside a specific chest guard provided by
the manufacturer. Similarly, heart rate bands are located at chest level
through elastic bands. Afterwards, the data are downloaded to a person-
al computer and analysed using software provided by the manufacturer.
The variables obtained by these devices provide information on physical
and physiological performance of athletes.

Movement patterns — external load

The GPS devices attached to the players provided information about
the total distance covered during the game (DT), the maximum speed

81
 sp or t science

(Vmax) and average speed (Vmean) reached during matches and/or training
sessions, as well as the distance covered in each one of the six locomotor
categories with speed ranges adapted from previous studies:16,17 stand-
ing (0–2 km/h), walking (2–7 km/h), easy running (7–13 km/h), fast run-
ning (13–18 km/h), high-speed running (18–21 km/h) and sprinting (>21
km/h). The actions above 13 km/h (fast running, high-speed running and
sprinting) are defined as high-intensity running. The GPS software also
provided information about the number, average distance and maxi-
mum average speed of the sprints (Figure 1). These actions of high inten-
sity have been identified as the most determinant in the performance of
team sports.

In the same way, the GPS devices register the maximum acceleration
peaks, as well as the number of accelerations of the players in different
ranges of intensity: 1.5–2 m/s2; 2–2.5 m/s2; 2.5–2.75 m/s2; >2.75 m/s2. The
data of maximum peak and number of impacts during the game are ob-
tained using the triaxial accelerometer of 100 Hz incorporated to each
one of the GPS devices and classified into six categories based on the
G forces reached (1 G = 9.81 m/s2): light impact (5–6 G), light/moderate
impact (6–6.5 G), moderate/heavy impact (6.5–7 G), heavy impact (7–8
G), very heavy impact (8–10 G), and severe impact (>10 G). The control
of the number of accelerations and impacts on the athlete can allow
the development of specific training tasks adapted to real situations of
competition, reducing athlete injury risk. Like the previous variable, the
load of the player can be obtained from the combination of the accelera-
tions detected in the three planes of body motion (vertical, horizontal
and anteroposterior), using the software of the GPS. The load data are
presented in arbitrary units. The GPS detection of an imbalance of more
than 5% can alert to a decompensation in the motor behaviour of the
athlete derived from a potentially dangerous situation for the athlete’s
health. Finally, GPS software includes the ratio work:rest, which results
from the quotient between the distance covered at speeds above 4 km/h
(work phase) and the distance covered by the player at speeds below 4
km/h (rest phase).

Heart rate parameters — internal load

Internal load is an essential parameter for training control. For this, it
is necessary to know the individual maximum heart rate (HRmax) of each
athlete. Numerous specific tests are used in each sport to estimate this
variable. For example, in football, the HRmax value can be obtained in the

82
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

intermittent recovery yo-yo test to establish different categories of in-

tensity during a match or training session: <70% HRmax, 70–80% HRmax,
80–90% HRmax, 90–95% HRmax and >95% HRmax. The GPS software synchro-
nises with the HR bands and registers the time used by players in each
one of the zones described above. The data are presented as relative
percentages in relation to the total time played. However, the peak and
average values of heart rate reached during the game and 15 min inter-
vals are obtained in absolute (bpm) and relative (% HRmax) terms.

The monitoring of these variables will reveal the level of fatigue of the
athlete, undertaking interventionist protocols to prevent injuries during
practice (Figure 1). In this sense, the heterogeneity between the acute
and chronic load experienced by the athlete has been identified as the
main predictor of athlete injury incidence.18 The use of these devices will
allow the control of load, assuring homogeneity between these param-
eters and reducing injury risk in athletes.

Figure 1. Profile of a footballer’s activity during a real game situation

Tensiomyography (TMG)
Tensiomyography is a non-invasive evaluation technique that allows
the evaluation of contractile capacities of an athlete’s muscles. This
control tool allows the identification of muscle imbalances in athletes
and the discovery of neuromuscular fatigue after an effort.19 This device
measures: maximal displacement (Dm) given by the radial movement of
muscle belly expressed in mm and depends on the muscle tone or stiff-
ness; contraction time (Tc), the time between 10 and 90% of Dm; sustain
time (Ts), the time in which the muscle response remains >50% of Dm;
delay time (Td), also known as reaction or activation time, is the time
between the initiation and 10% of Dm; and half-relaxation time (Tr), the
time in which the muscle response decreases from 90 to 50% of Dm mus-
cle (Figure 2).

83
 sp or t science

Figure 2. Tensiomyography (TMG) parameters definition.19 Dm=maximal dis-

placement; Tc=contraction time, Ts=sustain time, Td=delay time and
Tr=half-relaxation time.

A small electric stimulation is produced on the required muscle. This

stimulus is measured by placing a digital transducer perpendicular to
the muscle belly. The stimulation of the selected muscle is made us-
ing two self-adhesive electrodes placed equidistant to the point where
the measurement will be made. The proximal electrode corresponds
to the anode and the distal to the cathode. The stimulus is produced
by a system electro-stimulator of 1 ms duration. The amplitude range
of the electrical stimulus can be from 0 to 110 mA. The measurement
protocol varies according to the muscle group evaluated. The compari-
son of variables in the same athlete will allow detecting neuromuscular
asymmetries, establishing the level of muscle recovery after a training
stimulus and discovering the fatigue accumulated in the muscle groups
evaluated.20 This information will guarantee an exhaustive control of the
physical load of the athlete and the risk of muscle injury during practice.
Finally, a periodic evaluation of these variables will clarify when an ath-
lete can return to competition after an injury.

Thermography
Thermography has become a powerful research instrument in many
applications at mechanical, electrical, military, construction and medi-
cal levels. Due to its non-invasive character, infrared thermography (IRT)
can be defined as data analysis by using non-contact thermal imaging
instruments. Thermographic cameras detect radiation within the elec-
tromagnetic spectrum range and produce images of this radiation called
thermograms. This method provides real-time images at a distance, with
measurements of the surface temperature, in this case, skin tempera-

84
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

ture. It is a highly sensitive diagnostic method, fast, totally safe, harm-

less, highly reproducible and does not involve radiation emission.21 This
technique allows the visualisation of the surface temperature with sen-
sitivity above 0.025°C and an accuracy of 1%, non-invasively and without
physical contact with the subject.21 These characteristics allow scientists
to obtain a general thermal profile of the subject and if carried out rou-
tinely, lead to a real-time monitoring of skin temperature, which allows
obtaining information on the complex thermoregulation system of the
human body.

Currently, there are some articles21,22 in the literature that indicate the
possibility of using thermography for: evaluating muscle recovery; pro-
gramming training sessions; modifying workouts; estimating maximum
oxygen consumption VO2max; identifying edemas, varicose veins or is-
chemia; and identifying muscle injuries. Since injuries trigger inflamma-
tory processes and inflammation causes heat by increasing basal me-
tabolism, the level of inflammation can be assessed using temperature
gradients. Since thermography allows for the detection of small tem-
perature variations, thermographic images show, prematurely, the start
of inflammatory processes that still have not shown symptoms (pain,
edema or paraesthesia), acting as a preventive tool.

Thermal symmetry of the human body has been defined as the ‘de-
gree of similarity’ of two regions of interest (ROI) reflected along the
main longitudinal axis of the human body, which are equal in form and
size and were taken at the same angle.23 Pathological conditions, like
musculoskeletal disorders and traumatic injuries, affect the thermal dis-
tribution of the surface temperature, producing changes in its natural
symmetrical pattern. Thermal symmetry values have been considered
important for the evaluation of athletes after an injury.24 Thermal asym-
metries do not usually exceed values over 0.25°C. Different studies have
shown that greater asymmetries (above 0.60°C) may correspond to mus-
culoskeletal injuries (Figure 3). For this reason, infrared thermography
has been consolidated as a valuable tool for clinically evaluating the pro-
gress and treatment of sports injuries and musculoskeletal disorders.

85
 sp or t science

Figure 3. Thermal asymmetry detected in a knee injury in a futsal player (IGOID

archive)

Conclusions
In conclusion, the use of new technologies applied to sport allows
a detailed control of accumulated load of an athlete during the season.
Maybe the most important one is the GPS technology about the tactical
and performance applications. The interpretation of this information fa-
cilitates the improvement of physical planning carried out by teams and/
or athletes, as well as for early injury diagnosis. Therefore, their incorpo-
ration in sport nowadays has become indispensable. The multifactorial
character of these protocols, together with additional technology asso-
ciated with heart rate variability, balance and plantar pressure among
others, as well as controlling contextual variables like the playing sur-
face,17,25 increases the validity and reliability of interventional method-
ologies used by this research group in relation with football performance
and injury prevention.

4. Growth and bone development

In the last few years, there has been an alarming increase in over-
weight individuals and obesity, especially among young people, due to,
among other factors, low levels of physical activity. Obesity in childhood
is closely related to adult obesity, because these children have twice the
risk of developing obesity in later life than those who are not obese.26
Studies such as Boreham et al.27 show that physical activity during child-
hood prevents obesity in later life. For these reasons, the prevention
of obesity in childhood is an international priority given the impact it
has on chronic diseases, general health, development, and well-being.
It is known that physical exercise and sport in general regulate fat and
muscle mass of children. On the other hand, the type of sport practised
influences the body composition development.28 Each sport due to its
characteristics and requirements is associated with different player pro-

86
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

files. Within this framework, the IGOID Group developed an experimen-

tal study to analyse the differences in fat mass and lean mass of girls
playing different sports, and which the main conclusion was: ‘Girls who
practise regular high-impact sports (football, basketball and handball)
and low-impact (swimming) have lower fat mass and higher lean mass
compared to the inactive controls at prepubescent and puberty.

Likewise, the interest in bone health in childhood has grown mean-

ingfully because of the increase of osteoporosis cases in adults. Up to
60% of osteoporosis cases in adult life are related with acquired bone
mineral content during adolescence.29 Osteoporosis is considered a pub-
lic health problem due to the high number of people who suffer from
it and its economic repercussions that are generated during treatment
and rehabilitation.30 An increment in the level of physical activity in chil-
dren would result in higher bone mass accrual and a decrease in the risk
of suffering bone fractures during adulthood.31 Research carried out by
Vicente-Rodriguez et al.32 showed that boys engaged in a high-impact
sport like football increased bone mass values compared to inactive
boys. Nevertheless, while practising low-impact sports, as for example
swimming33 and cycling34, the bones do not get so many stimuli, so the
bone density values are lower.

Therefore, due to lack of research with girls, we carried out a second

study35 with the objective of comparing the differences in bone mass in
girls according to the type of sport practised. The main conclusion we
obtained from this research was: ‘Girls who practise high-impact sports
(football, basketball and handball) have higher bone mass compared to
girls who practise low-impact sports (swimming) and controls at puber-
ty.’ On the other hand, muscle mass is known as an important factor for
predicting bone mass increment during puberty.36 Although the role of
muscle mass in the development of bone mass is evident, the associa-
tion between fat mass and bone mass generates certain debate. Some
studies in girls show that fat mass is positively related with bone mass.37-
39
However, other studies support the idea that bone mass is negatively
related with bone mineral density in boys.40

In this sense, through a third and fourth study using a dual-energy

X-ray absorptiometry scan (DXA),41,42 we analysed the influence of fat
mass, lean mass, maximum oxygen consumption, weekly training hours
and years of sports practice on bone mass in girls; and we examined

87
 sp or t science

the association between muscular and cardiorespiratory fitness with

bone mineral density and hip geometry in boys. The main conclusions
obtained were: ‘1) During puberty, bone mass in girls is related to lean
mass, cardiorespiratory fitness and weekly hours of sports practice; 2)
Vigorous physical activity does not seem to explain the association be-
tween fitness (cardiorespiratory and muscular) and bone parameters
(bone mineral density and hip geometry), while lean mass plays a key
role in this association in young males.’

Finally, we must also take extrinsic factors into account like the play-
ing surface. The affinity between the player and the playing surface is
important for player’s performance and health. Most researches have
focused on studying the influence on injury rate43 and performance17. Up
to date, only one transversal study in children has been done by Plaza-
Carmona et al.44 that has addressed the issue on outdoor surfaces. In
that study, no significant differences were found in bone mass in the
limb, pelvis and hip regions between practising football on dust pitches
and artificial turf. Thus, in a fifth study41 we compared the differences in
bone mass in girls playing on different surfaces, obtaining as a main con-
clusion: A hard playing surface, with less vertical deformation and force
reduction and with higher energy return, is associated with higher levels
of bone mass in girls, regardless of the sport they practise.

Conclusions
The conclusions of our studies are that regular practice of both high-
impact and low-impact sports is associated with a lower fat mass and
higher lean mass in girls during puberty. In addition, girls who practise
high-impact sports (football, basketball and handball) have higher bone
mass compared to those who practise low-impact sports (swimming).
During puberty, bone mass is related to lean mass, cardiorespiratory fit-
ness and weekly hours of sports practice. Also, vigorous physical activity
does not seem to explain the association between fitness (cardiorespira-
tory and muscular) and bone parameters (bone mineral density and hip
geometry), while lean mass plays a key role in this association in young
males. Finally, a hard-playing surface, with less vertical deformation and
force reduction and with higher energy return, is associated with higher
levels of bone mass in girls, regardless of the sport they practise. Fu-
ture research is needed to know the influence of the playing surfaces on
sports injuries, to find a balance between bone development and injury
prevention.

88
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

5. Optimisation of management in sports companies

Performance optimisation in sport is usually seen as improvement in
physical and sports performance. However, there is another field that
has been studied on performance improvement: sports management,
in this case, through the optimisation of business performance in sports
organisations. We understand that this field of study is vital for sports
development, as growth and quality improvement in sports organisa-
tions will help improve resources and installations available for athletes,
as well as reducing barriers for sports practice. Following the evolution
of research by the IGOID Group in these sections, we can find three work
areas in which the group has collaborated: i) management models and
systems; ii) finance of local sports management; iii) fitness sector man-
agement. We will divide this section into these areas.

Management models and systems

In this area we can find the first article of international relevance pub-
lished by this group.45 This paper analyses manager requirements for
sports facilities to obtain comprehensive sports management through
a qualitative research by a group of experts. In this paper, needs were
anticipated that are currently a reality, like management using mobile
devices or different billing elements, timetables and clients that are now
incorporated into popular CMS software (i.e. customer relationship man-
agement software, a kind of software usually used by sports centres and
marketed by several companies).

After which, the IGOID Group started to work on big data concepts ap-
plied to sport, in this case, from a management point of view. Taking ad-
vantage of the data published in the Spanish sports facilities census, two
studies were published. In the first research46 a performance measure-
ment system in terms of sports facility availability is presented, obtain-
ing a synthetic indicator that allows cataloguing Spanish regions based
on the development of their sports infrastructure. After a factorial analy-
sis, a total indicator is created, as well as three partials ones based on
density, quantity and quality of sports facilities. In the second research47
an analysis of the influence of the socioeconomic environment on the
sports infrastructure development was carried out, finding clear rela-
tionships. Thus, the level of physical activity practice and the economic
level of the population are significantly related with the development of
sports infrastructure (Table 2).

89
 sp or t science

GDP Unemployment Market Tourism Index of Physical

per rate share per index economic activity
capita capita per activity index
capita per
capita
Sports 0.541* -0.142 0.400 0.585* 0.385 0.561*
infrastructure
development

Note: * = p<0.005
Table 2. Pearson’s correlation between sports infrastructure development and
socioeconomic factors (adapted from47)

Finance of local sports management

Public administration has undergone many changes in its interests
since the implementation of New Public Management. Since then, in-
terest in performance assessment and measurement has increased.48
Within the most affected different local public services, we can find mu-
nicipal sports services given their growth over the years.47 Generally, lo-
cal public sports services, in most of the European and South American
countries, work to provide access to sports facilities and activities for all
sectors of the population and to achieve the highest levels of physical
activity among their population.49 To do this, they manage sports activi-
ties, as well as their own sports facilities.50,51

Therefore, a series of studies have been developed to show the reality

regarding the performance of municipal sports services, with a view of
improving their management and marking data that allows to set achiev-
able and realistic performances targets. In this sense, the researches52,53
are highlighted. In these studies, the IGOID Group shows national refer-
ence values and a benchmark of financial condition (i.e. system for ana-
lysing the economic performance of municipal services,54) in municipal
sports agencies from 2002 to 2011 (Table 3). Thanks to these results, it
was demonstrated that management performance results that theoreti-
cally were correct (and up to now were taken as reference points) under-
estimate or overestimate the results that could be considered as good
or bad depending on the reality of the sector. After this, the study is ex-
tended by García-Unanue et al.55, who analysed how the environment
can affect these indicators, in a way that the importance of each decision
on the performance results can be considered and which part can be at-
tributed to uncontrollable environmental factors. The study shows that
in the case of municipal sports services the political orientation does not

90
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

have a clear effect, while the population is determinant, making the ser-
vice management in big populations more difficult.

91
 sp or t science

Net Saving Index per capita

p25 p50 p75 % mean sd CV ANOVA

2002 -0.369 0.539 2.178 66.667 1.219 5.126 n.a.

2003 -0.426 0.797 1.961 68.041 0.855 5.914 n.a.

2004 -0.536 0.694 2.323 64.840 1.111 5.079 n.a.

2005 -0.032 0.699 2.360 73.934 1.767 6.467 n.a.

2006 -0.135 0.799 2.375 70.142 1.086 6.989 n.a.

2007 -0.481 0.700 2.793 67.111 2.221 7.660 n.a. F = 1.346 p = 0.208

2008 -0.851 0.546 2.608 64.186 0.872 5.063 n.a.

2009 -0.383 0.616 3.053 69.484 1.207 6.480 n.a.

2010 -0.713 0.585 3.083 64.384 1.424 6.038 n.a.

2011 -1.070 0.385 2.144 63.333 0.520 6.118 n.a.

WP -0.450 0.636 2.388 67.181 1.237 6.155 n.a.

Non-Financial Budgetary Result Index

p25 p50 p75 % mean sd CV ANOVA

2002 1.039 0.993 0.964 57.658 1.016 0.140 0.137

2003 1.034 0.997 0.958 55.155 1.004 0.120 0.120

2004 1.034 0.998 0.955 56.164 1.008 0.126 0.125

2005 1.018 0.991 0.942 63.981 0.993 0.150 0.151

2006 1.022 0.991 0.946 62.085 0.987 0.104 0.105

2007 1.024 0.993 0.947 61.778 0.984 0.123 0.125 F = 2.398 p = 0.010

2008 1.045 0.998 0.950 53.953 1.008 0.133 0.132

2009 1.032 0.998 0.955 55.399 1.015 0.173 0.171

2010 1.034 0.999 0.959 52.511 1.000 0.126 0.126

2011 1.056 1.000 0.970 52.857 1.032 0.171 0.165

WP 1.032 0.996 0.955 57.176 1.005 0.139 0.138

Note: % = Percentage of institutions with adequate theoretical results; WP = Whole Period.

Table 3. Descriptive statistics, benchmarks and ANOVA results53

92
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Self-Financing
p25 p50 p75 % mean sd CV ANOVA

0.182 0.306 0.460 21.622 0.355 0.272 0.767

0.176 0.278 0.462 22.165 0.352 0.283 0.803

0.173 0.295 0.477 21.005 0.353 0.274 0.778

0.169 0.284 0.441 19.431 0.325 0.233 0.717

0.173 0.285 0.455 19.905 0.325 0.228 0.702

0.175 0.264 0.432 17.333 0.316 0.222 0.702 F = 1.204 p = 0.288

0.187 0.265 0.412 17.674 0.317 0.211 0.666

0.183 0.283 0.448 20.657 0.326 0.212 0.650

0.188 0.303 0.484 23.288 0.362 0.298 0.822

0.193 0.304 0.486 24.286 0.358 0.239 0.667

0.178 0.287 0.453 20.711 0.339 0.249 0.735

Current Expenditures per capita

p25 p50 p75 % mean sd CV ANOVA

19.586 30.775 43.896 n.a. 36.890 28.074 0.761

19.909 31.508 44.290 n.a. 36.067 24.111 0.669

21.285 34.266 51.476 n.a. 41.031 34.859 0.850

21.270 34.705 50.609 n.a. 41.474 33.336 0.804

23.393 36.848 54.987 n.a. 44.392 38.325 0.863

25.253 40.312 61.266 n.a. 47.299 37.213 0.787 F = 4.130 p < 0.001

25.212 41.498 63.919 n.a. 49.620 38.797 0.782

27.277 41.991 63.150 n.a. 50.413 40.126 0.796

25.357 39.528 58.654 n.a. 48.386 44.188 0.913

23.440 38.555 57.126 n.a. 47.147 44.656 0.947

23.010 36.905 55.800 n.a. 44.337 37.208 0.839

93
 sp or t science

Fitness sector management

After analysing the performance of the public sports sector, the fol-
lowing question arises: What happens with the financial performance
of the private sector? First, it is necessary to point out that the private
sports sector orientated towards sport for-all is mainly developed for the
fitness sector. The fitness sector has grown enormously in recent years,
with strong chains and companies that each year have created aggressive
expansion plans.56 However, over the last years, performance problems
have begun, with annual client turnover rates of between 60.5 and 70%57
and a decline in profits.58 Therefore, the IGOID Group analysed the finan-
cial performance of the fitness sector over recent years, trying to detect
the focus of the problem to improve performance of this sector, and in
this way, improve access and quality of sports practice. Thus, Rodríguez-
Cañamero et al.59 detected lower net income over the last years, varying
by size of company, in a way that the sports centres managed by microen-
terprises or small companies mostly exhibited these lower profits. As we
can see in Figure 4, the most notable changes in the incomes and profits
are in 2008, the year of the beginning of the economic crisis, and in 2012,
when the VAT (value added tax) was increased. However, there is no clear
trend in medium and large companies. Therefore, public strategies and
policies should focus on microenterprises and small companies to facili-
tate the sustainability and survival of this sector.

94
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Figure 4. Evolution of fitness company results.59 The graphs are prepared at

different scales

Conclusions
This part has shown another vision of performance in sport, focusing
on sports management. Good management of sports companies, public
entities or clubs is vital for the health of sports systems. It is good man-
agement that breaks down practice barriers and endows the necessary
resources for sports development at all levels. For this reason, the IGOID
Group has materialised the research on performance of sports organi-
sation, as if they were athletes with the aim of improving their results.
Furthermore, the researches cover real problems in the different sectors

95
 sp or t science

of sports management such as sports facilities management, municipal

sports agencies or the fitness sector, providing lines of intervention and
improvement.

References
1. Burillo P, Gallardo L, Felipe JL, Gallardo AM. Artificial turf surfaces: Perception
of safety, sporting feature, satisfaction and preference of football users.
Eur J Sport Sci. 2014;14(sup1):437-47.
2. Felipe JL, Gallardo L, Burillo P, Gallardo A, Sánchez-Sánchez J, Plaza-Carmo-
na M. A qualitative vision of artificial turf football fields: elite players and
coaches. S Afr J Res Sport Ph. 2013;35(2):105-20.
3. FIFA. FIFA Quality Programme for Football Turf. Handbook of Test Methods.
Zurich: FIFA; 2015.
4. Burillo P, Gallardo L, Felipe, JL, Gallardo AM. Mechanical assessment of artifi-
cial turf football pitches: The consequences of no quality certification. Sci
Res Essays. 2012;7(28):2457-65.
5. Sánchez-Sánchez J, Felipe JL, Burillo P, del Corral J, Gallardo L. Effect of the
structural components of support on the loss of mechanical properties of
football fields of artificial turf. Proc IMechE, Part P: J Sports Engineering
and Technology. 2014;228(3):155-64.
6. Andersson H, Ekblom B, Krustrup P. (2008). Elite football on artificial turf
versus natural grass: Movement patterns, technical standards, and player
impressions. J Sports Sci. 2008;26(2):113-22
7. Nédélec M, McCall A, Carling C, Le Gall F, Berthoin S, Dupont G. (2013). Physi-
cal performance and subjective ratings after a soccer-specific exercise
simulation: Comparison of natural grass versus artificial turf. J Sports Sci.
2013;31(5):529-36.
8. Hughes MG, Birdsey L, Meyers R, Newcombe D, Oliver JL, Smith PM, et al. Ef-
fects of playing surface on physiological responses and performance vari-
ables in a controlled football simulation. J Sports Sci. 2013;31(8):878-86.
9. Brito J, Krustrup P, Rebelo A. The influence of the playing surface on the ex-
ercise intensity of small-sided recreational soccer games. Hum Mov Sci.
2012;31(4):946-56.
10. Stone KJ, Hughes MG, Stembridge MR, Meyers RW, Newcombe DJ, Oliver
JL. The influence of playing surface on physiological and performance re-
sponses during and after soccer simulation. Eur J Sport Sci. 2014;16(1):42-9.
11. Meyers MC. Incidence, Mechanisms, and Severity of Match-Related Colle-
giate Men’s Soccer Injuries on FieldTurf and Natural Grass Surfaces A 6-Year
Prospective Study. Am J Sports Med. 2016;doi:10.1177/0363546516671715.
12. Fleming P. Artificial turf systems for sport surfaces: current knowledge and
research needs. Proc IMechE, Part P: J Sports Engineering and Technology.
2011;225(2):43-63.
13. Zanetti EM, Bignardi C, Franceschini G, Audenino AL. Amateur football
pitches: mechanical properties of the natural ground and of different
artificial turf infills and their biomechanical implications. J Sports Sci.
2013;31(7):767-78.
14. López-Fernández J, Gallardo L, Fernández-Luna Á, Villacañas V, García-

96
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Unanue J, Sánchez-Sánchez J. Pitch size and Game Surface in Differ-

ent Small-Sided Games. Global Indicators, Activity Profile and Accel-
eration of Female Soccer Players. J Strength Cond Res. 2017;doi: 10.1519/
JSC.0000000000002090
15. López-Fernández J, Sánchez-Sánchez J, Gallardo L, García-Unanue J. Meta-
bolic Power of Female Footballers in Various Small-Sided Games with Dif-
ferent Pitch Surfaces and Sizes. Sports. 2017;5(2):1-24.
16. Sánchez-Sánchez J, García-Unanue J, Felipe JL, Jiménez-Reyes P, Viejo-
Romero D, Gómez-López M, et al. Physical and physiological responses
of amateur football players on 3rd generation artificial turf systems dur-
ing simulated game situations. J Strength Cond Res. 2016; doi: 10.1519/
JSC.0000000000001415
17. Sánchez-Sánchez J, García-Unanue J, Jiménez-Reyes P, Gallardo A, Burillo P,
Felipe JL, et al. Influence of the mechanical properties of third-generation
artificial turf systems on soccer players’ physiological and physical perfor-
mance and their perceptions. PLoS One. 2014;9(10):e111368.
18. Jones CM., Griffiths PC, Mellalieu SD. Training load and fatigue marker asso-
ciations with injury and illness: a systematic review of longitudinal studies.
Sports Med. 2017;47(5):943-74.
19. Carrasco L, Sañudo B, de Hoyo M, Ochiana G. Tensiomyographic character-
istics of Rectus Femoris after a single bout of intense exercise. Journal of
Social Sciences. 2011;7(3):354-57.
20. Ubago-Guisado E, Rodríguez-Cañamero S, López-Fernández J, Colino E,
Sánchez-Sánchez J, Gallardo L. (2017). Muscle contractile properties on dif-
ferent sport surfaces using tensiomyography. Journal of Human Sport and
Exercise. 2017;12(1):167-79.
21. Costello J, Stewart IB, Donnelly AE, Selfe J, Karki AI. Use of thermal imag-
ing in sports medicine research: a short report: short article. International
Sportmed J. 2013;14(2):94-8.
22. Hildebrandt C, Raschner C, Ammer K. An overview of recent application
of medical infrared thermography in sports medicine in Austria. Sensors.
2010;10(5):4700-15.
23. Vardasca R, Ring EFJ, Plassmann P, Jones CD. Thermal symmetry of the up-
per and lower extremities in healthy subjects. Thermology international.
2012;22(2):53-60.
24. Etehadtavakol M, Ng EY. Potential of Thermography in Pain Diagnosing and
Treatment Monitoring. In Application of Infrared to Biomedical Sciences.
Springer Singapore: Singapore; 2017. p. 19-32.
25. Felipe JL, Burillo P, Fernández Luna Á, García Unanue J. ¿Es viable el fútbol
de élite sobre césped artificial? El caso FIFA Women World Cup™. Revista de
Psicología del Deporte. 2016;25(3):81-4.
26. Serdula MK, Ivery D, Coates RJ, Freedman DS, Williamson DF, Byers T. Do
obese children become obese adults? A review of the literature. Prev Med.
1993;22(2):167-77.
27. Boreham C, Twisk J, Neville C, Savage M, Murray L, Gallagher A. Associations
between physical fitness and activity patterns during adolescence and car-
diovascular risk factors in young adulthood: the Northern Ireland Young
Hearts Project. Int J Sports Med. 2002;23(1):22-6.

97
 sp or t science

28. Bayios IA, Bergeles NK, Apostolidis NG, Noutsos KS, Koskolou MD. Anthro-
pometric, body composition and somatotype differences of Greek elite
female basketball, volleyball and handball players. J Sports Med Phys Fit-
ness. 2006;46(2):271-80.
29. Baroncelli GI, Bertelloni S, Sodini F, Saggese G. Osteoporosis in children and
adolescents. Paediatr Drugs. 2005;7(5),295-323.
30. Cruz JG, Martínez RF, Martínez JG, Gutiérrez ES, Serrano ME, De las Deses
CDH. Osteoporosis. Conceptos básicos para la práctica diaria. Revista de
Especialidades Médico Quirúrgicas. 2009;14(3):128-40.
31. Karlsson MK, Nordqvist A, Karlsson C. Physical activity increases bone mass
during growth. Food Nutr Res. 2008;52(1):1-10.
32. Vicente-Rodriguez G, Jimenez-Ramirez J, Ara I, Serrano-Sanchez JA, Dora-
do C, Calbet JA. Enhanced bone mass and physical fitness in prepubescent
footballers. Bone. 2003;33(5):853-9.
33. Gómez-Bruton A, González-Agüero A, Gómez-Cabello A, Matute-Llorente A,
Casajús JA, Vicente-Rodríguez G. The effects of swimming training on bone
tissue in adolescence. Scand J Med Sci Sports. 2015;31(8):878-86.
34. Vlachopoulos D, Barker AR, Williams CA, Arngrimsson SA, Knapp KM., Met-
calf BS, et al. The impact of sport participation on bone mass and geom-
etry in adolescent males. Med Sci Sports Exerc. 2016;49(2):317-26.
35. Ubago-Guisado E, Gómez-Cabello A, Sánchez-Sánchez J, García-Unanue J,
Gallardo L. Influence of different sports on bone mass in growing girls. J
Sports Sci. 2015;33(16):1710-8.
36. Gracia-Marco L, Ortega FB, Jiménez-Pavón D, Rodríguez G, Castillo MJ, Vi-
cente-Rodríguez G, Moreno LA. Adiposity and bone health in Spanish ado-
lescents. The HELENA study. Osteoporos Int. 2012;23(3):937-47.
37. El Hage R, Jacob C, Moussa E, Groussard C, Pineau JC, Benhamou CL, et
al. Influence of the weight status on bone mineral content and bone min-
eral density in a group of Lebanese adolescent girls. Joint Bone Spine.
2009;76(6):680-68
38. Pietrobelli A, Faith MS, Wang J, Brambilla P, Chiumello G, Heymsfield SB. As-
sociation of lean tissue and fat mass with bone mineral content in children
and adolescents. Obes Res. 2002;10(1):56-60.
39. Sayers A, Tobias JH. Fat mass exerts a greater effect on cortical bone mass
in girls than boys. J Clin Endocrinol Metab. 2010;95(2);699-706.
40. El Hage R, Moussa E, El Hage Z, Theunynck D, Jacob C. Influence of age and
morphological characteristics on whole body, lumbar spine, femoral neck
and 1/3 radius bone mineral apparent density in a group of Lebanese ado-
lescent boys. J Bone Miner Metab. 2011;29(4):477-83.
41. Ubago-Guisado E, García-Unanue J, López-Fernández J, Sánchez-Sánchez
J, & Gallardo L. Association of different types of playing surfaces with bone
mass in growing girls. J Sports Sci. 2017;35(15);1484-92.
42. Ubago-Guisado E, Martinez-Rodriguez A, Gallardo L, Sánchez-Sánchez
J. Bone mass in girls according to their BMI, VO2 max, hours and years of
practice. Eur J Sport Sci. 2016;16(8):1176-86.
43. Iacovelli JN, Jingzhen Y, Thomas G, Hongqian W, Schiltz T, Foster DT. The
effect of field condition and shoe type on lower extremity injuries in Ameri-
can Football. Br J Sports Med. 2013;47(12):789-93.

98
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

44. Plaza-Carmona M, Vicente-Rodriguez G, Martín-García M, Burillo P, Felipe

JL, Mata E, et al. Influence of hard vs. soft ground surfaces on bone accre-
tion in prepubertal footballers. Int J Sports Med. 2014;35(1):55-61.
45. Gallardo L, García-Tascón M, Burillo P. New sports management soft-
ware: A needs analysis by a panel of Spanish experts. Int J Inf Manage.
2008;28(4):235-45.
46. Gallardo L, Burillo P, García-Tascón M, Salinero JJ. The ranking of the regions
with regard to their sports facilities to improve their planning in sport: the
case of Spain. Soc Indic Res. 2009;94(2):297-317.
47. Burillo P, Barajas Á, Gallardo L, García-Tascón M. The influence of economic
factors in urban sports facility planning: a study on Spanish Regions. Eur
Plan Stud. 2011;19(10):1755-73.
48. Navarro-Galera A, Ortiz-Rodríguez D, López-Hernández AM. Identifying bar-
riers to the application of standardized performance indicators in Local
Government. Public Manag Rev. 2008;10(2):241-62.
49. Gallardo L, García-Tascón M, Burillo P, Salinero JJ. Differences of needs and
problematics between managers in public and private esports organisa-
tions in Spain. 2009; Paper presented at the The 16th EASM Conference,
Heidelberg.
50. Benito B, Solana J, Moreno MR. Assessing the efficiency of local entities in
the provision of public sports facilities. Int J Sport Financ. 2012;7:46-72.
51. Liu YD. Sport and social inclusion: evidence from the performance of public
leisure facilities. Soc Indic Res. 2009;90:325-37.
52. García-Unanue J, Felipe JL, Gallardo L. Autofinanciación en los organismos
autónomos deportivos municipales en España: cálculo y evolución. Revis-
ta Española de Educación Física y Deportes. 2014;(404):15-26.
53. García-Unanue JL Felipe J, del Corral J, Gallardo L. Assessing Financial Con-
dition of Municipal Sports Agencies: A Data Benchmarking Approach. The
Open Sports Sciences Journal. 2016;9(1).
54. Zafra-Gómez JL, López-Hernández AM, Hernández-Bastida A. Evaluating
financial performance in local government: maximizing the benchmarking
value. Int Rev Adm Sci. 2009;75:151-67.
55. García-Unanue J, Felipe JL, Gómez-González C, del Corral J, Gallardo L.
External constraints on Spanish municipal sports agencies’ finances. Aca-
demia Revista Latinoamericana de Administración. 2016;29(2):198-215.
56. IHRSA. The IHRSA European Market Rrport: the size and scape of the health
club industry. Boston: Club Business International; 2015
57. Bedford P. Retain & Gain: Keeping Your Members Engaged. 2009; Paper pre-
sented at the A Global Perspective: Challenges & Opportunities for Mem-
bership Growth & Retention, San Francisco.
58. LifeFitness. Estudio anual del mercado del fitness en españa 2015: Con-
texto, situación y tendencias del sector en España. Barcelona: Life Fitness
Iberia; 2015.
59. Rodríguez-Cañamero S, López-Fernández J, Villacañas V, Gallardo AM.
(2016). Evolución de los resultados del sector del fitness por tamaño de
empresa. In: Gálvez Ruiz, P. et al, editors. Economía, gestión y deporte. Una
visión actual de la investigación Cizur Menor: Thomson Reuters Aranzadi;
2016; p. 165-8.

99
 sp or t science

Propulsive forces and their efficient

application during swimming performance
Vassilios Gourgoulis1

This Chapter is dedicated to the memory of Prof. Robert E. Schleihauf

(1948 – 2016), one of the pioneers in the estimation of the propulsive forces
in swimming

When a swimmer’s body is immersed into the water, the force of grav-
ity acts on it and pulls it downward. Fortunately, the force of buoyancy
pushes the body upwards, against the force of gravity, helping the swim-
mer to float. During swimming, when the swimmer’s muscles contract to
flex or to extend the joints, moving the limbs, propulsive forces are gen-
erated to displace the body forward. However, because the limbs and
the body move through water, which is nearly 800 to 1000 times denser
than the air, resistive forces hinder the swimmer’s forward motion. Con-
sequently, any flaw is magnified exponentially in comparison with lo-
comotion through the air and the importance of the technique during
displacement through the water increases.1
Although it is not existing a single “perfect” technique for all swim-
mers, there are principles that could be applied to be decided if a swim-
mer’s technique is “good” or not. In this context, the forces acting on the
swimmer’s body should be examined. To increase forward speed, swim-
mers should reduce the resistive forces in comparison with the propul-
sive forces or overcome the resistive forces applying greater propulsive
forces. However, the propulsive forces should not only be greater than
the resistive forces, but should also be applied efficiently, with the least
waste in effort. To be applied efficiently, the resultant propulsive force
should be aimed as much as possible forward (along the desired swim-
ming direction) and should not be wasted in other directions, giving un-
necessary kinetic energy to the water.2,3,4,5,6

Resultant and effective force

When a swimmer’s arm moves through water, at each instant of time a
resultant propulsive force is produced by the swimmer’s hand (Figure 1).
However, not all of this force is directed horizontally forward, along the

1. Department of Physical Education and Sport Science, School of Physical

Education and Sport Science, Democritus University of Thrace, Komotini,
Greece
100
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

desired swimming direction. Only an amount is aimed horizontally for-

ward, and this is the component of the resultant propulsive force in the
horizontal axis, which coincides with the desired swimming direction. This
component is called effective force (Figure 2).

Figure 1: During an underwater arm stroke a resultant propulsive force is pro-

duced by the swimmer’s hand at each instant of time.

Figure 2: Not all, but only an amount of the resultant propulsive force is direct-
ed horizontally forward along the desired axis of swimming propulsion.
This component is called effective force.

101
 sp or t science

Swimmers should try as much as possible of the resultant force to

be aimed horizontally forward, along the desired swimming direction,
increasing the magnitude of the effective force.1 The magnitude of these
two forces, the resultant propulsive force and the effective force, should
be as close together as possible. In this case, a great fraction of the ex-
ternal power produced by the swimmer’s muscles is transformed into
useful thrust. On the contrary, when the force generated by the swimmer
is wasted in undesirable directions, the effective force is much smaller
than the resultant force. An amount of force is dissipated in giving un-
useful energy to the water, without contributing to propulsion, and may
cause undesirable sideways and vertical deviations of the body increas-
ing the resistive forces.4,6

Understanding the propulsive force

Although it is not yet quite clear the whole mechanism for the genera-
tion of the propulsive forces, it has been proved that when a swimmer’s
hand moves though water the pressure at the trailing (the back) side of
the hand is reduced and the pressure at the leading (the palmar) side
of the hand is increased. Thus, a pressure difference between the lead-
ing (high pressure area) and the trailing (low pressure area) side of the
hand is formed,7,8,9,10 which generates a complex collection of force vec-
tors that act from the water onto the hand11,12 and for simplifying reasons
could be represented as a resultant force.1
To understand this resultant propulsive force it can be decomposed,
as any vector, into its three perpendicular components, according to the
three fingers rule of the right hand, along the X, the Y and the Z axes
(Figure 3). The component of the resultant force along the X–axis is de-
fined to be opposite to the direction of the hand movement, which is de-
termined by its resultant velocity vector. This component is called drag
force.13 The rest two components of the resultant force along the Y and Z
axes create a force vector, which is perpendicular to the drag force vec-
tor and the direction of the hand’s motion. This force is called lift. Thus,
the lift force vector actually has two components and lies in a plane that
is perpendicular to the direction of motion.14,15,16 Resolving the resultant
force into its components (the drag and the lift forces), they can be used
to understand the forces that act on a swimmer’s hand (Figure 4).

102
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Figure 3: The resultant force vector can be projected along each axis (X, Y, Z)
of the coordinate system and be decomposed into its three perpendicular
components.

Figure 4: The resultant propulsive force vector can be resolved into the drag
and the lift force. The drag force is the component of the resultant force op-
posite to the resultant velocity of the hand and the lift force lies in a plane
perpendicular to the drag force and the resultant velocity of the hand.

During the underwater arm stroke, at each instant of time the result-
ant propulsive force can be broken down into these two components:
the drag force and the lift force. For instance, in front crawl swimming
(Figure 5), at the final part of the entry and catch phase (which lasted
from the hand’s entry into the water until the beginning of its backward
motion), because the hand moves downward the drag force (which by
definition is opposite to the direction of the hand’s motion) would be
aimed upward and the lift force (which by definition is perpendicular to
the motion of the hand and to the drag force) would be aimed forward. At
the end of the pull phase (which lasted from the beginning of the hand’s
backward motion until the time where the hand crosses the transverse

103
 sp or t science

plane passing through the shoulder joint), because the hand moves
backward and slightly upward, the drag force would be aimed forward
and slightly downward, and the lift force would be aimed upward and
slightly forward. At the final part of the push phase (which lasted from
the end of the pull phase until the hand’s exit out of the water), because
the hand moves upward, the drag force would be aimed downward, and
the lift force would be aimed forward. The resultant propulsive force is
always a combination of the drag and the lift propulsive forces.13,16,17,18

Figure 5: At each instant of time the resultant propulsive force (vector with
double line) can be decomposed into two components: the drag (vector
with solid line) and the lift force (vector with dashed line).

The drag force

The drag force is by definition opposite to the direction of the motion
of a limb. This means that, for instance, when the hand moves backward
the drag force would be directed forward. At each instant of time, the di-
rection of hand’s motion is determined by the direction of the resultant
velocity vector of the hand and thus, the drag force is always opposite to
the direction of hand’s velocity vector.13
The faster the motion of the hand and the bigger the surface area of
the hand, the greater is the drag force acting in the opposite direction
of the hand’s motion, because according to the hydrodynamic equation
(1), the drag force is proportional to the square of the resultant velocity
of the hand and the surface area of the hand:

104
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

  1 (1)
FDRAG = • C DRAG • ρ • v 2 • S
2

where, CDRAG is the hydrodynamic coefficient of drag, ρ is the density

of the water, v is the hand’s resultant velocity and S is the surface area
of the hand.

The lift force

The lift force lies by definition in a plane perpendicular to the direc-
tion of the hand’s motion and the drag force vector,15,16 and thus is always
perpendicular to the direction of the flow passing the moving hand.13
Similar to the drag force, the faster the motion of the hand and the
bigger the surface area of the hand, the greater is the lift force that is act-
ing perpendicular to the hand’s motion, because according to the hydro-
dynamic equation (2), the lift force is proportional to the square of the
resultant velocity of the hand and the surface area of the hand:

  1
FLIFT = • C LIFT • ρ • v 2 • S (2)
2
where, CLIFT is the hydrodynamic coefficient of lift, ρ is the density of
the water, v is the hand’s resultant velocity and S is the surface area of
the hand.

The direction of the resultant propulsive force

As it is already mentioned, when the swimmer’s hand moves through
water, the direction of its motion is determined by its resultant velocity
vector and due to the pressure difference between the leading and the
trailing side of the hand a resultant propulsive force is created. This force
can be decomposed into three components. The force component op-
posite to the resultant velocity vector of the hand (and the direction of
the motion of the hand) is called drag force. The other two components
can be composed into the lift force vector, which by definition lies al-
ways in a plane perpendicular to the drag force vector and the resultant
velocity vector of the hand. Thus, for reductive, simplifying and under-
standing reasons, the force generated by a swimmer’s hand can be re-
solved into the drag and the lift propulsive forces.1 However, actually, the
swimmers did not and cannot feel the drag and the lift forces. They feel
only the resultant force, due to the differential pressure.7 Moreover, this
approximation is a “quasi-steady state” approach, where it is assumed

105
 sp or t science

that the motion of the hand and the water flow velocity is constant and it
is not taken into account the unsteady nature of the swimmer’s propul-
sion,5,9,13,14,16,18 as will be discussed in detail at the end of this chapter.
For an efficient arm stroke, the resultant propulsive force should
be aimed as much as possible horizontally forward, along the desired
swimming direction, while forces acting in other directions are ineffi-
cient, require an amount of energy without contributing to propulsion,
and may cause undesirable sideways and vertical deviations of the body
increasing the resistive forces.4,6

Consequently, two questions arise:

ȕ
On what depends the direction of the resultant propulsive force?

ȕ
How can a swimmer modify the direction of the resultant propul-
sive force to be aimed mainly horizontally forward?

During an arm stroke, at each instant of time, the lift force is perpen-
dicular to the drag force and the resultant propulsive force is the vector
sum of the drag and the lift force vectors. Thus, modifying appropriately
the magnitude and the relative contribution of the drag and the lift force
vectors changed also the direction of their resultant force18 (Figure 6).

Figure 6: Modifying appropriately the magnitude and the relative contribution

of the drag and lift force vectors changed also the direction of their result-
ant force (a and b: during the pull phase; c and d: during the push phase)
(modified from Gourgoulis et al.18).

106
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

The next question arises is:

ȕ
How can a swimmer modify the relative contribution of the drag

and lift force?

To answer this question, the parameters that determine the magni-

tude of the drag and lift forces should be examined. According to the hy-
drodynamic equations (1) and (2), the drag and the lift force depend on
the density of the water (ρ), the surface area of the hand (S), the square
of the hand’s resultant velocity (v2) and the hydrodynamic coefficient
of drag (CDRAG) and the hydrodynamic coefficient of lift (CLIFT), respec-
tively.
For a swimming pool the density of the water (ρ) is given and for a
particular swimmer the surface area of the hand (S) is also unchange-
able, unless hand paddles are used, which increase the pulling surface of
the hand. However, their use is not allowed in competition. The resultant
velocity of the hand (v) can be changed, but this modification would af-
fect both the drag and the lift forces to the same extent. Thus, the only
remaining tool is to manage properly the drag and lift hydrodynamic co-
efficients.18
Both coefficients depend on the inclination and the orientation of the
hand, which is determined by two angles: the pitch angle and the sweep-
back angle of the hand.13

The pitch angle of the hand

The pitch angle, which is also called “angle of attack”, is the angle
formed between the plane of the hand and the resultant velocity vector
of the hand, which determines the direction of motion of the hand13,19,20
(Figure 7).

Figure 7: The pitch angle of the hand is the angle between its plane and its
resultant velocity vector.

107
 sp or t science

In swimming, during the underwater motion of the hand, at each in-

stant of time the inclination of the plane of the hand is changed, as well
as the direction of its resultant velocity vector. Consequently, the pitch
angle of the hand is also changed. For instance, in front crawl, during the
entry and catch phase the pitch angle is small, during the pull phase it is
large and during the final part of the push phase it becomes small again21
(Figure 8).

Figure 8: In front crawl swimming, during the entry and catch phase the pitch
angle is small, during the pull phase it is large and during the end of the
push phase it becomes small again.

The sweepback angle of the hand

The sweepback angle defines the leading edge of the hand mo-
tion, projecting the hand’s resultant velocity vector onto its plane. For
instance, when the thumb side leads the hand motion the sweepback
angle is zero (0) degrees, when the fingertips lead the motion the sweep-
back angle is 90 degrees, when the little finger side leads the motion the
sweepback angle is 180 degrees and when the wrist leads the hand mo-
tion the sweepback angle is 270 degrees13,19,20 (Figure 9).

108
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Figure 9: The sweepback angle of the hand defines the leading edge of the
hand motion.

During the arm stroke, for instance in front crawl swimming, the
sweepback angle is continuously changed, because different sides of the
hand lead the motion, and thus the sweepback angle varies (Figure 10).

Figure 10: The sweepback angle varies during the underwater arm stroke in
front crawl swimming because different sides of the hand lead the motion.

109
 sp or t science

The hydrodynamic coefficients of the hand

The hand’s hydrodynamic coefficients are dimensionless quantities,
which depend mainly on the inclination (pitch angle) and the orienta-
tion (sweepback angle) of the hand.
Schleihauf13 measured the forces acting onto a hand model and, us-
ing the hydrodynamic equations (1) and (2), determined drag and lift
hydrodynamic coefficients for discrete inclinations and orientations of
the hand model under steady – state conditions, in which the water flow
velocity was constant. The values of these coefficients are presented as
curves versus pitch angle for discrete sweepback angles of the hand (Fig-
ure 11).
However, it seems that Scheihauf13 decomposed the resultant force in
only two dimensions and calculated hydrodynamic coefficients for the
drag force and only one of the two lift force components. Schleihauf’s
work was extended by Berger et al.15 and Sanders16 calculating hydrody-
namic coefficients for all three components of the resultant force. Fur-
thermore, Sanders16 presented the drag and the two lift hydrodynamic
coefficients as continuous three-dimensional surfaces, enabling a more
accurate determination of the coefficients’ values for any possible pitch
and sweepback angle during actual swimming (Figure 12).

Figure 11: Lift and drag hydrodynamic coefficients of the hand versus the pitch
angle of the hand (0 to 90 degrees) for a given sweepback angle of 45 de-
grees (modified from Schleihauf 13).

110
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Figure 12: Three-dimensional surfaces of the hand’s hydrodynamic coefficients

for the drag force and the two components of the lift force, as function of
the pitch (-90 to 90 degrees) and the sweepback angle (0 to 360 degrees) of
the hand (modified from Sanders16).

Observing these coefficients, despite the discrepancies in their mag-

nitude, arises that when the pitch angle of the hand takes high values,
the values of the drag hydrodynamic coefficient are high. This means,
for instance, that during an underwater arm stroke when the plane of the
hand is almost perpendicular to the direction of the motion of the hand,
and the hand “presses” the water, the drag force values would be high.
On the contrary, when the pitch angle of the hand is small to medium,
and the hand “slices” through the water, the drag force generated by the
hand would be small. Concerning the values of the lift hydrodynamic co-
efficients it appears that when the pitch angle of the hand takes high val-
ues, the lift force values are low, while at small to medium pitch angles,
where the hand “slices” through the water, the lift force generated by the
hand is large.
To understand the major role and the importance of the pitch angle
of the hand on the magnitude of the drag and the lift hydrodynamic co-
efficients and consequently on the magnitude and the relative contribu-
tion of the drag and lift forces, a very simple experiment could be done.
Driving a car, if someone stretch his or her arm out of the window, hold-
ing the hand with the palm faced downward at a pitch angle of near zero
(0) degrees, the drag force of the air will push the hand and arm slightly
backward, without any elevation. This means that no lift force would
be experienced. If the pitch angle of the hand is changed with the palm
facing the direction of car’s motion to about 35 degrees, an increase in
pressure in the palmar side of the hand takes place and the magnitude
of both the drag and lift force components acting on the hand will in-
crease. Now the hand will be pulled back even more than before, due to
the increased drag force component, while simultaneously the lift force
component will push the hand upwards. If the pitch angle of the hand

111
 sp or t science

is changed to minus 35 degrees, the air above the back side of the hand
will have an increase in pressure, and the air pressure on the palmar side
of the hand will be decreased. Now, the lift force component acting on
the hand will be aimed downward, and the hand will move downward
and backward. If the hand is holding at a pitch angle of 90 degrees, only
the drag force of the air will push the hand and arm much more back-
ward than before, and no lift force will be experienced. So, it is obvious
that large pitch angle results in the generation of a large drag force, while
a medium to small pitch angle results in a large lift force.1, 22
From the above it is concluded that, since the magnitude of the drag
and lift hydrodynamic coefficients depend on the orientation and the in-
clination of the hand, changing the sweepback and, especially and more
importantly, the pitch angle of the hand, changed also the magnitude
and the relative contribution of the drag and lift forces.

The question arises now is:

ȕ
Why is it so important to know which of the two forces, the lift or the
drag force, makes the greater contribution to propulsion?

The stroking pattern

The knowledge of the type of the propulsive force, the drag or the lift
force, with the greater contribution has significant implications for swim-
ming technique, because it determines the emphasis of the propulsive
movements during the stroke and consequently the movement pattern
that should be used.1 If the drag force has the greater contribution to pro-
pulsion the stroking pattern should be primarily backwards and the swim-
mers should “press” the water with their hands using large pitch angles.
If the lift force has the greater contribution the emphasis should be given
to large lateral and vertical movements and the swimmers should “sweep”
their hands with small pitch angles.11
It is observed that high-skilled swimmers pull their hands primarily di-
agonally backwards against the water, rather than emphasize lateral move-
ments, in all strokes other than breaststroke.2,16 Even in breaststroke, during
the in-sweep phase a diagonally in and slightly backward motion results in
a better and more efficient orientation of the resultant force forward, along
the swimming direction, in comparison with a pure inward motion.13 When
the hand moves backwards with a large pitch angle the drag force is the
major contributor to the resultant force generated by the hand. On the con-
trary, the contribution of the lift force is maximized when the pitch angle of

112
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

the hand reaches medium values. However, even then, the contribution of
the drag force is also important as the contribution of the lift force.
Nevertheless, even though the stroking pattern of the hand should be
predominantly backward, swimmers do not and should not move their
hands exactly straight backwards. The stroking pattern aims not only to
create propulsive forces, but also to counteract rotational forces. Thus,
there is always some element of sideways movements that occur to sat-
isfy reaction accommodations, for instance, due to breathing, the body roll
and/or the recovering of the other arm. Moreover, a small amount of side-
ways movements contributes in a more effectively use of the muscle and
lever systems, increases the propulsive impulse lengthening the distance
travelled by the hand and helps to continually find still water to accelerate
backwards.2,11,23,24
To become efficient, for instance during front crawl swimming, swim-
mers should navigate their arms in such a manner that their palms should
always face toward their feet and actually “press” the water with their
hands primarily backwards with large pitch angles during the middle part
of the stroke and “sweep” their hands diagonally backward, outward and
upward, with small pitch angles and with a little-finger-ward out sweep mo-
tion during the final part of the underwater arm stroke.25,26 Pitching their
hands with such a manner, in the middle part of the stroke, during the pull
phase, swimmers are taking advantage of drag forces, with a smaller contri-
bution of the lift forces, due to the larger pitch angle of the hand, and in the
final part of the stroke, during the push phase, they generate more lift than
drag force, due to the smaller pitch angle of the hand.2,11,18,23,24 With such
navigation of the hand, the resultant force would be aimed more in the de-
sired direction (forward) and the effective force would not be much smaller
than the resultant force (Figure 13).
Consequently, in all swimming strokes other than breaststroke, swim-
mers should try to “press” the water primarily backwards toward their feet
and slightly diagonally outward, without a big and exaggerated “S”- shaped
pulling pattern (Figure 14). A slight “S” is acceptable, but large sweeping
motions, emphasizing lateral movements, should be avoided,2,27 whereas
in front crawl it is suggested that as the race distance decreases and the
swimming speed becomes more important, a straighter arm pull should
be adopted.28 The resultant force is always a combination of drag and lift
forces, but their relative contribution changes along the stroke, while the
statement of the overall domination of the drag or the lift force is an over-
simplification.18

113
 sp or t science

Figure 13: (a) Relative contribution of the drag and lift force in front crawl swim-
ming. (b) Resultant and effective force in front crawl swimming.

114
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Figure 14: Stroke pattern in front crawl swimming relative to swimmer’s body.
For an efficient application of the propulsive forces in front crawl the swim-
mers should “press” the water primarily backwards and slightly diagonally
outward, without emphasizing lateral movements.

Summarizing …
Question: How can a swimmer modify the direction of the resul-
tant force to be aimed mainly horizontally forward?
Answer: Modifying appropriately the magnitude and the relative
contribution of the drag and lift forces.

Question: How can a swimmer modify the relative contribution of

the drag and the lift forces?
Answer: Modifying the inclination and orientation of the hand, by
changing the pitch and the sweepback angle of the hand.

Question: How can a swimmer modify appropriately the pitch and

the sweepback angle of the hand?
Answer: Using an arm stroke pattern in a predominantly back-
ward direction.

Limitations and perspectives

The most popular procedure for estimating the propulsive forces
generated by a swimmer’s hand uses a combination of kinematic data
derived from underwater three-dimensional kinematic analysis of the
hand’s movement and predetermined hydrodynamic coefficients for
the hand obtained from water channel experiments with hand models.
Calculating, from the kinematic data, the pitch and the sweepback an-
gles of the swimmer’s hand at each instant of time during an underwater

115
 sp or t science

stroke, it is possible to determine the values of the drag and lift hydro-
dynamic coefficients, and using the hydrodynamic equations (1) and (2)
to estimate the magnitude of the drag and lift forces, the magnitude and
the direction of their resultant force and the magnitude of the effective
force.
However, using this approach there are a lot of limitations. Although
the directions of the drag and the lift forces can be determined with ac-
curacy, their exact magnitude is not easy to be specified.11 The drag and
lift forces are not measured directly. Instead, they are estimated indirect-
ly from kinematic data and specific hydrodynamic coefficients. There-
fore, there are some errors derived from the calibration of the recorded
space, from the distortion during underwater recordings, from the digi-
tizing of the selected points onto the hand and from the calculation of
pitch and sweepback angles of the hand.15,16,18 Moreover, the magnitude
of the hydrodynamic coefficients is not only a function of the pitch and
the sweepback angles of the hand, but also depends on the shape of the
hand’s model and the position of the fingers (especially the position of
the thump).29,30
Using the above mentioned methodology to estimate the propulsive
forces produced by a swimmer’s hand, which is called the “quasi-steady
state” approach,13,14 it is assumed that the flow under steady conditions
in a water channel is comparable to the flow during actual swimming.5
Moreover, only the hand’s speed, inclination and orientation relative to
the flow were considered,9 while the time-variant properties of force
generating mechanisms are ignored.31 It is assumed that the overall flow
field does not change very quickly with time. However, swimmers do not
move their hands and arms in steady conditions. Their stroke paths are
usually curved9,32 and their propelling segments accelerate, decelerate
and change their orientation, as they move through water. This means
that their motion, and consequently the water flow around them, is
unsteady. Thus, the “quasi-steady state” hydrodynamic theory is insuf-
ficient to describe the mechanisms by which humans propel themselves
through water.12,33 The propulsive forces generated by a swimmer’s hand
depend not only upon the size, the shape, the inclination, the orienta-
tion and the velocity of the hand, but also the effects of the accelera-
tion should be taken into account.23,32,34,35 Sanders16, by measuring the
forces acting on an accelerating hand showed that accelerations have
large effects on the total force and so must be considered in addition to
the instantaneous speed when estimating forces from time records of
hand motion. When the supplementary forces due to the acceleration

116
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

of the hand are considered the estimated magnitude of the propulsive

forces increased.18 On the contrary, when the acceleration of the hand is
not included, as in the “quasi–steady state” approach, these additional
forces are neglected.36
Moreover, in the “quasi–steady state” approach, the forces produced
due to the vortex production and the rotation of the arm are not con-
sidered. However, when a swimmer’s hand moves though water, trail-
ing vortices are created at the back side of the hand and detached from
the hand’s surface. These vortices, because of the high circulation of
the water, increase the flow velocity enhancing the reduction of the wa-
ter pressure behind the hand.7,9,12 A remarkable amount of momentum
is generated through this process of vortex production and shedding,
and swimmers can generate unsteady forces in addition to the steady
forces.8,12 Moreover, due to the fact that the arm movement is not only
a translation through water, but instead is mainly a complicated combi-
nation of translation and rotation, the tangential velocity near the hand
is higher than near the elbow and near the shoulder, causing a veloc-
ity gradient of the water close to the limb. This induces an accelerating
axial water flow component and an axial pressure gradient along the
arm and hand towards the fingertips, which seems to be more abrupt
at the trailing than at the leading side of the limb. Due to this axial pres-
sure gradient, the local pressure of the affected water close to the limb
is decreased towards the fingertips and the circumferential pressure dif-
ference between the leading (high pressure area) and the trailing (low
pressure area) side of the hand, due to its translational movement, is
enhanced increasing the magnitude of the propulsive forces.10
Furthermore, most of the attention has been paid to what the hand
does. However, the forearm,35,37 and even the lower part of the upper
arm,38 provide also effective propelling surfaces and their contribution
on the propulsive forces should also be considered.
Studying the water’s behaviour in the wakes and measuring its flow
around a swimmer’s limb or the whole body, it is likely to understand the
resulting forces that are generated. Thus, more sophisticated methods,
like Computational Fluid Dynamics (CFD), Particle Image Velocimetry
(PIV) and SWimming hUman Model (SWUM), should be used to under-
stand how the water reacts to the swimmer’s movements. CFD is a nu-
merical simulation technique to study the fluid flow, PIV is a quantitative
flow visualization method that can be used to provide measurements of
the instantaneous velocity vectors and related properties, such as nor-
mal and shear stress, in actual fluids and SWUM is a simulation model,

117
 sp or t science

which considers rigid body dynamics and unsteady fluid forces for the
whole body. Although these methods have also some limitations and de-
merits,9 combining the findings from CFD, PIV and SWUM, in conjunction
with direct pressure measurements, could help researchers, coaches
and swimmers greatly in visualizing and understanding the complicated
fluid dynamic mechanisms that generate propulsive forces in human
swimming.9,12,33

References
1. Schleihauf RΕ. Biomechanics of Human Movement. Bloomington, Indiana:
AuthorHouse; 2004.
2. Rushall BS, Sprigings EJ, Holt LE, Cappaert JM. A Re-evaluation of forces in
swimming. J Swim Res. 1994; 10: 6–30.
3. Sato Y, Hino T. A computational fluid dynamics analysis of hydrodynamic
force acting on a swimmer’s hand in a swimming competition. J Sport Sci
Med. 2013; 12: 679–689.
4. Toussaint HM, Hollander AP, Berg Cvd, Vorontsov AR. Biomechanics of
swimming. In: Garrett WE, Kirkendall DT, editors. Exercise and Sport Sci-
ence. Philadelphia, Lippincott: Williams & Wilkins, 2000; p. 639–660.
5. Toussaint HM, Truijens M. Biomechanical aspects of peak performance in
human swimming. Anim Biol. 2005; 55(1): 17–40.
6. Vorontsov AR, Rumyantsev VA. Propulsive forces in swimming. In: Zatsior-
sky VM, editor. Biomechanics in Sport – Performance, enhancement and
injury prevention. Oxford: Blackwell Science, 2000; p. 205–231.
7. Arellano R. Vortices and propulsion. In: Sanders R, Linsten J, editors. Swim-
ming: Applied Proceedings of the XVII International Symposium on Bio-
mechanics in Sports. Perth, Western Australia: School of Biomedical and
Sports Science, 1999; p. 53–66.
8. Matsuuchi K, Miwa T, Nomura T, Sakakibara J, Shintani H, Ungerechts BE.
Unsteady flow field around a human hand and propulsive force in swim-
ming. J Biomech. 2009; 42(1): 42–47.
9. Takagi H, Nakashima M, Sato Y, Matsuuchi K, Sanders RH. Numerical and
experimental investigations of human swimming motions. J Sports Sci.
2016; 34(16): 1564–1580.
10. Toussaint HM, Den Berg CV, Beek WJ. “Pumped-up propulsion” during
front crawl swimming. Med Sci Sports Exerc. 2002; 34: 314–319.
11. Maglischo EW. Swimming fastest. Champaign IL: Human Kinetics; 2003.
12. Takagi H, Nakashima M, Ozaki T, Matsuuchi K. Unsteady hydrodynamic
forces acting on a robotic arm and its flow field: Application to the crawl
stroke. J Biomech. 2014; 47(6): 1401–1408.
13. Schleihauf RE. A hydrodynamic analysis of swimming propulsion. In:
Terauds J, Bedingfield EW, editors. Swimming III. International Series of
Sports Science, Vol. 8. Baltimore, MD: University Park Press, 1979; p. 70–117.
14. Berger MAM, De Groot G, Hollander AP. Hydrodynamics drag and lift forces
on human hand/arm models. J Biomech. 1995; 28(2): 125–133.
15. Berger MAM, Hollander AP, De Groot G. Determining propulsive force in

118
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

front crawl swimming: A comparison of two methods. J Sports Sci. 1999;

17: 97–105.
16. Sanders RH. Hydrodynamic characteristics of a swimmer’s hand. J Appl
Biomech. 1999; 15: 3–26.
17. Gourgoulis V, Aggeloussis N, Vezos N, Kasimatis P, Antoniou P, Mavromatis
G. Estimation of hand forces and propelling efficiency during front crawl
swimming with hand paddles. J Biomech. 2008; 41: 208–215.
18. Gourgoulis V, Boli A, Aggeloussis N, Antoniou P, Toubekis A, Mavrommatis
G. The influence of the hand’s acceleration and the relative contribution
of drag and lift forces in front crawl swimming. J Sports Sci. 2015; 33(7):
696–712.
19. Schleihauf RE, Gray L, De Rose J. Three – dimensional analysis of hand
propulsion in the sprint front crawl stroke. In: Hollander AP, Huijing PA, de
Groot G, editors. Biomechanics and Medicine in Swimming. Champaign, IL:
Human Kinetics Publishers, 1983; p. 173–183.
20. Gourgoulis V, Antoniou P, Aggeloussis N, et al. Kinematic characteristics of
the stroke and orientation of the hand during front crawl resisted swim-
ming. J Sports Sci. 2010; 28(11): 1165–1173.
21. Gourgoulis V, Aggeloussis N, Vezos N, Antoniou P, Mavromatis G. Hand ori-
entation in hand paddle swimming. Int J Sports Med. 2008; 29: 429–434.
22. Sprigings EJ, Koehler JA. The choice between Bernoulli’s or Newton’s mod-
el in prediciting dynamic lift. Int J Sport Biomech. 1990; 6: 235–245.
23. Bixler B. Resistance and propulsion. In: Stager JM, Tanner DA, editors.
Swimming. Oxford, UK: Wiley-Blackwell Scientific Publications Limited,
2005; p. 59–101.
24. Sanders RH. Lift performance in aquatic sports. In: Riehle HJ, Vieten MM,
editors. Proceedings XVI International Symposium of Biomechanics in
Sports. Germany: University of Konstanz, 1998; p. 25–39.
25. Marinho DA, Barbosa TM, Kjendlie PL, et al. Swimming Simulation: A new
tool for swimming research and practical applications. In: Peters M, editor.
Computational fluid dynamics for sport simulation. Lecture Notes in Com-
putational Science and Engineering 72. Berlin Heidelberg: Springer-Verlag,
2009; p. 33–61.
26. Riewald S, Bixler B. CFD analysis of a swimmer’s arm and hand acceleration
and deceleration. In: Blackwell J, editor. XIX International Symposium on
biomechanics in sports. San Francisco: University of San Francisco, 2001;
p. 117–121.
27. von Loebbecke A, Mittal R. Comparative analysis of thrust production for
distinct arm-pull styles in competitive swimming. J Biomech Eng. 2012;
134(7): 074501-1 – 0745501- 4.
28. Nakashima M, Maeda S, Miwa T, Ichikawa H. Optimizing simulation of the
arm stroke in crawl swimming considering muscle strength characteristics
of athlete swimmers. J Biomech Sci Eng. 2012; 7(2): 102–117.
29. Marinho DA, Rouboa AI, Alves FB, et al. Hydrodynamic analysis of different
thumb positions in swimming. J Sport Sci Med. 2009; 8: 58–66.
30. Marinho DA, Barbosa TM, Reis VM, et al. Swimming propulsion forces are
enhanced by a small finger spread. J Appl Biomech. 2010; 26: 87–92.
31. Dickinson MH. Unsteady mechanisms of force generation in aquatic and

119
 sp or t science

aerial locomotion. Am Zool. 1996; 36: 537–554.

32. Bixler B, Schloder M. Computational Fluid Dynamics: An analytical tool for
the 21th century swimming scientist. J Swim Res. 1996; 11: 4–22.
33. Takagi H, Shimada S, Miwa T, Kudo S, Sanders R, Matsuuchi K. Unsteady
hydrodynamic forces acting on a hand and its flow field during sculling mo-
tion. Hum Mov Sci. 2014; 38: 133–142.
34. Pai Y-C, Hay JG. A hydrodynamic study of the oscillation motion in swim-
ming. Int J Sport Biomech. 1988; 4: 21–37.
35. Rouboa A, Silva A, Leal L, Rocha J, Alves F. The effect of swimmer’s hand/
forearm acceleration on propulsive forces generation using computational
fluid dynamics. J Biomech. 2006; 39: 1239–1248.
36. Kudo S, Vennell R, Wilson B. The effect of unsteady flow due to acceleration
on hydrodynamic forces acting on the hand in swimming. J Biomech. 2013;
46: 1697–1704.
37. Bixler B, Riewald S. Analysis of a swimmer’s hand and arm in steady flow
conditions using computational fluid dynamics. J Biomech. 2002; 35: 713–
717.
38. Lecrivain G, Slaouti A, Payton C, Kennedy I. Using reverse engineering and
computational fluid dynamics to investigate a lower arm amputee swim-
mer’s performance. J Biomech. 2008; 41: 2855–2859.

120
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Hyperthermia and dehydration in

competitive swimmers: current aspects and
recommendations
Orlando Laitano1,2, Lara Elena Gomes1, Alex J. Mattingly2 and
Flávio Antônio de Souza Castro3

Introduction
Swimming is one of the most popular Olympic sports, with individual
events in freestyle, backstroke, breaststroke, butterfly and medley. In
addition to these individual events, Olympic swimmers also participate
in freestyle and medley relays. The distances covered by a swimmer in
the aforementioned events range from 50 to 1500 m. In addition to these
events, since the Olympic Games in Beijing 2008, open water swimming
became an official event and its popularity has increased since then
with an explosion in global participation and the number of competitive
events. For instance, in London 2012 open water swimming was part of
the Olympic program, but as a swimming event. At the Rio 2016 Games,
open water swimming was considered a standalone modality, evidenc-
ing the world popularity it has achieved in the last few years.
The distances covered by open water swimmers at the Internation-
al Swimming Federation (FINA) World Championships vary from 5, 10
(Olympic distance) to 25 km. The 10 km event is described as the aquatic
equivalent of the marathon run and the duration of this event ranged
from 110 to 130 min for males and from 119 to 137 min for females’ elite
open water swimmers at the FINA/HOSA 10 km Marathon Swim. (World
Cup 2016 on Balatonfüred, Hungary). The open water events also attract
recreational swimmers who enjoy the challenges of prolonged physical
exertion mixed with unpredictable environmental conditions that are
not found in the controlled climate of the indoor swimming pool.
In spite of this, hyperthermia and dehydration are hardly considered
a problem in swimming even though training and competition, in many
times, involve prolonged high-intensity exercise1. Therefore, the aims of

1. Physical Education Collegiate, Federal University of Vale do São Francisco,

Petrolina/PE, Brazil
2. Department of Applied Physiology and Kinesiology, University of Florida,
Gainesville/FL, USA
3. Aquatic Sports Research Group, Federal University of Rio Grande do Sul,
Porto Alegre/RS, Brazil

121
 sp or t science

our chapter are: 1) to provide a summary of studies that have addressed

the issue of heat stress and fluid balance in swimmers; 2) to broaden the
framework of discussion regarding heat stress and fluid balance in swim-
mers to incite new studies addressing the unresolved issues in this field;
and 3) to provide evidence-based recommendations to coaches and ath-
letes to prevent hyperthermia and dehydration in swimming.

Hyperthermia in swimming
Skeletal muscle contraction increases metabolic heat production
in an intensity-dependent manner, which elevates core temperature2.
When exercise is performed on land, evaporation of sweat represents
the main mechanism by which the body maintains internal tempera-
ture within physiological range3. However, during exercise in water,
there is little possibility for evaporative heat loss because most of the
body is immersed in water; the areas of skin surface that are exposed to
air are wet, therefore sweat secretion would have a small contribution
for thermoregulation4. The main mechanisms for thermoregulation in
swimming are convection and conductance and therefore body’s abil-
ity to maintain core temperature is dependent mainly upon water tem-
perature as water is a more effective medium than air for convective heat
transfer due to its heat-transfer coefficient that is ~25 times greater than
air5, 6. In Figure 1 we report the main sources of heat gain and losses in
swimming as compared to land-based sports. Although the number of
heat gain sources is similar between them, the number of heat loss path-
ways is fewer in outdoor swimming.

Figure 1. Sources of heat gain/loss in outdoor/open water swimming versus on-

land outdoor sports.

122
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

While in indoor and some outdoor swimming pools the water is

typically kept under controlled temperatures, which reduces the risk of
hyperthermia, during open water swimming the water temperature de-
pends on the environmental conditions. A recent study demonstrated
that swimming at a water temperature of 33˚C was ineffective to induce
heat acclimation and was not ergogenic for swimmers7. When water
temperature is above skin temperature (~32-33˚C at rest) there will be
heat gain by the body6. In this regard, FINA has determined in 2013 a wa-
ter temperature limit of 31˚C for official open water swimming races for
safety reasons8. Nevertheless, ambient temperature is not the only fac-
tor that may increase the risk of hyperthermia in outdoor swimmers. For
instance, radiation has recently been shown to reduce exercise perfor-
mance9 and could be an additional factor that affects open water swim-
mers’ performance and those whose practices and races are performed
in outdoor swimming pools.
Likewise, under higher swimming intensities, metabolic heat pro-
duction will also increase body temperature. For example, if 1 L of O2
consumed generates about 20 kJ of heat, a swimmer with a VO2max of
50 ml/kg/min and body mass of 70 kg, swimming at an intensity corre-
spondent to 75% of VO2max, will generate ~3120 kJ (~750 kcal) of meta-
bolic heat/h of activity excluding environmental heat sources. Regarding
the intensity, along a 10 km open water event, heart rate was registered
as 141 ± 5 bpm, for adults with 24.4 ±36 years old15. Thus, given that an
average competitive open water swimmer races for ~2 h in the 10 km
event, one can estimate that metabolic heat production in combination
with increased water temperature will result in hyperthermia that may
hamper performance and risk athletes’ health.
There are reports of deaths due to heat illness during open water
swimming. During the 2010 FINA 10-km open-water World Cup in Fujai-
rah (Dubai), an elite athlete of team USA died10. Although medical details
of the autopsy have not been released, concerns have been raised about
the water temperature during the swimming competition11. FINA officials
claim that the water temperature exceeded 31˚C (near its finish at 11:30
AM). Furthermore, intriguing data indicate that the majority of deaths
in triathlon occur during the swim phase12, 13. The prevalence of deaths
by heat illness in swimming (e.g. open water swimming) is likely to be
underestimated14 because drowning is ultimately what leads to death.
As discussed in the next section, there has been a high prevalence of de-
hydration in swimmers during practices as demonstrated by changes in
body mass, elevated urine osmolality and/or specific gravity15, 16 and this

123
 sp or t science

may be another factor involved in the increased risk of heat-related ill-

nesses in swimmers.’

Dehydration in swimming
In exercise settings, significant increase in core temperature leads
to sweat-induced dehydration and hampers physical performance17.
There has been considerable interest in the effects of dehydration, de-
fined as acute reductions greater than 2% of body mass, on exercise
performance, and it has been fairly well established that higher levels of
sweat-induced dehydration can limit exercise performance17,18. Howev-
er, because sweating is not the main thermoregulatory pathway during
swimming, it is not entirely clear what swimmers’ sweat electrolyte com-
position and rate are. This information might be of great help to sports
dietitians to guide fluid replacement in swimmers. Likewise, whether
dehydration affects swimming performance remains a matter of debate.
Surprisingly, a rather limited number of studies have addressed these
questions because sweat electrolyte composition assessment is particu-
larly challenging in swimming settings.
In comparison with exercise on land, it is puzzling to define sweat
electrolyte composition in swimmers while in the swimming pool. In
studies where sweat electrolyte composition is determined, absorbent
sweat patches are attached to various anatomical sites after the skin is
thoroughly cleaned with deionized water and dried. Thereafter, bottles
containing fluid (e.g., sports drink and/or plain water) identified with
athletes’ names are weighed before the given practice. Athletes are in-
structed to drink only from their personal bottles and not to spit any of
the fluid out. They are also instructed to urinate in a container if needed
during the practice so that this lost mass can be considered for sweat
rate calculations. After the activity, sweat patches are removed and the
body is towelled dry before post- practice body mass is recorded. Final-
ly, bottles are re-weighed so the volume consumed during the training
session or race can be calculated and considered for sweat rate calcu-
lations. This method is used to determine sweat rate, ad libitum fluid
intake and percent change in body mass (i.e., fluid balance) and helps to
identify those athletes with high sweat sodium losses who may need to
pay attention to sodium replacement.
While this technique is widely used in land-based sports19, 20, its use
in swimming is limited because of the possibility of the pool water af-
fecting sweat composition by saturating the sweat absorbent pad. How-
ever, to date, only one research group4 investigated fluid and electrolyte

124
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

balance in well trained swimmers using the abovementioned technique

during 105 min long training sessions in a pool with water temperature
of 27.4˚C. Mean sweat volume was 548 ± 243 ml and a sweat rate of 310 ±
100 ml/h. Sweat Na+ was 43 ± 14 mmol/L and Cl- was 31 ± 9 mmol/L. De-
spite the relatively low sweat rate reported by the authors, it is possible
to identify salt losses through swimmers’ sweat even though sweat is not
the main thermoregulatory pathway for these athletes.
A study assessing sweat rate, but not composition, of swimmers re-
ported a range of 314 to 415 ml/h21, which was similar to another study
reporting a sweat rate of 480 ml/h during prolonged swimming22. Cade
et al. reported higher sweat rates ranging from 1180 to 1620 ml/h during
2 hours of high intensity swim training, but as the focus of the study was
on muscle damage, no further detail was provided regarding ambient
and water temperature23. These studies report sweat rate as millilitre
per hour and others have suggested that sweat rate should be expressed
as ml/km to better translate into educational messages for swimmers
since training sessions for such athletes are mostly based on distance to
be swum rather than time21.
Sweat rate and composition have been reported to be different in
swimmers. Henkin et al24 compared sweat rate and electrolyte concen-
tration in swimmers, runners, and non-athletes during cycling in the heat
(35˚C). Swimmers exhibited a lower sweat volume (900 ± 300 ml) than
runners (1500 ± 200 ml), but similar to non-athletes (600 ± 200 ml). Swim-
mers sweat Na+ (65.4 ± 5.5 mmol/L) and Cl- (61.2 ± 8.1 mmol/L) content
were higher than runners (Na+ = 45.2 ± 7.5; Cl- = 38.9 ± 8.3 mmol/L), but
similar to non-athletes (Na+ = 67.3 ± 8.5; Cl- = 58.3 ± 9.6 mmol/L). These
results suggest differential adaptations of sweat glands in swimmers
when compared to runners, which agrees with an earlier study suggest-
ing that swimmers may not develop the same level of heat acclimati-
zation as their training sessions are conducted in an environment that
does not allow sweat evaporation25 or probably simply because of less
heat stress. Even so, it is possible to speculate that swimmers exercising
for longer than 1 h on land might need to pay attention to electrolyte
losses due to their potential to lose more salts through sweating.
One recent study performed with adolescent swimmers reported
that more than 75% (out of 46 participants) of the swimmers were hy-
pohydrated before practice15. This data matches with reports from dif-
ferent sports modalities indicating that athletes are hypohydrated be-
fore starting training sessions26,27. Higham et al16 found that 85% of the
swimmers were hypohydrated before training sessions based on urine

125
 sp or t science

specific gravity (USG) analysis. Pre-hydration level is usually determined

by urine osmolality, USG, or urine colour scale28. To prevent dehydration,
coaches and athletes can perform these measurements bearing in mind
that urine colour may not be an indicator of post practice hydration level
as swimming can change urine osmolality regardless of the hydration
status15. The effect of water pressure against the skin decreases superfi-
cial vascular transmural pressure (pressure inside the vessel minus the
outside pressure), thus causing haemodilution during the initial phase
of immersion29, 30. As immersion continues (> 1/2 h) a reflex renal diuresis
occurs and haemoconcentration prevails.

Factors that increase the risk of hyperthermia and/or dehydra-

tion in swimmers
Hyperthermia and dehydration are more likely to occur in open water
than in indoor swimmers. However, there are extrinsic factors that can
increase the chance of hyperthermia and dehydration even in indoor
swimmers. For instance, there are multiple factors that may predispose
athletes to heat illnesses that are not discussed in the context of sports.
In Table 1 we list a number of factors that are known to increase the risk
of hyperthermia and dehydration in athletes. In addition, evidence sug-
gests that prior bouts of illness or inflammation may increase the risk
of exertional heat illness regardless of the age, acclimatization levels, or
other risk factors31,32. It is worth noting that most evidences come from
athletes of other sports or military personnel who are potentially ex-
posed to similar exertional levels as swimmers. Therefore, future studies
are needed to test the extent by which these factors can indeed be an
issue in swimming.

Risk Factor Etiology

>31˚C water temperature reduced/negligible cooling via convection
Sunlight increased radiant heat transfer
Overtraining/High competition causes hypohydration, reduced plasma volume &
frequency thermoregulatory capacity
Excessive alcohol consumption metabolic stimulation and inhibition of vasomotor
reflexes that normally facilitates heat dissipation
Recent viral/bacterial infection excess proinflammatory and pyrogenic cytokines,
reduced thermoregulatory function
Lower fluid and electrolyte reduced plasma volume
consumption

126
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Acute/chronic NSAID use impaired heat tolerance, increased intestinal

permeability, potentially increased chance of
endotoxemia

Table 1: Risk factors for hyperthermia and dehydration in swimmers

Previous bouts of infection/inflammation

Athletes exposed to high exercise training volumes and intensities
may develop immunosuppression that may increase the risks of de-
veloping viral or bacterial infection33-35. For instance, swimmers with a
lower pre-season salivary IgA were more likely to contract upper respira-
tory tract infection during a 7-month training period36,37. Infection and
inflammation are, sometimes, accompanied by fever38. Fever from a
pre-existing illness may increase the normal hyperthermic response to
exercise31. Dehydration has been discussed as one mechanism involved
in the increased risk of hyperthermia in athletes with a previous period
of infection, as it compromises thermoregulatory control mechanisms
and was found to be a risk factor in young male runners that suffered
heat stroke following respiratory or GI illness39. Although more research
is needed to establish the impact that a previous bout of inflammation
and/or infection has on swimmers’ thermoregulatory response, caution
should be taken to avoid hyperthermia when returning from these con-
ditions.

Use of Non-Steroidal Anti-Inflammatory Drugs (NSAIDS)

NSAIDs (e.g., naproxen and aspirin) are one of the most common pre-
scription and over the counter medications used to treat shoulder inju-
ries and muscle pain/discomfort in swimmers40. The use of NSAIDS has
a high prevalence among athletes and limited awareness of potential
side effects41. The effect of high environmental temperatures on NSAIDs
toxicity has not been well appreciated despite early reports showing in-
creased aspirin toxicity (measured as lethality) in rats passively exposed
to 36˚ C versus a thermoneutral environment42. One possible mecha-
nism for the increased risk of heat illness when hyperthermia, exercise,
and the use of NSAIDs are superimposed is the potential for enhanced
intestinal permeability as NSAIDs can increase intestinal permeability
within 24h of ingestion43 and hyperthermia can increase ischemia-rep-
erfusion induced intestinal damage44. This combination may lead to en-
dotoxemia and advance for systemic inflammatory response syndrome.
Thus, swimmers taking NSAIDs might have increased risk of heat illness-
es although further work is needed to determine the magnitude of risk.

127
 sp or t science

Nutritional aspects
The use of alcohol is often intimately associated with sport. Although
alcohol can provide a source of energy to the working muscle, the nega-
tive effects on metabolic, cardiovascular, thermoregulatory, and neu-
romuscular aspects that may hinder performance are well known45.
An early report demonstrated moderate alcohol consumption among
young swimmers ranging from rarely to 2.1 L of beer per week46. Alcohol
enhances diuresis and may lead to dehydration47. The magnitude of diu-
resis experienced is thought to be dependent upon the amount of alco-
hol consumed, and its effects on hydration status will be influenced by
the concentration of alcohol relative to water in ingested beverages48-50.
Shirreffs and Maughan48 investigated effects of several beverages with
a low alcohol (0, 1, 2 and 4%) content (2.212 ± 153 ml) on the restoration
of water and electrolyte balance immediately after exercise-induced de-
hydration. They concluded that there was no difference in the recovery
of body water content from dehydration with a beverage of up to 2%
alcohol, but that the 4% alcohol drink increased urine output compared
with the alcohol-free drink. This is consistent with the results of a study
demonstrating that an additional 100 ml of urine would be produced for
each 10 g of alcohol ingested, though this estimate was based on data
from only a single subject49. Another early study was able to show more
convincingly that the diuretic response in rats was proportional to the
concentration of alcohol in a fluid bolus delivered by gastric intubation50.
Alcohol consumption is an independent risk factor for heat illnesses due
to the metabolic stimulation and inhibition of vasomotor reflexes that
normally facilitates heat dissipation31.
The use of creatine as a nutritional supplement is common by ath-
letes including swimmers. While primarily used for potential benefits
in sports of short duration and high intensity, its use has been seen in
longer duration activities as well. Older theory suggested that creatine
supplementation, through increasing intramuscular creatine phosphate
stores, could potentially cause dehydration via increased osmotic pull of
water from the extracellular fluid into the intracellular compartment51.
It was thought that this impairment of ability to shift fluid into the extra-
cellular space could be detrimental to maintenance of plasma volume
during periods of high sweat rates required for thermoregulation51. This
theory has been abandoned, as existing data on the effects of creatine
on hydration status indicate an increase in total body water, typically
without concurrent shifts in compartmental distribution52-54. Data indi-
cates acute increases in total body water and ICF within 3 days55, with

128
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

longer time periods (~28 days) yielding no differences in fluid distribu-

tion54. Studies show that creatine does not hinder thermoregulation51, 56,
57
, in either men or women58, and during exercise in the heat59, with some
data supporting improvements in thermoregulatory measures including
core temperature and heart rate60, 61. Current data supports the theory
that by encouraging a hyperhydrated state via increased total body wa-
ter, thermoregulatory mechanisms are not impaired, and may poten-
tially be improved.
Caffeine consumption is pervasive in everyday food items and taken
purposefully as a nutritional supplement. Caffeine is a known diuretic
and has been previously thought to potentially exacerbate dehydra-
tion and/or electrolyte imbalance, which could contribute to impaired
thermoregulation during exercise in the heat. However, current data
indicates caffeine to lack negative effects on fluid and electrolyte bal-
ance62-64, including during exercise in heat65,66, in chronic consumers67,
and during post workout rehydration68. Use of caffeine is not detrimen-
tal, and may indeed be a useful adjunct to a comprehensive nutritional
plan in long duration open water swimming events69.

Overdrinking, hyponatremia, and drink temperature

Besides the aspects discussed above, many other aspects can have
potential impact on swimming performance, but they still deserve more
research. For instance, overdrinking can be as harmful as dehydration71.
However, it is challenging to precisely monitor the volume of water in-
take in swimmers because they often involuntarily drink water from the
pool/sea/river. One major concern related to overdrinking fluids with
low electrolyte concentration is hyponatremia (low serum sodium con-
centration – below 135 mEq/L), which can impair brain function due to
edema. The incidence of hyponatremia in swimmers is thought to be
higher in females than in males open-water swimmers.
For instance, Wagner et al.72 investigated the prevalence of exercise-
associated hyponatremia in 25 male and 11 female open-water ultra-en-
durance swimmers participating in the ‘Marathon-Swim’ in Lake Zurich,
Switzerland, covering 26.4 km. They found that two males (8%) and four
females (36%) developed hyponatremia where one female was sympto-
matic with plasma sodium [Na+] of 127 mmol/L. Interestingly, monitored
fluid intake was neither associated with changes in body mass, post-race
plasma sodium or the change in plasma sodium, suggesting that swim-
mers drank water from other sources other than that provided by their
staff. It is important to highlight that overdrinking is not the only cause

129
 sp or t science

of hyponatremia. Other factors such as inadequate secretion of suppres-

sion of antidiuretic hormone, which leads to excessive fluid retention
and fail to mobilize osmotically inactive sodium from internal sources
are potential factors involved in the development of hyponatremia72.
Whether there is an optimal drink temperature to enhance perfor-
mance in swimmers is another question that deserves attention. Hue
et al.15 assessed the effects of ingesting drinks with different tempera-
tures (cold = 1.1˚C or neutral = 28˚C) during open water swimming in a
tropical environment. Cold-water ingestion significantly decreased core
temperature. However, heart rate and performance were not affected by
drink temperature. Although these results suggest that drink tempera-
ture do not affect overall performance, it may be a useful strategy to pre-
vent hyperthermia in open water swimmers, mainly when competition
takes place in hot weather.

Practical recommendations to avoid significant dehydration and

heat stress in swimmers
The following recommendations aim to prevent significant
dehydration and heat stress in swimmers considering the evidences re-
ported in this review.

ғ
Education about the importance of fluid ingestion to swimmers’

performance is fundamental and should start in the early stages
of their career.
ғ
Track changes in body mass during training and races in different
environmental conditions to determine individual sweating rates
and hydration habits. This will help identify swimmers who are at
risk of significant dehydration.
ғ
Individualize the hydration strategy based on swimmers’ sweat-
ing rate and drink preferences (e.g., beverage type and flavour to
promote voluntary fluid intake).
ғ
Swimmers should drink enough fluid during training/races to
prevent >2% dehydration. One strategy to achieve this goal in
swimmers is to encourage them to drink throughout a day of
competition. Overdrinking relative to sweat losses should also
be avoided.
ғ
After exercise, if dehydration is severe (>5% of body mass) or rap-
id rehydration is needed (e.g., < 24 h before next practice or race)
drink ~1.5 L of fluid for each 1 kg of body mass deficit70.
ғ
Consuming a beverage with sodium or sodium-containing snacks

130
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

/ foods, helps replace sweat sodium losses, stimulate thirst and

retain the ingested fluids.
ғ
Caffeine and/or creatine use are unlikely to affect thermoregula-
tory mechanisms negatively.
ғ
Avoid high intensity training sessions and/or competition when
returning from periods of infection or prolonged NSAIDs use.
ғ
Be aware that alcohol is likely to increase the risk of heat-related
illnesses.

Conclusions
Swimmers and coaches often neglect hyperthermia and dehy-
dration probably because swimming is a sport performed in an environ-
ment that favours heat loss and because sweat evaporation is not the
main thermoregulatory pathway involved. Even so, evidences suggest
that salt losses occur through swimmers’ sweat despite a lower sweat
rate compared to athletes from other land-based sports. Evidences also
suggest that swimmers are likely to start practices and races already hy-
pohydrated, which further enhances the danger of heat related illnesses
and impairments in performance. Open water swimmers are, often-
times, exposed to environments that increase the risk of hyperthermia
and dehydration due to the duration of the races and the unpredictable
weather conditions. We identified factors that may enhance the risk of
hyperthermia and dehydration in swimmers such as the use of NSAIDs
and use of alcohol. In addition, swimmers returning from periods of in-
fection/inflammation must bear in mind the potential for compromised
thermoregulatory function during training and races. Creatine supple-
mentation and caffeine consumption are unlikely to compromise ther-
moregulatory response. In fact, the latter may be a useful adjunct to a
comprehensive nutritional plan in long duration swimming events.

References
1. Mujika I, Chatard JC, Busso T, Geyssant A, Barale F, Lacoste L. Effects of
training on performance in competitive swimming. Can J Appl Physiol.
1995;20(4):395-406.
2. Gonzalez-Alonso J, Quistorff B, Krustrup P, Bangsbo J, Saltin B. Heat pro-
duction in human skeletal muscle at the onset of intense dynamic exercise.
J Physiol. 2000;524 Pt 2:603-615.
3. Shibasaki M, Wilson TE, Crandall CG. Neural control and mechanisms of
eccrine sweating during heat stress and exercise. J Appl Physiol (1985).
2006;100(5):1692-1701.
4. Maughan RJ, Dargavel LA, Hares R, Shirreffs SM. Water and salt bal-

131
 sp or t science

ance of well-trained swimmers in training. Int J Sport Nutr Exerc Metab.

2009;19(6):598-606.
5. Nielsen B, Davies CT. Temperature regulation during exercise in water and
air. Acta Physiol Scand. 1976;98(4):500-508.
6. Costill DL, Cahill PJ, Eddy D. Metabolic responses to submaximal exercise in
three water temperatures. J Appl Physiol. 1967;22(4):628-632.
7. Bradford CD, Lucas SJ, Gerrard DF, Cotter JD. Swimming in warm water is
ineffective in heat acclimation and is non-ergogenic for swimmers. Scand J
Med Sci Sports. 2015;25 Suppl 1:277-286.
8. FINA. FINA Open Water Swimming Rules. 2015; http://www.fina.org/sites/
default/files/finaowrules_20152017.pdf. Accessed 18th June 2016, 2016.
9. Otani H, Kaya M, Tamaki A, Watson P, Maughan RJ. Effects of solar radiation
on endurance exercise capacity in a hot environment. Eur J Appl Physiol.
2016;116(4):769-779.
10. Croucher M. Death of US swimmer Fran Crippen was ‘a disaster waiting to
happen’. 2012; http://www.thenational.ae/news/uae-news/death-of-us-
swimmer-fran-crippen-was-a-disaster-waiting-to-happen. Accessed June
18th, 2016.
11. Macaluso F, Barone R, Isaacs AW, Farina F, Morici G, Di Felice V. Heat stroke
risk for open-water swimmers during long-distance events. Wilderness En-
viron Med. 2013;24(4):362-365.
12. Harris KM, Henry JT, Rohman E, Haas TS, Maron BJ. Sudden death during
the triathlon. JAMA. 2010;303(13):1255-1257.
13. Eichner ER. The mystery of swimming deaths in athletes. Curr Sports Med
Rep. 2011;10(1):3-4.
14. Asplund CA, Creswell LL. Hypothesised mechanisms of swimming-related
death: a systematic review. Br J Sports Med. 2016.
15. Hue O, Monjo R, Riera F. Imposed Cold-water Ingestion during Open Water
Swimming in Internationally Ranked Swimmers. Int J Sports Med. 2015; 36
(11), 941-946.
16. Higham DG, Naughton GA, Burt LA, Shi X. Comparison of fluid balance be-
tween competitive swimmers and less active adolescents. Int J Sport Nutr
Exerc Metab. 2009;19(3):259-274.
17. Wendt D, van Loon LJ, Lichtenbelt WD. Thermoregulation during exercise in
the heat: strategies for maintaining health and performance. Sports Med.
2007;37(8):669-682.
18. Sawka MN, Noakes TD. Does dehydration impair exercise performance?
Med Sci Sports Exerc. 2007;39(8):1209-1217.
19. Shirreffs SM, Aragon-Vargas LF, Chamorro M, Maughan RJ, Serratosa L,
Zachwieja JJ. The sweating response of elite professional soccer players to
training in the heat. Int J Sports Med. 2005;26(2):90-95.
20. Maughan RJ, Shirreffs SM, Merson SJ, Horswill CA. Fluid and electrolyte bal-
ance in elite male football (soccer) players training in a cool environment. J
Sports Sci. 2005;23(1):73-79.
21. Cox GR, Broad EM, Riley MD, Burke LM. Body mass changes and voluntary
fluid intakes of elite level water polo players and swimmers. J Sci Med
Sport. 2002;5(3):183-193.
22. Lemon PW, Deutsch DT, Payne WR. Urea production during prolonged

132
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

swimming. J Sports Sci. 1989;7(3):241-246.

23. Cade JR, Reese RH, Privette RM, Hommen NM, Rogers JL, Fregly MJ. Dietary
intervention and training in swimmers. Eur J Appl Physiol Occup Physiol.
1991;63(3-4):210-215.
24. Henkin SD, Sehl PL, Meyer F. Sweat rate and electrolyte concentration
in swimmers, runners, and nonathletes. Int J Sports Physiol Perform.
2010;5(3):359-366.
25. McMurray RG, Horvath SM. Thermoregulation in swimmers and runners. J
Appl Physiol Respir Environ Exerc Physiol. 1979;46(6):1086-1092.
26. Logan-Sprenger HM, Palmer MS, Spriet LL. Estimated fluid and sodium bal-
ance and drink preferences in elite male junior players during an ice hockey
game. Appl Physiol Nutr Metab. 2011;36(1):145-152.
27. Osterberg KL, Horswill CA, Baker LB. Pregame urine specific gravity and
fluid intake by National Basketball Association players during competition.
J Athl Train. 2009;44(1):53-57.
28. Armstrong LE, Pumerantz AC, Fiala KA, et al. Human hydration indices:
acute and longitudinal reference values. Int J Sport Nutr Exerc Metab.
2010;20(2):145-153.
29. Khosla SS, DuBois AB. Fluid shifts during initial phase of immersion diuresis
in man. J Appl Physiol Respir Environ Exerc Physiol. 1979;46(4):703-708.
30. Epstein M. Renal effects of head-out water immersion in man: implications
for an understanding of volume homeostasis. Physiol Rev. 1978;58(3):529-
581.
31. Leon LR, Bouchama A. Heat stroke. Compr Physiol. 2015;5(2):611-647.
32. Gardner JW, Kark JA, Karnei K, et al. Risk factors predicting exertional heat
illness in male Marine Corps recruits. Med Sci Sports Exerc. 1996;28(8):939-
944.
33. Atamaniuk J, Vidotto C, Tschan H, Bachl N, Stuhlmeier KM, Muller MM. In-
creased concentrations of cell-free plasma DNA after exhaustive exercise.
Clin Chem. 2004;50(9):1668-1670.
34. Brahmi Z, Thomas JE, Park M, Park M, Dowdeswell IR. The effect of acute
exercise on natural killer-cell activity of trained and sedentary human sub-
jects. J Clin Immunol. 1985;5(5):321-328.
35. Edwards AJ, Bacon TH, Elms CA, Verardi R, Felder M, Knight SC. Changes
in the populations of lymphoid cells in human peripheral blood following
physical exercise. Clin Exp Immunol. 1984;58(2):420-427.
36. Pyne DB, Gleeson M. Effects of intensive exercise training on immunity in
athletes. Int J Sports Med. 1998;19 Suppl 3:S183-191; discussion S191-184.
37. Gleeson M, Hall ST, McDonald WA, Flanagan AJ, Clancy RL. Salivary
IgA subclasses and infection risk in elite swimmers. Immunol Cell Biol.
1999;77(4):351-355.
38. Kluger MJ, Kozak W, Conn CA, Leon LR, Soszynski D. Role of fever in disease.
Ann N Y Acad Sci. 1998;856:224-233.
39. Richards R, Richards D, Whittaker R. Method of predicting the number of
casualties in the Sydney City-to-Surf fun runs. Med J Aust. 1984;141(12-
13):805-808.
40. Alaranta A, Alaranta H, Helenius I. Use of prescription drugs in athletes.
Sports Med. 2008;38(6):449-463.

133
 sp or t science

41. Gorski T, Cadore EL, Pinto SS, et al. Use of NSAIDs in triathletes: prevalence,
level of awareness and reasons for use. Br J Sports Med. 2011;45(2):85-90.
42. Gordon CJ. Temperature Regulation in Laboratory Rodents: Cambridge
University Press; 1993.
43. Bjarnason I, Takeuchi K. Intestinal permeability in the pathogenesis of
NSAID-induced enteropathy. J Gastroenterol. 2009;44 Suppl 19:23-29.
44. Oliver SR, Phillips NA, Novosad VL, Bakos MP, Talbert EE, Clanton TL. Hy-
perthermia induces injury to the intestinal mucosa in the mouse: evidence
for an oxidative stress mechanism. Am J Physiol Regul Integr Comp Physiol.
2012;302(7):R845-853.
45. Flores-Salamanca R, Aragon-Vargas LF. Postexercise rehydration with beer
impairs fluid retention, reaction time, and balance. Appl Physiol Nutr Me-
tab. 2014;39(10):1175-1181.
46. Smith MP, Mendez J, Druckenmiller M, Kris-Etherton PM. Exercise intensity,
dietary intake, and high-density lipoprotein cholesterol in young female
competitive swimmers. Am J Clin Nutr. 1982;36(2):251-255.
47. Hobson RM, Maughan RJ. Hydration status and the diuretic action of a
small dose of alcohol. Alcohol Alcohol. 2010;45(4):366-373.
48. Shirreffs SM, Maughan RJ. Restoration of fluid balance after exercise-in-
duced dehydration: effects of alcohol consumption. J Appl Physiol (1985).
1997;83(4):1152-1158.
49. Eggleton MG. The diuretic action of alcohol in man. J Physiol. 1942;101(2):172-
191.
50. Crow LT. Diencephalic Influence in Alcohol Diuresis. Physiol Behav.
1968;3(2):319-&.
51. Lopez RM, Casa DJ, McDermott BP, Ganio MS, Armstrong LE, Maresh CM.
Does creatine supplementation hinder exercise heat tolerance or hydration
status? A systematic review with meta-analyses. J Athl Train. 2009;44(2):215-
223.
52. Sobolewski EJ, Thompson BJ, Smith AE, Ryan ED. The Physiological Effects
of Creatine Supplementation on Hydration: A Review. American Journal of
Lifestyle Medicine. 2011;5(4):320-327.
53. Dalbo VJ, Roberts MD, Stout JR, Kerksick CM. Putting to rest the myth of
creatine supplementation leading to muscle cramps and dehydration. Br J
Sports Med. 2008;42(7):567-573.
54. Powers ME, Arnold BL, Weltman AL, et al. Creatine Supplementation In-
creases Total Body Water Without Altering Fluid Distribution. J Athl Train.
2003;38(1):44-50.
55. Ziegenfuss TN, Lowery LM, Lemon PWR. Acute Fluid Volume Changes in
Men During Three Days of Creatine Supplementation. Journal of Exercise
Physiology Online. 1998;1(3).
56. Mendel RW, Blegen M, Cheatham C, Antonio J, Ziegenfuss T. Effects of crea-
tine on thermoregulatory responses while exercising in the heat. Nutrition.
2005;21(3):301-307.
57. Weiss BA, Powers ME. Creatine supplementation does not impair the ther-
moregulatory response during a bout of exercise in the heat. J Sports Med
Phys Fitness. 2006;46(4):555-563.
58. Rosene JM, Whitman SA, Fogarty TD. A Comparison of Thermoregulation

134
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

With Creatine Supplementation Between the Sexes in a Thermoneutral En-

vironment. J Athl Train. 2004;39(1):50-55.
59. Watson G, Casa DJ, Fiala KA, et al. Creatine use and exercise heat tolerance
in dehydrated men. J Athl Train. 2006;41(1):18-29.
60. Kern M, Podewils LJ, Vukovich M, Buono MJ. Physiological Response to Ex-
ercise in the Heat Following Creatine Supplementation. Journal of Exercise
Physiology Online. 2001;4(2):18-27.
61. Kilduff LP, Georgiades E, James N, et al. The effects of creatine supplemen-
tation on cardiovascular, metabolic, and thermoregulatory responses dur-
ing exercise in the heat in endurance-trained humans. Int J Sport Nutr Ex-
erc Metab. 2004;14(4):443-460.
62. Armstrong LE. Caffeine, body fluid-electrolyte balance, and exercise per-
formance. Int J Sport Nutr Exerc Metab. 2002;12(2):189-206.
63. Armstrong LE, Casa DJ, Maresh CM, Ganio MS. Caffeine, fluid-electrolyte
balance, temperature regulation, and exercise-heat tolerance. Exerc Sport
Sci Rev. 2007;35(3):135-140.
64. Falk B, Burstein R, Rosenblum J, Shapiro Y, Zylber-Katz E, Bashan N. Effects
of caffeine ingestion on body fluid balance and thermoregulation during
exercise. Can J Physiol Pharmacol. 1990;68(7):889-892.
65. Del Coso J, Estevez E, Mora-Rodriguez R. Caffeine effects on short-term
performance during prolonged exercise in the heat. Med Sci Sports Exerc.
2008;40(4):744-751.
66. Del Coso J, Estevez E, Mora-Rodriguez R. Caffeine during exercise in the
heat: thermoregulation and fluid-electrolyte balance. Med Sci Sports Ex-
erc. 2009;41(1):164-173.
67. Roti MW, Casa DJ, Pumerantz AC, et al. Thermoregulatory responses to ex-
ercise in the heat: chronic caffeine intake has no effect. Aviat Space Environ
Med. 2006;77(2):124-129.
68. Dias JC, Roti MW, Pumerantz AC, et al. Rehydration after exercise dehydra-
tion in heat: Effects of caffeine intake. J Sport Rehabil. 2005;14(4):294-300.
69. Shaw G, Koivisto A, Gerrard D, Burke LM. Nutrition considerations for open-
water swimming. Int J Sport Nutr Exerc Metab. 2014;24(4):373-381.
70. Shirreffs SM, Sawka MN. Fluid and electrolyte needs for training, competi-
tion, and recovery. J Sports Sci. 2011; 29 Suppl 1:S39-46
71. Beltrami FG, Hew-Butler T, Noakes TD. Drinking policies and exercise-
associated hyponatraemia: is anyone still promoting overdrinking? Br
J Sports Med. 2008; 42(10):796-501. Review. Erratum in: Br J Sports
Med. 2009; 43(4):310-1.
72. Wagner S, Knechtle B, Knechtle P, Rüst CA, Rosemann T. Higher prevalence
of exercise-associated hyponatremia in female than in male open-water
ultra-endurance swimmers: the ‘Marathon-Swim’ in Lake Zurich. Eur J Appl
Physiol. 2012; 112(3):1095-106.
73. Noakes TD, Sharwood K, Speedy D, Hew T, Reid S, Dugas J, Almond C,
Wharam P, Weschler L. Three independent biological mechanisms cause
exercise-associated hyponatremia: evidence from 2,135 weighed competi-
tive athletic performances. Proc Natl Acad Sci USA. 2005; 102:18550–55.

135
 sp or t science

Biomechanics of relay swimming starts: state

of the art
K. de Jesus1,2,3, K. de Jesus1,2,3, L. Mourão 1,2,4, L. S. Coelho5,6, L.
Machado1,2, R.J. Fernandes1,2, J.P. Vilas-Boas1,2

Introduction
In competitive swimming, starting techniques performed in competi-
tive events differ according to the received starting signal, with the audi-
tory and visual stimulus being used for individual and first swimmer of
relays, and for the second, third and fourth changeovers, respectively
([1]; Figure 1). In fact, the first swimmer should be stationary until the
starting signal is given (FINA SW 4.1 and 4.2 rules) and the second, third
and fourth swimmers should be in contact with the platform when their
incoming teammate touches the starting wall (FINA SW 2.6.5 rule) but
can move him/herself to leave the block in a more explosive way. Starts
performed in individual ventral and dorsal events have been extensively
studied in comparison to the relay starting techniques, probably due to
the greater number of official individual events [2] and the higher com-
plexity that relay experimental research protocols require (with two
swimmers simultaneously performing different tasks; [3]).

1. Centre of Research, Education, Innovation and Intervention in Sport, Faculty

of Sport, University of Porto, Portugal;
2. Porto Biomechanics Laboratory, University of Porto, Porto, Portugal;
3. Faculty of Physical Education and Physiotherapy, Federal University of Ama-
zonas, Amazonas, Manaus, Brazil;
4. Superior School of Industrial Studies and Management, Porto Polytechnic In-
stitute, Vila do Conde, Portugal;
5. Industrial and Systems Engineering Graduate Program, Pontifical Catholic
University of Paraná, Brazil
6. Electrical Engineering Graduate Program, Federal University of Paraná, Brazil

136
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Figure 1. Official swimming starting conditions and technical solutions most

commonly used (adapted from Vilas-Boas and Fernandes [1]).

Saavedra et al. [4] have pointed out that, from a range of 19 interna-
tional competitions held between 2000 to 2012, high proficient starters
achieved better performances in international swimming events (spe-
cially in 4x100 m race), evidencing the need for deeper research on the
topic. Moreover, although three out of 17 events in the Olympic swim-
ming programme are relays, coaches remain allocating only a small-
time percentage for the specific training of the relay starts prior to na-
tional and international events [2]. Furthermore, since 2013, new relay
races with mixed gender teams have been added to World Champion-
ship events (FINA SW 10.10 rule)
On the relay swimming events, FINA authorizes the second, third and
fourth swimmers to begin their starting motion before the incoming
swimmer has touched the wall ([5]), just requiring that one foot is in con-
tact with the starting block at the time the former swimmer finishes his/
her course (the relay exchange timing tolerance is -0.03 s; [4]). To reduce
the potential for a false start, leading to an automatic disqualification,
some swimmers prefer to use the conventional non-step start technique,
involving a circular backswing upper limb movement [2,3,6]. Here, the
risk of going out before the colleague touches the wall is much lower, but
the impulse of the upper limbs is increased comparing to the grab start
[7]. In the past two decades, new step starting techniques with one or
two steps approach before driving off the platform (Figure 2) have been
observed during international competitions and later examined from a
scientific point of view [3,6]. Findings indicated that the step starting in-

137
 sp or t science

volves longer exchange block times than non-step techniques [6] and
the effective use of the double-step motion depends on the swimmer’s
ability to take longer steps [3].

Figure 2. Relay start techniques: One and two steps (left and right panels, re-
spectively).

The importance of the swimming starts for enabling better perfor-

mance on relay events evidences the need to synthesise the scientific
knowledge regarding biomechanical behaviour of relay start techniques.
Literature review is available regarding ventral [8] and dorsal (back-
stroke events) start techniques [9], but no survey has been conducted
for a comprehensive knowledge of the relay start techniques literature.
The aim of this chapter is to review and discuss the state of the art re-
garding relay start biomechanics, using peer-reviewed journal articles
and relevant congresses and symposium series (as the Biomechanics
and Medicine in Swimming Symposiums and the International Society
of Biomechanics in Sports Conferences).

Development
Search strategy
The literature search was performed using PubMed and Scopus elec-
tronic databases, and only for English written documents published
before March 2017. The reference lists of the articles were also used to
ensure, as far as practically possible, that all appropriate studies were
considered for inclusion. In addition, it was also analysed the Proceed-
ings of the Biomechanics and Medicine in Swimming (BMS) Symposiums

138
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

and of the International Society of Biomechanics in Sports (ISBS) Con-

gresses from 1980 to 2016. Key words including “swimming”, “relay” and
“start” were used to locate documents.

Inclusion and exclusion criteria

Included studies were experimental biomechanical approaches with
able-bodied swimmers. Documents available only as abstracts and du-
plicated studies from original investigations were excluded.

Results and Discussion

Fifteen studies were obtained from the preliminary search, but only
six studies have met the inclusion criteria, being one from a swimming
specific journal, four from peer-review journals and another from the
Proceedings book of the BMS conference (Table 1).

Author (s) Main aim Sample Setting

Gambrel To identify the mechanical Seven college Experimental
et al. [18] characteristics of the step male swimmers
start and to compare it with a
conventional relay start
McLean et To compare the three step starts Ten collegiate Experimental
al. [3] (with a full step length) and the male swimmers
non-step start. To evaluate the
restriction of step length on
performance
Takeda et To evaluate the effectiveness of the Eight well-trained Experimental
al. [6] no-step, single-step and double- male college
step start techniques to determine swimmers
the relay start performance
Saavedra To analyse the association 827 relay race Competition
et al. [4] between relay exchange block records from
time and final performance in Olympic, World,
relay events in international European,
championships Commonwealth
and Pan Pacific
Games swimmers
Skorski et To investigate if swimming 166 elite male Competition
al. [5] performance is better in a relay swimmers
race than in the corresponding
individual race.
Fischer et To compare the offensive Twelve male and Experimental
al. [2] (minimizing change-over time) twelve female
and conservative (maximizing national level
horizontal peak force) strategies swimmers
using the non-step start technique

139
 sp or t science

Table 1. Descriptive analysis of the six included studies regarding swimming

starting relays, displaying the authors, main goal, swimmers’ sample profi-
ciency and data collection setting.

Table 1 reveals that three experimental studies analysed the effects

of non-step and step swimming starting techniques [3,6,13], and anoth-
er compared two intervention strategies considered feasible to be im-
plemented one week ahead of competition [2]. Comparisons between
non-step and step starts have revealed different results for performance,
however McLean et al. [3] and Takeda’s et al. [6] findings have corrobo-
rated about the importance of proper step length. Fischer et al. [2] no-
ticed that while minimizing changeover time may eventually enforce the
risk of a false start, keeping it within a safe range may prevent the relay
team from disqualification. Researchers have also focused at the influ-
ence of the changeover time on the final relay race performance at high
calibre swimming events as a function of sex and classification (medal-
lists, first to third, and non-medallists, fourth to eight) and at the confir-
mation if starting at relay events provide better performance than indi-
vidual ones ([4,5]; respectively). Despite being a small amount of studies
performed in relay techniques, they have raised and answered relevant
coach questions before selecting a start position that can result in better
performance [10].
The studies conducted in the laboratory setting adopted a small
number of swimmers, which undermines the effect size and results
reproducibility [11]. However, ~ 10 swimmers have been considered a
common sample size number in swimming start studies due to the avail-
ability for familiarization and testing protocols using complex data col-
lection methodology. From the four studies using the experimental set-
ting, three implemented a relay start techniques familiarisation and one
refused this period to analyse the effects of the feet position usage.
A small variation in experimental design has been noticed, being four
studies conducted in experimental and two in competition setting. It has
been observed a considerable number of sophisticated and useful so-
lutions for relay swimming start analysis (i.e. video cameras and force
plates). However, these methods must cover the scientific demands of
validity, reliability and accuracy along with the more practical issues like
range of usage, complexity and costs [12]
The data collection setting determines the biomechanical variables
possible to be assessed and their study relevance. Data collected in the
experimental setting are more accurate and reliable, but the validity can
be substantially restricted. The opposite is observed in the field, since

140
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

variables can be defined with high validity, but restricted accuracy and
reliability [12] Two biomechanical branches were predominantly used
for data recordings, kinematics (video camera images) and kinetics
(force-plates). The pioneer study considering relay start techniques used
a high-speed cinematography camera (LoCAm, model 51, 16 mm) to as-
sess block, flight and 10 m time, horizontal and vertical centre of mass
position, centre of mass take-off angle, height and velocity and vertical
centre of mass position at water entry [13]. The effectiveness of dive
starts has been measured by the time to a set distance, ranging from 1.52
to 25 m (cf. [14]). The start performance is commonly measured between
the start signal and the moment when the swimmer’s head reaches the
15th meter [15]. The 10, 7.5 and 5 m mark has been pointed out as ap-
propriate to assess start performance because the effects of other swim
variables are avoided, being only the block, flight, entry and glide time
included in most of these distances (e.g. [3]). The choice of performance
measure might affect the conclusions drawn from a technique analysis
[16], thus researchers should define a common variable to quantity per-
formance in relay starts.
The changeover time measured at experimental and competition
setting (i.e. from the incoming swimmer wall contact and the outgoing
swimmer’s toe-off; [4,5]) has been considered the most important varia-
ble for shorter 15 m start time [4]. The perfect exchange time is achieved
when the toes of the outgoing swimmer leave the starting platform while
the incoming swimmer touches the wall [5]. Researches on relay starts,
so far, have implicitly assumed that minimizing the changeover time
would be the best strategy for a relay start prior to the race [2]. These au-
thors evidenced that improvements in the relay start time (i.e. between
the wall contact of the incoming swimmer and the head passage of the
relay starter at 7.5 m) are, for the most part, related to decreases in the
changeover time. In addition, they found that feedback on the horizon-
tal peak force provided superior effects on the reaction start time when
compared with changeover information. For them, independent of the
start technique, differences in the attentional focus (i.e. changeover time
or horizontal peak force) during the relay start may influence relay start
time.
Temporal data has been presented as the most common variables
measured in relay start techniques, however, following Gambrel and
co-authors [13] the most recent studies have used deterministic factors
identified from an individual swimming start model [18] from the take-
off until swimmers’ immersion. Authors have adopted distinct methodo-

141
 sp or t science

logical procedures to assess the take-off angle and horizontal velocity

using cinematographic, digital video cameras and ground reaction force
measurements [13, 2,6]. In addition, McLean et al. [3] and Takeda et al. [6]
assessed the vertical velocity, and ground reaction forces, respectively,
which allows an overall understanding about the swimmers’ locomotion
over the block and three-dimensional forces generated before the take-
off instant. The variables measured during swimmer block contact as the
take-off angle, velocities and external forces are considered flight time
influencers [17]. In fact, if a limb is in contact to a force platform, the re-
spective three-dimensional force and momentum should be measured
to maximise training efficiency [18].
The effects of the step start were firstly assessed by Gambrel et al. [13]
who verified no differences when compared to the conventional no-step
start. Despite these findings, authors pointed out that the flight distance
and time to 10 m could be covered longer and faster using the step start.
McLean and co-authors [3] also found that time to 10 m was not short-
er when compared step and no-step start, although some mechanical
advantages were observed. The double and single step start increased
horizontal take-off velocity and take-off height, respectively, and the
double, single step apart and single step together increased the take-off
vertical velocity. Authors included the analysis of restricting step length
by 50% and noticed little effect on step starts with exception of horizon-
tal and vertical take-off velocity, reduced and increased, respectively.
Data suggested that step starts offered performance improvements, but
those were not widespread and were dependent on the ability to take
longer steps.
Contrarily to McLean et al. [3], further comparison between non-step,
single-step and double step start revealed similar horizontal take-off
velocity, although relay time (i.e. from the swimmer’s contact until the
take-off) decreased in the order no-step, single and double-step start [6],
suggesting the former as better for relay starts performance. According
to these authors the familiarisation period might not have been enough
since swimmers have incorrectly performed the foot placement during
step starts.
The most recent relay starts studies formulated three different re-
search questions to characterise available relay start techniques. Saave-
dra and co-authors [4] analysing the association between relay exchange
block time and swimming final performance found that the former was
especially relevant for women’s relay medallists in the 4x100 freestyle
and medley. Authors considered that the exchange block time should be

142
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

considered as one of the relay starts performance variables, being need-

ed to be included as part of training. In opposition to Saavedra et al. [4],
horizontal peak force, as a conservative relay strategy, showed a clear
advantage over the changeover time strategy to reduce relay time [2].
Skorski et al. [5] studied the individual swimming performance re-
lated to the corresponding individual race and verified that during re-
lays, highly trained swimmers competing in 1st position did not show any
different compared with their corresponding individual performance.
However, swimmers competing in second to fourth relay team position
demonstrated faster times in the 100 m and first half of the 200 m relays
than in their individual events, which was nullified when finishing times
were adjusted for the flying start.

Future suggestions
The presented studies evidenced important results for biomechanics,
coaches and swimmers, but further research questions should be raised.
Firstly, researchers should consider the effects of the current starting
block configuration on biomechanical variables when swimmers per-
form the step starts. Secondly, an electromyographical approach should
be implemented for better understanding about the muscular activa-
tion sequence and intensity. Thirdly, linear and non-linear mathemati-
cal methods should be built using traditional biomechanical variables to
model and predict relay start performance.

Conclusions
This chapter have exhibited the research background in relay start
techniques, summing up six included studies following the inclusion cri-
teria. Biomechanics, coaches and swimmers have been provided with
some objective evidence about kinematics and kinetics data of four
relay start techniques, no-step, one-step apart, one-step together and
two-step. Further methodological advances can be made for an exten-
sive comprehension of biomechanics during relay start techniques con-
sidering the back plate and respective effects, the electromyographic
analysis and mathematical prediction models.

Acknowledgments
Coordination for the Improvement of Higher Education Personnel
and Foundation for Science and Technology 039-2014 (grant number
99999.008578/2014-01)

143
 sp or t science

References
1. Vilas-Boas, J.P., Fernandes, R.J. (2003). Swimming starts and turns: deter-
minant factors of swimming performance. Actes des 3èmes journées spé-
cialisées de natation. Université Lille, pp. 84–95
2. Fisher, S., Braun, C., Kibele, A. (2017). Learning relay start strategies in
swimming: what feedback is best? European Journal of Sport Science,
17(3), 257-263.
3. McLean, S.P., Holthe, M.J., Vint, P.F., Beckett, K.D., Hinrichs, R.N. (2010).
Additional of an approach to a swimming relay start. Journal of Applied
Biomechanics, 16, 342-355.
4. Saavedra, J.M., García-Hermoso, A., Escalante, Y., Domingues, A.M., Arella-
no, R., Navarro, F. (2014). Relationships between exchange block time in
swim starts and final performance in relay races in international champi-
onships. Journal of Sports Sciences, 32(19), 1783-1789.
5. Skorski, S., Etxebarria, N., Thompson, K.G. (2016). Breaking the myth that
relay swimming is faster than individual swimming. International Journal
of Sports Physiology and Performance, 11, 410-413.
6. Takeda, T., Takagi, H., Tsubakimoto, S. (2010). Comparison among three
relay starts in competitive swimming. In: P-L Kjendlie, R. K. Stallman, J.
Cabri (eds), XIth International Symposium on Biomechanics and Medicine
in Swimming, pp. 170-172.
7. Ayalon, A., Van Gheluwe, B., Kanitz, M. (1975). A comparison of four styles
of racing starts in swimming. In. L. Lewillie and J.P. Clarys (Eds.). Swim-
ming II (pp 233-240). Baltimore University Park Press.
8. Vantorre, J., Chollet, D., Seifert, L. (2014). Biomechanical analysis of the
swim-start: a review. Journal of Sports Science and Medicine, 13, 223-231.
9. De Jesus, K., de Jesus, K., Fernandes, R.J., Vilas-Boas, J.P., Sanders, R.
(2014). The backstroke swimming start: state of the art. Journal of Human
Kinetics, 10(42), 27-40.
10. Seifert, L., Vantorre, J., Lemaitre, F., Chollet, D., Toussaint, H.M., Vilas-
Boas, J.P. (2010). Different profiels of the aerial start phase in fornt crawl.
Journal of Strength and Conditioning Research, 24(2), 507-516.
11. Button, K. S., Ioannidis, J.P.A., Mokrysz, C., Nosek, B.A., Flint, J., Robinson,
E.S. (2013). Power failure: why small sample undermines the reliability of
neuroscience. Nature reviews. Neuroscience, 14, 365, 376.
12. Schwameder, H. (2008). Aspects and challenges of applied sport biome-
chanics research. In: Y-H. Kwon, J. Shim, J.K. Shim, I-S. Shin (eds.), XXVI
International Symposium on Biomechanics in Sports. Seoul National Uni-
versity, pp 25-28.
13. Gambrel, D.W., Blanke, D., Mellion, M.B. (1991). A biomechanical compari-
son of two relay starts in swimming. Journal of Swimming Research, 2, 5-9.
14. Galbraith, H., Scurr, J., Hencken, C., Wood, L., Graham-Smith, P. (2008). Bi-
omechanical comparison of the track start and the modified one-handed
track start in competitive swimming: an intervention study. Journal of Ap-
plied Biomechanics, 24, 307-315.
15. Elipot, M., Dietrich, G., Hellard, P., Houel, N. (2010). High level swimmers’
kinetic effciency during the underwater phase of a grab start. Journal of

144
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Applied Biomechanics, 26, 501-507.

16. Bezodis, N.E., Salo, A.I., Trewartha, G. Choice of Sprint start performance
measure affects the performance-based ranking withing a group of sprint-
ers: which is the most appropriate measure? Sports Biomechanics, 9(4),
258-269.
17. Guimarães, A.C.S., Hay, J.G. (1985). A mechanical analysis of the grab start-
ing technique in swimming. International Journal of Sport Biomechanics,
1, 25-35.
18. Mourão, L., de Jesus, K., Viriato, N., Fernandes, R.J., Vilas-Boas, J.P., Vaz,
M.A.P. (2016). Design and construction of a 3D force plate prototype for
developing an instrumented swimming start block. Journal of Biomedical
Engineering and Informatics, 2(2), 17-22.

145
 sp or t science

Competition analysis in swimming: inter-

and intra-individual variability of speed and
stroking parameters management
David Simbaña-Escobar1,2, Philippe Hellard1,3, Marta Avalos4,5,
Ludovic Seifert2

Introduction
To achieve performance goals in competitive swimming there is a
need to strike a delicate balance between movement pattern stability
and variability because, although swimmers need to achieve consistent
outcomes, they also need to be able to successfully adapt their move-
ments to changes in the performance environment (Seifert, Komar,
Barbosa, Toussaint, Millet, et al., 2014a). To achieve these aims, the eco-
logical dynamics framework advocates that there is an intertwined rela-
tionship between the specific intentions, perceptions and actions of in-
dividual athletes which constrains this relationship between movement
pattern stability and variability in each individual performer (Seifert &
Davids, 2012). This inter-twined relation between an individual’s inten-
tions, perception and action processes needs to be carefully understood
because of the insights it provides on expert performance in swimming.
Traditionally, a high-level of expertise in sport has been associated
with the capacity to be able to reproduce a specific movement pattern
consistently and to reduce attention demands during performance by
increasing the automaticity of movement (Schmidt & Lee, 2011). It was
assumed that the central nervous system functioned as an executive or-
ganizer and prescriber of motor programs and action plans charged with
the task of producing stable movement patterns from an individual’s ef-
fector system (Schmidt & Lee, 2011; Summers & Anson, 2009). From that

1. Research Department, French Federation of Swimming, Pantin, France

2. Centre d’Etudes des Transformations des Activités Physiques et Sportives,
EA-3832, Faculty of Sport Sciences, University of Rouen Normandy, Rouen,
France
3. Institut de Recherche Biomédicale et d’Epidémiologie du Sport, National In-
stitute of Sport, Expertise and Performance (INSEP), Paris, France
4. INRIA SISTM, Team, F-33405 Talence, France
5. University of Bordeaux, INSERM, Bordeaux Population Health Research Cent-
er UMR 1219, F-33000, Bordeaux, France

146
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

viewpoint, expertise in sport was associated with a reduction in devia-

tions in task performance from an ideal standard or movement template,
which was represented in the central nervous system. By harnessing in-
tegrated feedback systems, swimmers were considered to modify the
motor program entry parameters until expert behaviour was eventually
achieved after many hours of practice (Schmidt & Lee, 2011). Tradition-
ally, therefore, movement variability was considered as noise in perfor-
mance and learning which should be minimized or eradicated to enable
the production of highly functional movement programs (i.e. an artefact
limiting an individual system’s processing of information from input to
output)(Davids, Bennett, & Newell, 2006; Newell & Corcos, 1993).
However, research in ecological dynamics has shown that move-
ment system variability should not necessarily be construed as noise,
detrimental to performance (Davids et al., 2006). Nor should it always
be viewed as error, or a deviation from a putative expert model, which
should be constantly corrected in swimmers. Inspired by Bernstein’s
work research program (Bernstein, 1996), movement system variabil-
ity instead is now considered to exemplify the functional flexibility of
a skilled athlete to respond to changes in dynamic performance con-
straints (Seifert, Button, & Davids, 2013; Seifert, Komar, Barbosa, Tous-
saint, Millet, et al., 2014a). A key idea is that movement pattern vari-
ability can be viewed as a functional property of skilled swimmers to
help them adapt their movement behaviours to changing task and en-
vironmental constraints (such as glide, swimming speed, effect of start,
turn, finish, fatigue (Hellard et al., 2008; Seifert, Komar, Barbosa, Tous-
saint, Millet, et al., 2014a; Seifert, Komar, Araújo, & Davids, 2016; Seifert,
Komar, Crettenand, & Millet, 2014b). Given these theoretical advances
in understanding behavioural variability, research within the ecologi-
cal dynamics framework have argued that there is no ideal behavioural
and race management solution that all swimmers should aspire during
training and competition. Rather, functional behaviour emerged during
training and race from the interaction of constraints on each individual
swimmer (task, environmental and organismic; Newell, 1986); leading
to intra-individual and inter-individual swimming speed variability as
consistent performance outcomes are achieved. For instance, in 400-m
international swimming competition, Mytton, Archer, Turner, Skorski,
Renfree, Thompson, and St Clair Gibson, (2015a) emphasized that the
variance of speed profiles tended to increase as the race progressed.
Similar results were also observed by Robertson, Pyne, Hopkins, and An-
son, (2009) when analysing between-swimmers average and standard

147
 sp or t science

deviation times for the 200-m freestyle for males and the 400-m freestyle
for both gender. Mytton, Archer, St Clair Gibson and Thompson, (2014),
Mytton, Archer, Turner, Skorski, Renfree, Thompson, and St Clair Gibson,
(2015a) related this higher variance of swimming speed in the final lap
to an indication of fatigue and deterioration in mechanical efficiency as
the race progresses. In addition, Mytton et al., (2014) recently showed
that performance in the final lap in 400-m (collected in international
competitions between 2005 and 2011), and especially the differences
in absolute, normalized, and relative speed can differentiate between
medallists and non-medallists. These authors suggested that the suc-
cess associated with a more pronounced end spurt could mean that
medallists were able to call on reserves of energy not available to non-
medallists three-quarters of the way through the race. These studies as
well as other recent research on speed management in cyclic activities
(e.g., running, swimming, cycling, skating, kayaking, etc) showed that
pacing is a key element of performance and more importantly of win-
ning the race (Foster et al., 1993, 2003, 2004, 2009; Thompson, Haljand, &
MacLaren, 2000; Thompson, MacLaren, Lees, & Atkinson, 2004; Tucker &
Lambert, 2006). However, these researches only consider lap times (i.e.
average value of speed for 50-m lap) and even 100 m times for the 400-m
freestyle event analysed by Mytton et al., (2014), Mytton, Archer, Turner,
Skorski, Renfree, Thompson, & St Clair Gibson, (2015b) and Thompson
et al. (2015), neglecting the possible effect of turning in the pacing. The
main limitations of these studies might be the impossibility to accurately
determine the inter-individual variability of race management.
Our study attempted to overcome this limitation by considering cy-
cle-to-cycle variability to draw individual profile because recently, the
functional role of behavioural variability has been supported by neuro-
scientific research highlighting the property of neurobiological system
degeneracy. Degeneracy as nothing to do this the degradation of a sys-
tem but could be defined by several authors from cognitive anatomy and
theoretical neurobiology (Edelman & Gally, 2001; Price & Friston, 2002;
Tononi, Sporns, & Edelman, 1999) as the capacity of system components
that differ in structure to achieve the same function or performance
output. This structural property in humans indicates the availability of
an abundance of motor system degrees of freedom, which can take on
different roles when assembling functional actions during sport perfor-
mance. Research in ecological dynamics has demonstrated that degen-
eracy in complex perception-action systems provides the neurobiologi-
cal basis for diversity of actions required to negotiate information-rich

148
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

and dynamic environments for task goal attainment (Komar, Chow,

Chollet, & Seifert, 2015; Rein, Davids, & Button, 2010; Seifert, Wattebled,
Herault, Poizat, Adé, et al., 2014c; Seifert, Komar, Crettenand, & Millet,
2014b). These studies have shown that, more than simply ensuring sta-
bility against perturbations and adaptations to dynamic performance
environments, the degenerate architecture of neurobiological systems
can help individuals exhibit adaptability and creativity. Based on these
ideas on neurobiological degeneracy, research on sport performance
optimisation has begun to examine and to explain why expert perform-
ers often display high levels of intra- individual and inter-individual be-
havioural variability in sport, data traditionally viewed as counter-intui-
tive. The behavioural variability exhibited by skilled swimmers can play
a functional role; for instance, it highlights an expert swimmer’s capacity
to perform several types of movement and/or to adopt one of a num-
ber of co-existing modes of coordination in order to achieve the same
functional performance outcomes (swimming against waves or current,
approaching the wall, preparing the turn, overcoming drag)(Guignard et
al., 2017; Seifert, Komar, Barbosa, Toussaint, Millet, et al., 2014a). In the
past years, empirical research on sport performance has clearly exempli-
fied how intra-individual and inter-individual behavioural variability can
play a functional role in swimming performance; however, these studies
were limited to training and learning contexts (for a review, see Seifert,
Komar, Barbosa, Toussaint, Millet, et al., 2014a). Therefore, the aim of
our study was to consider cycle-to-cycle data set to investigate swim-
ming speed (S) management and its subsequent stroking parameters
(SR, stroke rate; SL, stroke length). More precisely, our study investigates
the inter- and intra-individual variability of swimming speed (and its
subsequent SR and SL) in top-level swimmers of a 50, 100 and 200-m
freestyle National and World Championships, to determine individual
profile of race management by using latent class mixed model, which is
an innovative method coming from data mining.

Material and Methods

The performances of 32 top-level swimmers were video-analysed
during the 50, 100 and 200-m freestyle finals of the Fédération Interna-
tionale de Natation Amateur (FINA) World Championships (Kazan, Rus-
sia, 2015) (8 males, 8 females) and the French National Championships
(Montpellier, France 2016) (8 males, 8 females). The procedures (data col-
lection and analysis) were approved by the French Federation of Swim-
ming, FINA and the local University ethics committee, and conformed to

149
 sp or t science

the declaration of Helsinki. The procedures were explained to the swim-

mers, who then gave their written informed consent to participate.

The 50, 100 and 200-m freestyle events were videoed with a Black
Magic 4K camera (25 Hz, Ultra HD, 3840Í2160 pixels â Black Magic de-
sign, Melbourne, Australia) equipped with a 10-mm focal lens (f/4.5-5.6 IS
STM) (âCanon Inc, Ôta, Tokyo, Japan) to visualize the entire pool (50 x 25
m). The camera was focused on the middle of the pool (25 m) and located
in the top row of spectator seats approximately 25 m above and 15 m away
from the side of the pool.
Race analysis software compiled in Matlab 2012 (The MathWorks,
Inc., Natick, MA, USA) was used for calibration and image processing to
obtain stroke rate (SR), stroke length (SL) and swim speed (S) for every
stroke cycle. Four poolside marks in the swimming lane were chosen to
calibrate the pool using 2D direct linear transformation algorithms (Ab-
del-Aaziz & Karara, 1971; Benarab, Napoléon, Alfalou, & Verney, 2017).
The operator for each event made this calibration and then manually
digitalized the head position at the beginning of each stroke cycle (right
hand entry). The stroking phase was defined as the first and last right-
hand entry after the start or underwater phase and before the turn in
each lap. To quantify the reproducibility of the tracking procedure, the
same race and same swimmer was digitalized ten times by eight opera-
tors. The root mean square (RMS) was calculated to estimate the track-
ing measure reproducibility (RMS = 0.0121 ± 0.0084 m).
The longitudinal analysis of speed was conducted on 32 swimmers
for each 50-m lap in all races (50, 100 and 200-m freestyle). The speed
dataset of each participant was normalized by the average speed of
the race for all races events (to minimize the effect of gender and skill
level, i.e. all males and females national and international swimmers
were included). The latent class mixed model (LCMM) was used to iden-
tify trajectory classes of normalized speed over the different races. This
model seeks potential latent profiles in heterogeneous populations. It
combines a latent class model to identify homogenous latent classes
of participants and a mixed model to describe the mean trajectory over
the swimming race in each latent group, while taking account the indi-
vidual correlation between repeated measures (Carrière et al., 2016).
The trajectories of normalized speed were described using cubic spline
functions without adjustment for baseline covariates. Cubic splines are
piece-wise cubic polynomials smoothly assembled in the knot points
(frequently, knots are located at suitably chosen quantiles). Not only

150
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

pieces must connect but their slopes must match to ensure the curve
is smooth. The shape of the class-specific and subject-specific trajecto-
ries were determined by comparing models with increasing number of
model parameters (one to four knots in the spline functions, a diagonal
or unstructured random-effect covariance matrix and a class-specific or
proportional random-effect covariance matrix). The best model among
models with the same number of classes (one- to four- latent classes)
was selected using the Bayesian Information Criterion (BIC). The opti-
mal number of latent class was determined by the BIC combined with
the interpretability of the distinct classes. The longitudinal analysis al-
lowed us to investigate the evolution of normalized speed in each latent
class (decreased, stabilized or increased) for each lap. The classes size
does not need to be identical. Finally, latent classes were characterized
in terms of covariates (specialty, sex, distance race). The function HLME
of the LCMM R-package version 1.7.6 was used to estimate the model
parameters (Proust-Lima et al., 2015).

Results
The main findings of our study showed that the spline with three de-
grees of freedom was the best model to the normalized speed over the
distance. This spline is composed by three pieces in each lap. Roughly,
coefficients associated to spline components give us information about
the first part of the race after the breakout (first movement after the un-
derwater phases), the middle of the race and the turn in (approach to
the wall before the flip turn) or to the end of the race (when the swimmer
touches the wall).
Based on BIC and interpretability of the normalized values of speed
(normalization is done by dividing absolute speed by the average speed
on the race), two classes appeared for the 50-m race, then two classes for
the first lap (L1) and one class for the second lap (L2) appeared for the
100 m race. Finally, two classes for L1 and L2 and three classes for L3 and
L4 appeared for the 200-m race. When the raw values of speed were con-
sidered in respect to gender, two classes for male and four classes for fe-
male appeared in the 50-m race; three classes appeared for both gender
in the 100 m race and four classes appeared for both gender in the 200-m
race. These results (i.e., both the normalized and raw values of speed)
suggested that longer was the race, higher was the number of classes i.e.
higher was the variability of race management profiles, probably due to
a strong effect of the turn-in and –out. This was well emphasized by the
cubic spline model as the dominant class occurrence.

151
 sp or t science

For the 50-m freestyle, the figure 1 showed two classes for the nor-
malized values of speed: C1 with 25 swimmers (78.1% of the sample) and
C2 with 7 swimmers (21.9% of the sample). The gender distribution in-
side each cluster was: 15 male and 10 female swimmers. Male swimmers
have an average speed of 2.11 ± 0.04 m.s-1 with a CV S of 5.43 ± 0.65 %
while females have a lower average speed 1.89 ± 0.04 m.s-1 and a CV S of
4.97 ± 0.76 %. Concerning C2, only 1 male and 6 female swimmers com-
posed this cluster. The average absolute speed of the male was 2.13 m.s-1
with a CV S of 2.84 % and the average speed for the female swimmers
was 1.88 ± 0.03 m.s-1 with a CV S of 3.28 ± 0.62 %. Based on the param-
eters of the model to cluster the swimmers, both C1 and C2 exhibited a
quite stable speed after the dive, then a decrease of speed until the end
of the race. As C2 started with a lower speed, its decrease of speed ap-
peared lower than for C1 (Fig. 1).

Figure 1. Normalized values of speed throw the 50-m freestyle. The black line
corresponds to class 1 and the grey line to class 2, while the dashed lines
correspond to their respective confidence interval (CI).

For the 100 m freestyle, the first lap (L1) was composed by two classes
(Fig. 2 left panel): C1 represented 62.5% and C2 37.5% of the sample. The
gender distribution inside each cluster was: 15 male and 5 female swim-
mers in C1. Male swimmers have an average speed of 2.05 ± 0.02 m.s-1
with a CV S of 4.05 ± 0.53 %, while females have lower average speed
1.86 ± 0.03 m.s-1 and a CV S of 3.69 ± 0.88 %. Concerning C2, only 1 male

152
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

and 11 female swimmers composed this cluster. The average absolute

speed for the male was 2.05 m.s-1 with a CV S of 3.14 % while the average
speed for female swimmers was 1.81 ± 0.02 m.s-1 and a speed variation of
2.86 ± 0.60 %. At L1, swimmers from C1 swam faster after the start than
swimmers of C2, but these later exhibited a slightly greater speed at the
end of the lap when they approached the wall. C2 was characterized by
an increase of speed after the start, then a quite constant speed until the
middle of the lap and finally a decrease of speed.
At L2, all the swimmers were in the same class, showing a constant
speed until the middle of the lap, and then a decrease of speed occurred
(Fig. 2 right panel). The average speed was 1.89 ± 0.03 m.s-1 with CV S
equals to 4.91 ± 1.31 m.s-1 for the 16 male swimmers while the average
speed was 1.71 ± 0.03 m.s-1 with 4.88 ± 0.76 % of speed variation for fe-
males.

Figure 2. Normalized values of speed through the 100 m freestyle (left panel
for L1 and right panel for L2). The black line corresponds to class 1 and the
grey line to class 2, while the dashed lines correspond to their respective
confidence interval (CI).

The mean values of S and the technical parameters (SR, SL, SI) with
their standard deviation (s.d.) are summarized in Table 1 for the differ-
ent classes of each lap, gender and of each race event (50, 100, 200-m
freestyle). The Table 2 summarized the coefficient of variation (CV) for
all parameters.

153
 sp or t science

Race Lap Class Gender n S s.d. SR s.d. SL s.d.

50-m 1 M 15 2.11 0.04 60.95 1.70 2.08 0.04
1 F 10 1.89 0.04 58.86 4.33 1.94 0.14
2 M 1 2.13 60.18 2.13
2 F 6 1.88 0.03 58.12 2.40 1.94 0.08
100 m L1 1 M 15 2.05 0.03 52.46 2.04 2.35 0.07
1 F 5 1.86 0.03 50.99 2.97 2.20 0.16
2 M 1 2.06 52.23 2.36
2 F 11 1.82 0.03 51.31 2.39 2.13 0.11
L2 1 M 16 1.89 0.03 50.41 1.32 2.25 0.05
1 F 16 1.71 0.03 49.45 2.41 2.09 0.10
200-m L1 1 M 12 1.89 0.03 43.58 2.35 2.60 0.13
1 F 16 1.72 0.03 45.60 2.79 2.27 0.13
2 M 4 1.92 0.04 44.26 1.41 2.60 0.11
L2 1 M 12 1.80 0.02 42.16 2.28 2.58 0.14
1 F 15 1.65 0.02 43.52 3.06 2.28 0.15
2 M 4 1.82 0.02 42.70 0.78 2.57 0.02
2 F 1 1.68 40.38 2.49
L3 1 M 13 1.77 0.03 42.91 2.75 2.48 0.15
1 F 10 1.60 0.02 42.98 1.68 2.23 0.09
2 M 1 1.78 45.08 2.37
2 F 6 1.62 0.01 44.59 3.28 2.20 0.16
3 M 2 1.78 0.01 43.90 1.01 2.43 0.03
L4 1 M 8 1.73 0.05 44.91 2.16 2.32 0.13
1 F 15 1.59 0.04 45.16 2.37 2.11 0.09
2 F 1 1.60 43.57 2.21
3 M 8 1.74 0.05 46.67 2.15 2.25 0.11

Table 1. Average values of speed (S) and stroking parameters: stroke rate (SR),
stroke length (SL) of the different classes for each lap, each race event (50,
100, 200-m freestyle) and for each gender (male, M and female, F). n is the
number of swimmers in each class, and s.d. is the standard deviation.

154
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Race Lap Class Gender n CV S s.d. CV s.d. CV SL s.d.

SR
50-m 1 M 15 5.43 0.65 4.20 0.95 3.90 1.07
1 F 10 4.97 0.76 4.15 1.16 4.03 0.67
2 M 1 2.84 2.64 3.62
2 F 6 3.28 0.62 4.09 0.65 3.18 0.94
100 m L1 1 M 15 4.06 0.53 4.49 1.66 3.63 0.95
1 F 5 3.69 0.89 4.10 1.56 3.50 0.87
2 M 1 3.15 2.43 3.48
2 F 11 2.86 0.60 4.53 1.70 4.64 1.44
L2 1 M 16 4.91 1.31 3.47 0.95 4.64 1.15
1 F 16 4.88 0.76 3.69 1.00 5.68 1.80
200-m L1 1 M 12 3.68 0.83 3.95 1.40 3.86 1.05
1 F 16 3.48 0.80 4.61 1.33 3.98 0.97
2 M 4 6.55 1.22 4.73 0.50 4.62 0.99
L2 1 M 12 3.20 0.60 4.01 2.63 4.51 2.05
1 F 15 2.70 0.97 3.02 1.13 3.63 1.65
2 M 4 4.30 1.20 3.88 1.79 4.58 1.65
2 F 1 2.87 3.26 2.06
L3 1 M 13 3.62 1.14 3.34 1.40 4.35 1.08
1 F 10 3.00 0.73 3.11 1.35 3.75 1.17
2 M 1 2.54 2.20 3.07
2 F 6 2.41 0.58 2.64 1.14 2.98 0.84
3 M 2 6.49 0.76 4.07 2.36 7.09 0.57
L4 1 M 8 3.09 0.67 3.39 1.35 4.09 1.20
1 F 15 3.27 1.08 2.65 4.29 1.94
2 F 1 5.01 3.55 1.82 3.22
3 M 8 4.26 0.75 3.39 1.35 4.78 0.85

Table 2. Coefficient of variation (CV) of speed (S), stroke rate (SR) and stroke
length (SL) of the different classes for each lap, each race event (50, 100, 200-m
freestyle) and for each gender (male, M and female, F). n is the number of swim-
mers in each class, and s.d. is the standard deviation.

For the 200-m freestyle, two profiles of race management occurred

at L1 and L2, while three profiles of race management appeared at L3
and L4.
At L1, two classes were identified as follows: C1 was composed by
28 swimmers (87.5% of the sample) and C2 was composed by 4 swim-

155
 sp or t science

mers (12.5% of the sample). For C1, the speed decreased quite linearly
until approaching of the wall. This cluster C1 was composed by 12 males
with an average speed of 1.89 ± 0.03 m.s-1 and the coefficient of variation
of speed (3.63 ± 0.83 %) which were higher than those of the 16 female
swimmers (S: 1.72 ± 0.03 m.s-1 and CV S: 3.48 ± 0.80 %). Only 4 male swim-
mers composed C2 (S: 1.92 ± 0.04 m.s-1 and CV S: 6.55 ± 1.22 %). The
cluster C2 demonstrated a greater variability of speed and a faster start
than C1. However, the speed further decreased than C2 from the middle
to the end of the lap.

Figure 3. Normalized values of speed through the 200-m freestyle (upper left
panel for L1; upper right panel for L2; lower left panel for L3 and lower right
panel for L4). The black line corresponds to class 1, the grey line to class
2 and the light grey to class 3, while the dashed lines correspond to their
respective confidence interval (CI).

In the second lap (L2), the first class C1 was composed by 84.3 % of
the population. The 12 male swimmers have a greater average speed
and CV S than the 15 females (S: 1.80 ± 0.02 vs.1.65 ±0.02 m.s-1 exponent
and CV S: 3.20 ± 0.60 vs. 2.70 ± 0.97 % respectively) in the first class C1.
The same tendency was founded in C2 with a 4 male and 1 female. (S:
1.82 ± 0.02 vs. 1.68 m.s-1 exponent and 4.30 ± 1.20 vs. 2.87 % respec-
tively). The swimmers of C2 increased their speed after the turn and then
they decreased it from 75-m to the wall. The lower variation of speed in

156
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

C1, revealed a quite constant speed until 80-m.

At the third lap (L3), three classes of swimmers were observed. Twen-
ty swimmers (71.9 % of the sample) composed C1, including 13 males
and 10 females. Males have higher speed and CV S than females (S: 1.77
± 0.03 vs. 1.60 ±0.02 m.s-1 exponent and CV S: 3.62 ± 1.14 vs. 3 ±, 0.73
% respectively). The second class C2 exhibited 7 swimmers (21.8 % of
the sample) with only 1 male and 6 females. These later demonstrated a
lower speed and variation of speed (S: 1.78 vs. 1.62 ±0.01 m.s-1 exponent
and 2.54 vs. 2.41 ± 0.58 % respectively). Only 2 male swimmers (6.3 %)
composed the class C3 for whom S: 1.78 ±0.01 m.s-1 exponent and CV
S: 6.49 ± 0.76 %. In C1, the speed stayed quite constant and decreased
when the swimmer approached the wall. In C2, the speed stayed quite
constant through the lap. The two swimmers of C3 decreased their speed
after the turn, then their speed increased at the middle and decreased
again at the end of the lap. In summary, this third lap showed three dif-
ferent profiles of race management: C3 exhibited a fast speed after the
turn, a short increase and finally a great decrease of speed, which looks
like a “zig-zag” management close to cubic model; C2 demonstrated a
quite constant speed through the lap, reflecting a linear model; C1 exhib-
ited a drop of speed at the end of the lap, reflecting a parabolic profile.
At L4, three classes of swimmers also occurred with similar profiles
than observed at L3. The class C1 was composed by 23 swimmers (71.8
% of the sample) including 8 males with an average speed of 1.73 ± 0.05
m.s-1 and CV S of 3.09 ± 0.67 %, and 15 females with an average speed of
1.59 ± 0.04 m.s-1 exponent and a CV S of 3.27 ± 1.08 %. Those swimmers
kept their speed almost constant until 180-m, and then they slightly de-
creased their speed until the end of the race. The class C2 was composed
by only 1 female swimmer, who had an average speed of 1.60 m.s-1 ex-
ponent. Surprisingly she had a very high coefficient of variation of speed
(5.01 %) and increased her speed until the end of the race. Finally, C3
was composed by 8 male swimmers with an average speed of 1.74 ±0.05
m.s-1 exponent and a high CV S: 4.26 ± 0.75 %. Those males increased
their speed after the turn, slightly decreased it until 180-m and kept the
speed almost constant until the end of the race. In sum, C3 exhibited an
‘increase-decrease-slightly increase’ of speed profile, reflecting a cubic
model, while the quite constant speed of C1 could correspond to a linear
model.

157
 sp or t science

Discussion
The novelty of our approach was the use of latent class mixed model
(LCMM) to investigate the intra- and inter-individual variability based on
time series (i.e. cycle-to-cycle swimming speed) in competitive swim-
ming. The coefficients associated to spline components gave us infor-
mation about the different sections of the race: after the turn (i.e. arm
strokes after the underwater phase of the turn-out), the middle section
of the race (i.e. clean swimming part) and before the turn or the finish
(i.e., approach of the wall).
The first main finding of our study was the similarity of race manage-
ment and the low number of profiles (only two classes) observed for the
first lap of 50, 100 and 200-m freestyle. In particular, the first profile dem-
onstrated the capacity of swimming fast after the underwater phase of
the start, which was followed by a high decrease of speed until the wall.
Conversely, the second profile reflected a lower speed after the under-
water phase of the start than the first profile but that remained more
stable or even increased until the 25-m (for the 100 and 200-m), and then
followed by a smaller decrease of speed than in the first profile. This sec-
ond profile is well exemplified in figure 4 where the female S1 adopted
the same race management for the 50-m and the first lap of the 100 m.
Similarly, the male S3 adopted the same race management for L1 of the
100 m and the 200-m, and for L2 of the 100 m and the 200-m.

Figure 4. Normalized values of speed for two different freestyle races for the
same swimmer: 100 m for S3 (upper left panel) and 200-m for S3 (lower left
panel); 50-m for S1 (upper right panel) and 100 m (lower right panel).

158
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

The second main finding is the specificity of the race management

for the 200-m; particularly the higher number of profiles (i.e. three
classes) for the two last laps in comparison to the two first laps of the
200-m (i.e. two classes). Notably, it must be emphasized that only male
swimmers composed the class C3, exhibiting the highest variation of
speed with a fast speed after the turn and a great decrease of speed
(for an example, see swimmer S3 in Fig. 5). The high variation of speed
within the lap for the swimmers of C3 suggested an important effect of
turn-in, turn-out, fatigue and/or a too fast start (this later could lead
to great decrease of phosphocreatine). This class C3 suggested that
higher was the swimming speed in comparison to the speed during the
turn, more able were the swimmers to generate speed after the turn;
however, it appeared hard to maintain this high speed to approach the
wall.
The higher number of profiles for L3 and L4 of the 200-m also sug-
gested that the swimmers might switch among various profiles be-
tween laps. For instance, the figure 5 exemplified three different pro-
files of race management of three international male swimmers (S1, S2
and S3) for the 200-m freestyle (Fig. 5).
Interestingly, those three international male swimmers showed
close final time (106.81 ± 0.56 s) and high world ranking for the 200-
m race between July 2015 and April 2016, suggesting that swimmers
could exhibit similar performance outcome but with various race man-
agements. The swimmer S2 managed his four laps in the same way (as
they are all included in C1); those laps exhibited a quite stable or slight-
ly increase of speed at the beginning of the lap, and then decrease of
speed from the middle of the lap to the wall. Conversely, the swimmer
S1 and S3 used the same speed management for L1 and L2; in partic-
ular a great decrease of speed, which was characterized by the class
C2. Then, at L3 and L4, the swimmer S1 switched to C1 profile of man-
agement (i.e. a more stable speed within the lap than in C2 profile),
whereas the swimmer S3 switched to C3 profile of management (i.e.
an ‘increase-decrease’ of speed within the lap). These three examples
suggested that (i) the swimmers could either maintain the same man-
agement all along the race or switch from one profile (let say one class
of race management) to another profile (let say to another class of race
management), and (ii) the swimmers could exhibited similarities of
speed profile as they shared the same management for some laps of
the race, but they could also showed differences of speed profile for
other laps of the race. Interestingly, this inter-individual variability

159
 sp or t science

suggested that even at top level, various profiles of race management

could coexist and could lead to high swimming speed, however with-
out providing any prediction on winning the race.

Figure 5. Trajectory of normalized values of speed for the four laps of the 200-
m freestyle for three international male swimmers (S1, S2 and S3). The final
time (FT) and the world ranking (WR) at the 200-m freestyle between July
2015 and April 2016. Black refers to the first class (C1), grey refers to the sec-
ond class (C2) and light grey refers to the third class (C3).

In summary, our LCMM well emphasized that the different profiles

of race management mainly related to in on hand the start, the turn
and the finish effects (as already highlighted by Veiga and Roig (2016)
and in another hand, the fatigue effect (as already mentioned by Aujo-
uannet, Bonifazi, Hintzy, Vuillerme, & Rouard, 2006; Toussaint, Carol,
Kranenborg, & Truijens, 2006). According to Veiga and Roig (2016), the
speed at the start and at the turn was higher than the average speed

160
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

(i.e. during the clean part of the swim) during the 100 m at the 2013 FINA
World Championships. Our study reinforced those results by showing
that some swimmers generated speed after the turn, but also lost speed
when approaching the wall and turning. Our results confirmed the previ-
ous effect of turn (i.e. the turn itself but also the underwater phase after
the turn until starting swimming) highlighted for instance by Elipot, Di-
etrich, & Hellard (2010), Houel, Elipot, André, & Hellard, (2013)Elipot, Di-
etrich, & Hellard (2010), Houel, Elipot, André, & Hellard, (2013) and Puel
et al., (2012). Puel et al., (2012). However, our study also emphasized that
there was not only one profile of race management as some swimmers
exhibited a more linear management of the speed with a more or less
constant speed or a more or less constant decrease of speed. It could
be hypothesized that these different profiles of race management relate
to the various profiles of turn and of underwater phase after the turn.
For instance, Puel et al., (2012) suggested that longer wall push-off times
could lead to faster velocities but mentioned that too much time spent
on the wall could be directly prejudicial to performers by an increase
of the turn time and a decrease of the overall performance. Therefore,
there is not only one profile of turn, but instead the swimmers must find
the best compromise to apply the highest force in the shortest turn time.
This necessary compromise led to significant inter-individual variability
of the turn time and turn distance during competition (e.g. Veiga & Roig,
2016). Similar results have been found about the effect of start on swim-
ming speed (Seifert et al., 2010; Vantorre, Seifert, Fernandes, Vilas-Boas,
& Chollet, 2010b; Vantorre, Seifert, Fernandes, Boas, & Chollet, 2010a).
In particular, both for aerial and underwater phase, Seifert et al., (2010),
Vantorre et al., (2010a) showed that different profiles could lead to simi-
lar short 15-m start time. For instance, to generate great take off speed,
the swimmers must make a compromise between a long time spent on
the block to create more force and a short time on the block to minimize
the time deficit (Seifert et al., 2010).
A second hypothesis to understand higher inter-individual variabil-
ity of race management of the 200-m freestyle related to fatigue (Figue-
iredo, Pendergast, Vilas-Boas, & Fernandes, 2013a), (Figueiredo, Rouard,
Vilas-Boas, & Fernandes, 2013b). As already emphasized in previous
publications, fatigue led the decrease in swimming speed that is related
to a decrease in SL and in SR through the event (Alberty, Sidney, Pelayo,
& Toussaint, 2009; Aujouannet et al., 2006; Sidney, Alberty, Leblanc, &
Chollet, 2011; Toussaint et al., 2006) which was explained by a decrease
in mechanical power output (e.g., 24% decrease of the mechanical pow-

161
 sp or t science

er output was observed for the 100 m by Toussaint et al., (2006). Our
results in table 1 confirmed this hypothesis as SL and SR decreased be-
tween laps. This decrease of SL and mechanical power output appeared
concomitant with the decrease of anaerobic alactic contribution and
the increase of aerobic contribution (Figueiredo, Zamparo, Sousa, Vilas-
Boas, & Fernandes, 2010). The works simulated muscular metabolism
(Mader, 2003),(Rodriguez & Mader, 2011) showed that relative rest phas-
es such as during underwater phase after the turn (~9 to 10 s) could ena-
ble a partial recovery of phosphocreatine. From there, the high decrease
of phosphocreatine during the first 50-m of the 200-m could explain the
decrease of swimming speed(Rodriguez & Mader, 2011). At L2 and L3, the
partial recovery of phosphocreatine during the turn as well as the higher
contribution of aerobic system could explain the speed increase occur-
ring until the middle of the lap (as observed in class C3). At L4 of the 200-
m, the increase of the lactic anaerobic contribution until 80% (Rodriguez
& Mader, 2011) associated to a decrease of the arm stroke efficiency and
an increase of 16% of energy cost of locomotion Figueiredo, Pendergast,
Vilas-Boas, & Fernandes (2013a),
Figueiredo, Rouard, Vilas-Boas, & Fernandes (2013b). Figuereido,
Toussaint, Vilas-Boas, & Fernandes (2013c), (Figueiredo, Rouard, Vilas-
Boas, & Fernandes, 2013b) could explain the decrease of speed observed
in our study. The inter-individual variability observed for this energy sys-
tem model (e.g. an increase of 40% of the standard deviation for the lac-
tic anaerobic contribution in L4 of the 200-m; (Figueiredo et al., 2010)
might also explain the inter-individual variability that we observed for
the speed management. One hypothesis could be that the swimmers
with high aerobic capacity could maintain a more constant swimming
speed in comparison to swimmers with higher anaerobic contribution
that could involve muscular fatigue and higher decrease of swimming
speed at the end of L4.

In conclusion, our LCMM helped to explain and understand the effect

of start, turn and finish, associated to fatigue, about the race manage-
ment in competitive swimming. By distinguishing various profiles of race
management in top elite swimmers, we emphasized the functional role
of inter- and intra-individual variability because it might reflect continu-
ous and dynamic interaction between an individual and the environ-
ment to achieve the task-goal. Indeed, according to the ecological dy-
namics framework (Davids, Araújo, Seifert, & Orth, 2015; Seifert et al.,
2013; Seifert & Davids, 2012) the intertwined relationship between inten-

162
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

tions, perceptions and actions constrains the direction and restrain the
range of movement possibilities available for each individual perform-
er. Thus, skilled swimmers showed the capacity to functionally adapt
their behaviour by using a stable management or by being flexible (i.e.
switching form profile to another profile) to satisfy these key constraints
to achieve high performance outcome. Thus, our study suggested that
expertise in swimming could be expressed through the relationship be-
tween behavioural flexibility (i.e., functional variability to adapt to a set
of constraints) and stability (i.e., robustness of motor functions undergo-
ing internal and external disturbances, such as fatigue) under interacting
performance constraints (e.g., task, environment and personal)(Newell,
1986; Seifert et al., 2013; Warren, 2006). Skilled swimmers can individu-
ally and functionally adapt their race management during performance,
exhibiting degenerate behaviours. Indeed, as already demonstrated in
cognitive and perceptual-motor systems (Edelman & Gally, 2001; Mason,
2010; Price & Friston, 2002; Whitacre, 2010) the achievement of the same
task-goal (i.e. to perform one function) by using various race manage-
ment profiles (i.e. use of many behavioural structures) reflects the ca-
pacity to exploit degeneracy property in neurobiological system (Seifert
et al., 2016). To explain this exploitation of degeneracy in race manage-
ment, our hypothesis is that expert swimmers developed a more func-
tional relationship with the performance environment, predicated on
perception and action coupling. The development of expertise leads to
the enhanced capacity for skilled performers to be attuned to functional
properties of the environment (i.e. informational variable to approach
the wall in order to turn without losing too much speed)(Davids et al.,
2015; Fajen, Riley, & Turvey, 2009). This is because experts are more ca-
pable of exploiting information about environmental and task-related
constraints to functionally (re)organise and regulate their behaviour,
continuously, to achieve consistent performance outcomes.

Bibliography
Abdel-Aaziz, Y. I., & Karara, H. M. (1971). Direct linear transformation from com-
parator coordinates into object space coordinates in close-range photo-
grammetry (pp. 1–18). Presented at the ASP Symposium on close range
photogrammetry, ASP Symposium on close range photogrammetry. http://
doi.org/10.1080/10671188.1967.10616517
Alberty, M., Sidney, M., Pelayo, P., & Toussaint, H. M. (2009). Stroking Charac-
teristics during Time to Exhaustion Tests. Medicine & Science in Sports &
Exercise, 41(3), 637–644. http://doi.org/10.1249/MSS.0b013e31818acfba
Aujouannet, Y. A., Bonifazi, M., Hintzy, F., Vuillerme, N., & Rouard, A. H. (2006).
Effects of a high-intensity swim test on kinematic parameters in high-lev-

163
 sp or t science

el athletes. Applied Physiology, Nutrition, and Metabolism, 31(2), 150–158.

http://doi.org/10.1139/h05-012
Benarab, D., Napoléon, T., Alfalou, A., & Verney, A. (2017). Optimized swim-
mer tracking system based on a novel multi-related-targets approach.
Optics and Lasers in Enginieering, 89, 195–202. http://doi.org/10.1016/j.opt-
laseng.2016.05.013
Bernstein, N. A. (1996). Dexterity and Its Development. (M. L. Latash & M. T. Tur-
vey, Eds.). Mahwah, New Jersey: Lawrence Erlhaurn Associates Inc.
Carrière, I., Farré, A., Proust-Lima, C., Ryan, J., Ancelin, M. L., & Ritchie, K. (2016).
Chronic and remitting trajectories of depressive symptoms in the elderly.
Characterisation and risk factors. Epidemiology and Psychiatric Sciences,
1–11. http://doi.org/10.1017/S2045796015001122
Davids, K., Araújo, D., Seifert, L., & Orth, D. (2015). An ecological dynamics per-
spective. In J. Baker & D. Farrow (Eds.), Routledge Handbook of Sport Exper-
tise (pp. 130–144). London.
Davids, K., Bennett, S., & Newell, K. M. (2006). Movement System Variability.
Human Kinetics.
Edelman, G. M., & Gally, J. A. (2001). Degeneracy and complexity in biological sys-
tems. Proceedings of the National Academy of Sciences of the United States
of America, 98(24), 13763–13768. http://doi.org/10.1073/pnas.231499798
Elipot, M., Dietrich, G., & Hellard, P. (2010). High-level swimmers’ kinetic effi-
ciency during the underwater phase of a grab start. Journal of Applied Bio-
mechanics, 26(4), 501–507. http://doi.org/10.1123/jab.26.4.501
Fajen, B. R., Riley, M. A., & Turvey, M. T. (2009). Information, affordances, and the
control of action in sport. International Journal of ….
Figueiredo, P., Pendergast, D., Vilas-Boas, J., & Fernandes, R. (2013a). Interplay
of biomechanical, energetic, coordinative, and muscular factors in a 200
m front crawl swim. BioMed Research International Int, 897232. http://doi.
org/10.1155/2013/897232
Figueiredo, P., Rouard, A., Vilas-Boas, J., & Fernandes, R. (2013b). Upper- and
lower-limb muscular fatigue during the 200-m front crawl. Applied Physi-
ology, Nutrition, and Metabolism, 38(7), 716–724. http://doi.org/10.1139/
apnm-2012-0263
Figuereido, P., Toussaint, H. M., Vilas-Boas, J. P., & Fernandes, R. J. (2013c). Rela-
tion between efficiency and energy cost with coordination in aquatic loco-
motion. European Journal of Applied Physiology, 113(3), 651–9.
Foster, C., de Koning, J. J., Hettinga, F., Lampen, J., La Clair, K. L., Dodge, C., et al.
(2003). Pattern of Energy Expenditure during Simulated Competition. Medi-
cine & Science in Sports & Exercise, 35(5), 826–831. http://doi.org/10.1249/01.
MSS.0000065001.17658.68
Foster, C., deKoning, J. J., Hettinga, F., Lampen, J., Dodge, C., Bobbert, M., &
Porcari, J. P. (2004). Effect of competitive distance on energy expenditure
during simulated competition. International Journal of Sports Medicine,
25(3), 198–204. http://doi.org/10.1055/s-2003-45260
Foster, C., Hendrickson, K. J., Peyer, K., Reiner, B., deKoning, J. J., Lucia, A., et al.
(2009). Pattern of developing the performance template. British Journal of
Sports Medicine, 43(10), 765–769. http://doi.org/10.1136/bjsm.2008.054841
Foster, C., Snyder, A. C., Thompson, N. N., Green, M. A., Foley, M., & Schrager, M.

164
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

(1993). Effect of pacing strategy on cycle time trial performance. Medicine &
Science in Sports & Exercise, 25(3), 383-388. http://doi.org/10.1249/00005768-
199303000-00014
Guignard, B., Rouard, A., Chollet, D., Hart, J., Davids, K., & Seifert, L. (2017). Indi-
vidual–Environment Interactions in Swimming: The Smallest Unit for Ana-
lysing the Emergence of Coordination Dynamics in Performance? Sports
Medicine, 1–13. http://doi.org/10.1007/s40279-017-0684-4
Hellard, P., Dekerle, J., Avalos, M., Caudal, N., Knopp, M., & Hausswirth, C. (2008).
Kinematic measures and stroke rate variability in elite female 200-m swim-
mers in the four swimming techniques: Athens 2004 Olympic semi-finalists
and French National 2004 Championship semi-finalists. Journal of Sports
Sciences, 26(1), 35–46. http://doi.org/10.1080/02640410701332515
Houel, N., Elipot, M., André, F., & Hellard, P. (2013). Influence of angles of attack,
frequency and kick amplitude on swimmer’s horizontal velocity during
underwater phase of a grab start. Journal of Applied Biomechanics, 29(1),
49–54.
Komar, J., Chow, J.-Y., Chollet, D., & Seifert, L. (2015). Neurobiological degen-
eracy: Supporting stability, flexibility and pluripotentiality in complex
motor skill. Acta Psychologica, 154, 26–35. http://doi.org/10.1016/j.actp-
sy.2014.11.002
Mader, A. (2003). Glycolysis and oxidative phosphorylation as a function of cy-
tosolic phosphorylation state and power output of the muscle cell. Euro-
pean Journal of Applied Physiology, 88(4), 317–338. http://doi.org/10.1007/
s00421-002-0676-3
Mason, P. H. (2010). Degeneracy at Multiple Levels of Complexity. Biological The-
ory, 5(3), 277–288. http://doi.org/10.1162/BIOT_a_00041
Mytton, G. J., Archer, D. T., Gibson, A. S. C., & Thompson, K. G. (2014). Reliability
and Stability of Performances in 400-m Swimming and 1500-m Running.
International Journal of Sports Physiology and Performance, 9(4), 674–679.
http://doi.org/10.1123/ijspp.2013-0240
Mytton, G. J., Archer, D. T., Turner, L., Skorski, S., Renfree, A., Thompson, K. G., &
Gibson, A. S. C. (2015a). Increased Variability of Lap Speeds: Differentiating
Medalists and Nonmedalists in Middle-Distance Running and Swimming
Events. International Journal of Sports Physiology and Performance, 10(3),
369–373. http://doi.org/10.1123/ijspp.2014-0207
Mytton, G. J., Archer, D. T., Turner, L., Skorski, S., Renfree, A., Thompson, K. G., &
St Clair Gibson, A. (2015b). Increased variability of lap speeds: differentiat-
ing medalists and nonmedalists in middle-distance running and swimming
events. International Journal of Sports Physiology and Performance, 10(3),
369–373. http://doi.org/10.1123/ijspp.2014-0207
Newell, K. M. (1986). Constrains on the development of coordination. In M. G.
Wade & H. T. A. Whiting (Eds.), Motor Develepment in Children Aspects of Co-
ordination and Control (pp. 341–360). Dordrecht.
Newell, K. M., & Corcos, D. M. (1993). Variability and motor control (pp. 1–12).
Champaign:Human Kinetics Publishers.
Price, C. J., & Friston, K. J. (2002). Degeneracy and cognitive anatomy.
Trends in Cognitive Sciences, 6(10), 416–421. http://doi.org/10.1016/S1364-
6613(02)01976-9

165
 sp or t science

Proust-Lima, C., Philipps, V., & Liquet, B. (2015, March 3). Estimation of extend-
ed mixed models using latent classes and latent processes: the R package
lcmm. arXiv.org.
Puel, F., Morlier, J., Avalos, M., Mesnard, M., Cid, M., & Hellard, P. (2012). 3D kin-
ematic and dynamic analysis of the front crawl tumble turn in elite male
swimmers. Journal of Biomechanics, 45(3), 510–515. http://doi.org/10.1016/j.
jbiomech.2011.11.043
Rein, R., Davids, K., & Button, C. (2010). Adaptive and phase transition behavior
in performance of discrete multi-articular actions by degenerate neurobio-
logical systems. Experimental Brain Research, 201(2), 307–322. http://doi.
org/10.1007/s00221-009-2040-x
Robertson, E. Y., Pyne, D. B., Hopkins, W. G., & Anson, J. M. (2009). Analysis of lap
times in international swimming competitions. Journal of Sports Sciences,
27(4), 387–395. http://doi.org/10.1080/02640410802641400
Rodriguez, F., & Mader, A. (2011). Energy systems in swimming. In L. Seifert, D.
Chollet, & I. Mujika (Eds.), World book of swimming: From Science to Perfor-
mance (pp. 225–240). New York: Nova Science.
Schmidt, R. A., & Lee, T. D. (2011). Motor Control and Learning (5 ed.). New York:
Human Kinetics.
Seifert, L., & Davids, K. (2012). Intentions, perceptions and actions constrain
functional intra-and inter-individual variability in the acquisition of ex-
pertise in individual sports. Open Sports Sciences Journal, 5(Suppl 1-M8),
68–75.
Seifert, L., Button, C., & Davids, K. (2013). Key properties of expert movement
systems in sport : an ecological dynamics perspective. Sports Medicine,
43(3), 167–178. http://doi.org/10.1007/s40279-012-0011-z
Seifert, L., Komar, J., Araújo, D., & Davids, K. (2016). Neurobiological degenera-
cy: A key property for functional adaptations of perception and action to
constraints. Neuroscience and Biobehavioral Reviews, 69, 159–165. http://
doi.org/10.1016/j.neubiorev.2016.08.006
Seifert, L., Komar, J., Barbosa, T., Toussaint, H., Millet, G., & Davids, K. (2014a).
Coordination Pattern Variability Provides Functional Adaptations to Con-
straints in Swimming Performance. Sports Medicine, 44(10), 1333–1345.
http://doi.org/10.1007/s40279-014-0210-x
Seifert, L., Komar, J., Crettenand, F., & Millet, G. (2014b). Coordination Pat-
tern Adaptability: Energy Cost of Degenerate Behaviors. PLoS ONE, 9(9),
e107839–9. http://doi.org/10.1371/journal.pone.0107839
Seifert, L., Vantorre, J., Lemaitre, F., Chollet, D., Toussaint, H. M., & Vilas-Boas,
J. P. (2010). Different profiles of the aerial start phase in front crawl. Journal
of Strength and Conditioning Research / National Strength & Conditioning As-
sociation, 24(2), 507–516. http://doi.org/10.1519/JSC.0b013e3181c06a0e
Seifert, L., Wattebled, L., Herault, R., Poizat, G., Adé, D., Gal-Petitfaux, N., & Da-
vids, K. (2014c). Neurobiological Degeneracy and Affordance Perception
Support Functional Intra-Individual Variability of Inter-Limb Coordination
during Ice Climbing. PLoS ONE, 9(2), e89865–12. http://doi.org/10.1371/jour-
nal.pone.0089865
Sidney, M., Alberty, M., Leblanc, H., & Chollet, D. (2011). Stroking Parameters
during Competition. In World book of swimming: From Science to Perfor-

166
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

mance (pp. 443–458). Nova Science.

Summers, J. J., & Anson, J. G. (2009). Current status of the motor program: Re-
visited. Human Movement Science, 28(5), 566–577. http://doi.org/10.1016/j.
humov.2009.01.002
Thompson, K. (2015). Pacing:Individual Strategies for Optimal Performance.
Champaign, Illinois: Human Kinetics.
Thompson, K. G., Haljand, R., & MacLaren, D. P. (2000). An analysis of selected
kinematic variables in national and elite male and female 100- m and 200-
m breastroke swimmers. Journal of Sports Sciences, 18(6), 1–13. http://doi.
org/10.1080/02640410050074359
Thompson, K. G., MacLaren, D. P., Lees, A., & Atkinson, G. (2004). The effects
of changing pace on metabolism and stroke characteristics during high-
speed breaststroke swimming. Journal of Sports Sciences, 22(2), 149–157.
http://doi.org/10.1080/02640410310001641467
Tononi, G., Sporns, O., & Edelman, G. M. (1999). Measures of degeneracy and
redundancy in biological networks. Proceedings of the National Academy
of Sciences of the United States of America, 96(6), 3257–3262. http://doi.
org/10.1073/pnas.96.6.3257
Toussaint, H. M., Carol, A., Kranenborg, H., & Truijens, M. J. (2006). Effect of
Fatigue on Stroking Characteristics in an Arms-Only 100 m Front-Crawl
Race. Medicine & Science in Sports & Exercise, 38(9), 1635–1642. http://doi.
org/10.1249/01.mss.0000230209.53333.31
Tucker, R., & Lambert, M. I. (2006). An analysis of pacing strategies during men’s
world-record performances in track athletics. International Journal of
Sports Physiology and Performance.
Vantorre, J., Seifert, L., Fernandes, R. J., Boas, J. P. V., & Chollet, D. (2010a). Kin-
ematical profiling of the front crawl start. International Journal of Sports
Medicine, 31(1), 16–21. http://doi.org/10.1055/s-0029-1241208
Vantorre, J., Seifert, L., Fernandes, R. J., Vilas-Boas, J. P., & Chollet, D.
(2010b). Comparison of Grab Start between Elite and Trained Swim-
mers. International Journal of Sports Medicine, 31(12), 887–893. http://doi.
org/10.1055/s-0030-1265150
Veiga, S., & Roig, A. (2016). Effect of the starting and turning performances on
the subsequent swimming parameters of elite swimmers. Sports Biome-
chanics, 16(1), 34–44. http://doi.org/10.1080/14763141.2016.1179782
Warren, W. H. (2006). The dynamics of perception and action. Psychological Re-
view, 113(2), 358–389. http://doi.org/10.1037/0033-295X.113.2.358
Whitacre, J. M. (2010). Degeneracy: a link between evolvability, robustness and
complexity in biological systems. Theoretical Biology and Medical Model-
ling, 7(1), 6. http://doi.org/10.1186/1742-4682-7-6

167
 sp or t science

Acute effects of various post-activation

potentiation protocols on 15-m and 25-m
freestyle swimming
Daniel Juárez Santos-García1, Enrique Alonso Plaza1 and José
María González Ravé1.

Introduction
The phenomenon of post-activation potentiation (PAP) has been
a topic of considerable interest in recent years (Docherty, Robbins
and Hodgson, 2004; Ebben and Watts, 1998), with its foundations being
the following: exercise with heavy loads increases motor neuron excit-
ability and the reflection of potentiation, which can create optimal con-
ditions for subsequent exercise (Chu, 1996; Fees, 1997; Fleck and Kontor,
1986). A possible explanation for this phenomenon is the improvement
in prestimulation of motor neuron excitability (increased recruitment of
motor units, better synchronisation or reduction in presynaptic inhibi-
tion) (Aagaard, 2003; Aagaard, Simonsen, Andersen, Magnusson and
Dyhre-Poulsen, 2002; Gullich and Schmidtbleicher, 1996; Trimble and
Harp, 1998). In this respect, PAP could be a strategy to improve perfor-
mance in various sports. However, although there are clear benefits to
using this principle as a long-term training intervention, there is also the
potential of incorporating it into a warm-up prior to performance (Mat-
thews, Matthews and Snook, 2004).
The power developed during swimming is a fundamental aspect (es-
pecially in the specialties that include associated short tests) for achiev-
ing optimal performance in competition, due to its close relationship
with swimming speed and, consequently, the time needed to complete
the distance (González-Ravé et al., 2011). Systems and elements that en-
able the implementation of greater swimming resistance have tradition-
ally been employed to develop specific swimming strength (Girold, Cam-
els, Maurin, Milhau and Chatard 2006, 2007; Patnott, Post and Northius,
2003; Wright, Bramer and Stager, 2009).
Girold et al. (2006) indicated that sprint training with resistance bands
in swimming was more effective than traditional training or assisted
sprint, improving times in the 100 m freestyle and enhancing muscle

1. Sport Training Lab, Faculty of Sport Sciences, University of Castilla-La Man-

cha, Spain.

168
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

strength and power. These studies refer to training consisting of several

sessions; however, there are few studies on the acute effects of a single
session in swimming. Kilduff et al. (2011) observed that after a series of
three repetitions at 87% of 1-repetition maximum, there were no differ-
ences in the times for the 15-m swim length at maximum velocity. Juárez
et al. (2013) found no significant differences in the times for the 25 m
length after various tasks were performed 30 s before swimming (rubber
band traction for 10 s, 12 m of swimming with 30% of 1-repetition maxi-
mum resistance in this task). Each group repeated this protocol on four
occasions, with 2 minutes of rest between repetitions. No significant dif-
ferences were found either between the various groups or between the
repetitions performed by each group.
Another study with swimmers (Cuenca-Fernández, López-Contreras
and Arellano, 2015) compared two PAP protocols on launch perfor-
mance. In the first protocol, the swimmers performed three lunges at
85% 1-repetition maximum. In the second, they performed four repeti-
tions simulating the launch in swimming but using a Yo-Yo device. The
study found better performance results in the launch with the latter
protocol. On the other hand, Sarramian, Turner and Greenhalgh (2015)
compared the effect of 4 types of warm-up on 50-m freestyle times: tra-
ditional warm-up, PAP focused on the upper body, PAP focused on the
lower body and combined PAP. The authors observed that a traditional
warm-up achieved better times when compared with the upper body
PAP warm-up, with no significant differences between the rest of the
warm-ups. However, when the data were analysed according to sex, the
authors found that the men achieved better times with the upper body
PAP warm-up than with the traditional warm-up or combined PAP.
In this respect, more studies are needed to analyse these issues, con-
sidering the importance that any meaningful improvement, even small,
has for short tests. Therefore, the objectives for this study were the fol-
lowing: (i) to analyse the acute responses caused by various tasks with
high loads (semi-squat, pulldown to the chest and both) compared with
traditional warm-up in the swim speed for 15 m and 25 m; and (ii) to ana-
lyse the differences in the implementation of this method according to
sex. The hypotheses were the following: (i) performing tasks with heavy
loads beforehand positively influences the swimming times in the 15-
and 25-m lengths; and (ii) applying the PAP training method produces
the same effects regardless of sex.

169
 sp or t science

Materials and methods

Participants
Ten swimmers participated in the study: six men and four women
(mean age, 16.5 ± 1.8 years; mean weight, 67.75 ± 6.89 kg; mean height,
177.90 ± 8.08 cm). All the swimmers were national-level athletes, having
participated in the Spanish Championship in their category during the
season in which the study was conducted. All were finalists, and four of
them had won a medal in this competition. At the time when the study
was conducted, the swimmers trained six times a week, with 2 hours of
training in the swimming pool and three days of dry training (gym) of
approximately 1 hour. All the swimmers had accumulated five to seven
years of high-performance training. The swimmers, their trainers and
the swimmers’ parents were informed of the study characteristics. All
the parents signed an informed consent document. The study was con-
ducted according to the Declaration of Helsinki.

Procedure
The study participants conducted five assessment sessions. The
first session determined the 6-repetition maximum in semi-squat and
lat pulldown, increasing the weight until the swimmers were no longer
able to perform more than six repetitions with a specific load. These two
exercises were selected because they were standard upper- and lower-
limb exercises for these swimmers. In a subsequent session, the swim-
ming times for 25 m were recorded (with partial measurement in the 15
m) after a standard swimming warm-up for these swimmers: 200 m of
swimming, 4 x 50 m legs, 4 x 50 m (25 sculling/25 swimming), 4 x 25 m
changes in pace, 4 x 25 m progressive and 100 m smooth. The times were
recorded manually by a trained expert with a stopwatch.
In the third session, all the swimmers performed a 6-repetition maxi-
mum series of semi-squats, which, according to the Brzycki prediction
equation (1993), represents 86.1% of 1-repetition maximum. After 8 min
of recovery (Bewan et al., 2009; Kilduff et al., 2011, Cuenca-Fernández
et al., 2015), the times for 25 m were recorded (with a partial measure-
ment at 15 m) to observe the direct influence of a leg strength exercise
on swimming speed. In the fourth session, the swimmers performed a
6-repetition maximum series of lat pulldowns. Eight minutes later, they
performed 25 m of swimming at maximum speed. In the last study ses-
sion, the swimmers performed a 6-repetition maximum series of squats
followed by 6-repetition maximum in lat pulldown and, after 8 min, 25 m
of swimming. In this last session, half of the swimmers started with the

170
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

squat series, and the other half started with the lat pulldowns, in a ran-
domised manner. Passive recovery was performed in all sessions.

Statistical analysis
The SPSS v.19.0 statistical program was employed for the data analy-
sis. We calculated the means and standard deviations of the study end-
points. The normality of the sample was checked with the Shapiro-Wilk
test. To analyse the differences in the 15- and 25-m times among the var-
ious measurements, we applied a one-way, repeated measures analysis
of variance (ANOVA), establishing a significance level of p<.05. To analyse
the differences according to sex, we applied a two-way repeated meas-
ures ANOVA (warm-up condition and sex), verifying the homogeneity of
variances using Levene’s test. In addition, a partial eta squared (η2) was
calculated as an index of effect size, and was interpreted as small (0.01),
moderate (0.06) and large (0.14) (Cohen, 1988).

Results
The mean weight lifted in the 6-repetition maximum in the semi-
squat test was 74.5 ± 13.3 kg, whereas the mean weight in the 6-repeti-
tion maximum in the lat pulldown test was 63.0 ± 9.2 kg. Table 1 shows
the results for the 15 m and 25 m. There were no differences in the times
after each of the tasks in the 15 m (η2=0.08). However, we observed a
difference (p<.01; η2=0.43) between the time in the 25 m after the 6-rep-
etition maximum lat pulldown task and the time in the 25 m after the
6-repetition maximum semi-squat plus 6-repetition maximum pulldown
task. When analysing the results according to sex, we found that in no
case there were differences (sex x warm-up condition) in time in the 15 m
(η2=0.07) and 25 m (η2=0.04) in each of the evaluation sessions for either
the men or the women. However, if sex is the only factor considered, we
found a partial eta square of 0.15 for the 15 m and 0.09 for the 25 m.

Task Participants 15 m 25 m
mean ± SD mean ± SD
Traditional warm All subjects (N=10) 7.05 ± 0.71 12.75 ± 1.21
up Men (N=6) 6.84 ± 0.66 12.48 ± 1.30
Women (N=4) 7.37 ± 0.75 13.16 ± 1.11
6-repetition All subjects (N=10) 7.03 ± 0.79 12.78 ± 1.26
maximum Men (N=6) 6.85 ± 0.77 12.53 ± 1.39
Squat
Women (N=4) 7.30 ± 0.86 13.16 ± 1.10

171
 sp or t science

6-repetition All subjects (N=10) 6.96 ± 0.67 12.62 ± 1.09

maximum Men (N=6) 6.74 ± 0.64 12.37 ± 1.20
Pulldown
Women (N=4) 7.31 ± 0.62 12.99 ± 0.94
6-repetition All subjects (N=10) 7.04 ± 0.72 12.83 ± 1.13
maximum Squat Men (N=6) 6.80 ± 0.63 12.54 ± 1.18
+ 6-repetition
maximum Women (N=4) 7.40 ± 0.78 13.27 ± 1.03
Pulldown

Table 1. Time in the 15 m and 25 m after each task.

Discussion
Our aims were to analyse the acute responses caused by various
tasks with heavy loads (semi-squat, lat pulldown and both) compared
with traditional warm-up in the swim speed for 15 m and 25 m and to
compare the results according to sex. We found no improvements in the
times for the 15- and 25-m lengths at maximum speed after previously
performing one or several strength exercises with heavy loads prior to
swimming when compared with the implementation of a traditional
warm-up. This result is in line with Sarramian et al. (2015) who found no
differences in the times for the 50-m freestyle after a traditional warm-
up compared with various PAP warm-ups, observing in some cases bet-
ter times after the traditional warm-up. First, we must consider the task
performed, because it was not an explosive action such as a swimming
start, but rather a longer-lasting action. Although it was also not a task
with the duration of a 15-m or 25-m swimming sprint, we have in any
case shown that the complex method could be interesting for improving
sprint performance.
Matthews et al. (2004) investigated the acute effect of a pre-com-
petition resistance training warm-up on subsequent 20-m sprint per-
formance, observing improvements in 20-m sprints after a 5-repetition
maximum in squats in rugby players. Despite involving a similar dis-
tance to those evaluated in our study, in that case, the 20-m sprints are
in running and not in swimming. Gullich and Schmidtbleicher (1996) and
González-Ravé et al. (2009) stated that the potentiation phenomenon is
only effective when the participants were experts and belong to a profes-
sional group. The current study supports this statement, given that the
selected sample was not an elite group. In contrast Young et al. (1998)
and Evans et al. (2000) obtained gains in potentiation work with midlevel
participants, although the participants did have experience in working
with loads. This differs partially from our study because our participants

172
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

who performed the squat and those who performed the 6-repetition
maximum did not typically work with heavy loads.
The intensity of the load employed in our study was 86.1% of 1-rep-
etition maximum (6-repetition maximum). Kilduff et al. (2011) worked
with three repetitions at 87% of 1-repetition maximum in swimming,
obtaining significant improvements at 8 min in the countermovement
jump but not in the 15-m sprint. This finding is within the 75%–85% mar-
gin of 1-repetition maximum (Radcliffe and Radcliffe, 1996) needed to
promote potentiation, although it is true that these authors proposed
four sets of four repetitions versus the single sets of six repetitions that
was performed in the present study. However, our study is supported by
studies that obtained PAP improvements with a single set of 5-repetition
maximum (Young et al., 1998; Evans et al., 2000; Matthews et al., 2004;
Scott and Docherty, 2004).
There is no clear evidence of a different functioning of the PAP mech-
anism between men and women. It is difficult to compare differences
due to the limited samples and availability of women participants. The
number of studies that treat both groups and establish differences is
limited. It has been suggested that the PAP mechanism could be more
effective in men (Radcliffe and Radcliffe, 1996). A study by Sarramian
et al. (2015) observed that men achieved better times in the 50 m free-
style after an upper-body PAP warm-up compared with a combined PAP
warm-up or a traditional warm-up. In our study, although no differences
were observed, the large and moderate effect sizes according to sex, of
15 m and 25 m, respectively, can suggest the possibility of differences
between sex, if studied with a large sample. In any case, it is unclear
whether this phenomenon affects men and women differently (O’Leary,
Hope, and Sale, 1998).

Conclusions
We can state that there are no significant differences in the times for
the 15 m and 25 m freestyle at maximum speed after performing the vari-
ous tasks with heavy loads (semi-squat, lat pulldown and both) 8 min
before swimming. We can therefore confirm that there is insufficient sci-
entific evidence to confirm that the PAP exercise with these characteris-
tics improves speed performance in a 25-m swimming test. There are no
significant differences according to sex in the recorded sessions or in the
15- and 25-m lengths. There is therefore no apparent difference in the
functioning of the PAP mechanism between men and women, consider-
ing the study characteristics.

173
 sp or t science

References
Aagaard, P. (2003). Training induced changes in neural function. Exercise and
Sport Science Review 31(2), 61-67.
Aagaard, P., Simonsen, E. B., Andersen, J. L., Magnusson, P. and Dyhre-Poulsen,
P. (2002). Neural adaptations to resistance training: evoked V-wave and H-
reflex responses. Journal of Applied Physiology 92(6), 2309-2318.
Bevan, H.R., Owen, N.J., Cunningham, D.J., Kingsley, M.I.C. and Kilduff, L.P.
(2009). Complex training in professional rugby players influence of recov-
ery time on upper-body power output. Journal of Strength and Conditional
Research 23(6), 1780–1785.
Brzycki, M. (1993) Strength testing-predicting a one-rep max from reps-to-fa-
tigue. JOPERD, 68: 88-90.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2ª ed).
NewYork: Academic Press.
Chu, D. A (1996). Explosive power and strength: complex training for maximum
results. Champaign: Human Kinetics.
Cuenca-Fernández, F., López-Contreras, G. and Arellano, R. (2015). Effect on
swimming start performance of two types of activation protocols: lunge
and YoYo squat. Journal of Strength and Conditioning Research, 29(3): 647-
655.
Docherty, D., Robbins, D., and Hodgson, M (2004). Complex training revisited:
a review of its current status as a viable training approach. Strength and
Conditioning Journal 26(6), 52-57.
Ebben, W. P. and Blackard, D. O. (1998). Paired for strength: a look at combin-
ing weight training with plyometric exercises with a focus on vertical jump
improvement. Training and Conditioning 8(3), 55-63.
Evans, A.K., Hodgkins, T.D., Durham, M.P., Berning, J.M. and Adams, K.J. (2000).
The acute effects of a 5RM bench press on power output. Medicine and Sci-
ence in Sports and Exercise 32(5), 312.
Fees, M.A. (1997). Complex training. Athletic therapy today 2(1), 18.
Fleck, S.J. and Kontor, K. (1986). Complex training. National Strength and Con-
ditioning Association Journal 8(5), 66-68.
Girold, S., Camels, P., Maurin, D., Milhau, N. and Chatard, J.C. (2006). Assisted
and resisted sprint training in swimming. Journal of Strength and Condi-
tional Research 20(3), 547-554.
Girold, S., Camels, P., Maurin, D., Milhau, N. and Chatard, J.C. (2007). Effects of
dry-land vs. resisted and assisted sprint exercise on swimming sprint per-
formance. Journal of Strength and Conditional Research 21(2), 599-605.
González-Ravé, J.M., Machado, L., Navarro-Valdivielso, F. and Vilas-Boas, J.P.
(2009). Acute effects of heavy-load exercises, stretching exercises, and
heavy-load plus stretching exercises on squat jump and countermovement
jump performance. Journal of Strength and Conditional Research 23(2),
472–479.
González-Ravé, J.M., Juárez, D., Clemente, V., Muñoz, V., Díaz, G., Carrasco, M.
and Navarro-Valdivielso, F. (2011). Análisis de la potencia máxima especí-
fica en nadadores de diferente nivel. Revista de Entrenamiento Deportivo
25(3), 11-15.

174
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Gullich, A. and Schmidtbleicher, D. (1996). MVC-induced short-term potentia-

tion of explosive force. New Studies in Athletics 11(4), 67-81.
Juárez, D., González-Ravé, J. M., Legaz, A., Portillo, J., Clemente, V. and Newton,
R. U. (2013). Acute effects of two resisted exercises on 25m swimming per-
formance. Isokinetics and Exercise Science, 21(1): 29-35.
Kilduff, L.P., Cunningham, D.J., Owen, N.J., West, D.J,. Bracken, R.M. and Cook,
C.J. (2011). Effect of postactivation potentiation on swimming starts in
international sprint swimmers. Journal of Strength and Conditioning Re-
search 25(9), 2418-2423.
Matthews, M.J., Matthews, H.P. and Snook, B. (2004). The acute effects of a re-
sistance training warm up on sprint performance. Research in Sports Medi-
cine 12(2), 151-159.
O’Leary, D.D., Hope, K. y Sale, D.G. (1998). Influence of gender on post-tetanic
potentiation in human dorsiflexors. Canadian Journal of Physiology and
Pharmacology 76(7-8), 772-779.
Patnott, J.R., Post, K. and Northius, M.E. (2003). Muscular power changes in col-
legiate swimmers. Medicine and Science in Sports and Exercise 35(5). Sup-
plement abstract 1454.
Radcliffe, J.C. and Radcliffe, J.L. (1996). Effects of different warm-up protocols
on peak power output during a single response jump task. Medicine and
Science in Sports and Exercise 28(5), 189.
Sarramian, V.G., Turner, A. and Greenhalgh, A.K. (2015). Effect of postactiva-
tion potentiation on fifty-meter freestyle in national swimmers. Journal of
Strength and Conditioning Research, 29(4): 1003-1009.
Scott, S. and Docherty, D. (2004). Acute effects of heavy pre-loading on vertical
and horizontal jump performance. Journal of Strength and Conditioning
Research 18(2), 201-205.
Trimble, M. and Harp, S. (1998). Postexercise potentiation of the H-reflex in hu-
mans. Medicine and Science in Sports and Exercise 30(6), 933-941.
Wright, B.V., Bramer, C.L. and Stager, J.M. (2009). Five-week assessment of in
water output in competitive swimmers. ACSM 56th. Annual Meeting, Seat-
tle, Washington.
Young, W.B., Jenner, A. and Griffiths, K. (1998). Acute enhancement of power
performance from heavy load squats. Journal of Strength and Condition-
ing Research 12(2), 82-84.

175
 sp or t science

Altitude training for sea level performance: a

systematic review
Ferran A. Rodríguez1 and Sergi Àvila1

Introduction
Altitude training (AT) at natural environment has been a matter of
extensive research for half a century and, despite some sceptical views
(1-3), it continues to play an important role in the preparation of elite
and sub-elite athletes in many countries (4, 5). Paradoxically, there is a
remarkable lack of controlled and adequately powered studies on natu-
ral AT in the scientific literature, particularly in elite athletes, and there is
no clear evidence that AT enhances performance more than training at
sea level (SL) (1, 5-7). The theoretical concept behind this practice is the
independent and combined effects of the physiological processes of ac-
climatization to chronic hypoxia and those derived from training under
the additional stress imposed by exercising in a hypoxic environment
(8). In accordance with some investigations, altitude acclimatization re-
sults in central and peripheral adaptations, i.e. augmented red cell vol-
ume, haemoglobin (Hb) mass and maximal oxygen uptake (V̇ O2max) that
improve primarily systemic oxygen delivery (“erythropoietic paradigm”)
(9, 10), while others argue against this view and support the concept of
“nonhematological” adaptations such as improved muscle efficiency,
greater muscle buffering and the ability to tolerate lactic acid produc-
tion (11, 12).
Conversely, the combination of intense training and hypoxia may
have a negative impact on athlete’s performance capacity and health
status, causing unfavourable effects such as acute mountain sick-
ness (13), immune suppression (14), iron-deficient erythropoiesis (15),
cathecolamine mediated glycogen depletion (16) and increased oxida-
tive stress and tissue damage (17), among others. Interestingly, a recent
meta-analysis concluded that AT performance gains could be related to
a placebo or nocebo effect (7).
There are different strategies to train at altitude. The classical ap-
proach (“live high-train high”, Hi-Hi), used since the late 1960s, involves

1. INEFC-Barcelona Sport Sciences Research Group, Institut Nacional d’Educació

Física de Catalunya (INEFC), Universitat de Barcelona, Spain

176
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

SL-resident athletes who travel to and subsequently live and train at

moderate altitude. Despite being used by many elite swimmers and
coaches, there is no clear evidence that training at natural altitude en-
hances performance more than training at SL (6). In a milestone study
published in the mid-1990s, Levine & Stray-Gundersen provided evi-
dence that the “live high-train low” (Hi-Lo) strategy can improve 3000- to
5000-m running performance in collegiate/club runners (9). Later, this
approach was modified to limit the low-altitude training sessions to
only high-intensity workouts and was subsequently termed “live high-
train high and low” (Hi-HiLo). The improvement in running performance
was associated with increase in red cell mass, the subsequent increase
in maximal oxygen uptake (V̇ O2max), the “high altitude effect”, and the
maintenance of high-intensity training velocities and oxygen flux to the
muscles “the low altitude effect” (18). This paradigm has been sustained
by later investigation in elite endurance athletes performing different
sports including running (19), orienteering (20), and cycling (21). How-
ever, these studies are difficult to compare with each other directly given
the many differences in experimental design (12).
The aims of this systematic review are: 1) to collate and to critically
evaluate the empirical evidence sustaining the use of natural AT in ath-
letes with the main goal of improving SL performance; and 2) to derive
which of the different natural AT strategies is more efficient for enhanc-
ing SL performance when the athletes come back to SL training and
competition. To achieve these goals, we systematically reviewed con-
trolled and uncontrolled studies through the PubMed and SPORTDiscus
databases. The studied participants were athletes from regional to elite
level, the exposure of interest was natural AT, and the main outcome of
interest was performance.

Methods
Literature search
This systematic review followed the PRISMA statement guidelines
(22). To achieve this, a systematic literature search was conducted for
studies in any language indexed in the PubMed and SPORTDiscus da-
tabases (up to March 2017). This search was performed using the fol-
lowing selected keywords: ALL FIELDS, altitude training AND sport AND
performance, NOT simulated OR artificial OR normobaric, NOT review.
To manage the bibliographic references the EndNote (ver. X7) software
was used.

177
 sp or t science

Eligibility criteria and study selection

In order to be considered eligible for inclusion, studies had to meet
the following criteria: 1) participants were healthy adult competitive ath-
letes; 2) studies were controlled and uncontrolled; 3) altitude exposure
was natural (classic or terrestrial, not artificial or simulated); 4) primary
focus was SL performance (articles with no performance measures were
excluded); 5) original studies were used only (no reviews); and 6) studies
published in any language.
The flow chart of literature screening approach and study identifica-
tion is displayed in figure 1. From the 325 articles initially identified, we
removed 51 duplicated articles, 224 were excluded from title and ab-
stract information, and 30 articles were excluded after reading the full
text for not meeting the eligibility criteria or due to impossibility to ac-
cess them. Finally, 20 studies were included in the qualitative review.

Figure 1. Flow chart of literature screening approach and study identification.

178
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Data extraction
We developed a data extraction table classifying the type/level of
athletes, altitude strategy, sample size, study design, follow up during
the intervention, living and training characteristics and effects on perfor-
mance measures. Performance data extraction was done from time tri-
als, race results, power output and total work capacity (cycle ergometry)
and time scores. Weight-adjusted V̇ O2max values and selected haemato-
logical data (Hb mass, Hb blood concentration and haematocrit ratio)
were also extracted.

Results
Study characteristics
A total of 20 studies, 7 controlled and 13 uncontrolled, were identi-
fied for inclusion in the systematic review. These studies involved 439
athletes, from which 173 took part in the controlled studies and 266 in
the uncontrolled studies. 247 out of the total 439 participants were elite
athletes (56%), and the rest were from regional/interregional to subelite
level athletes (44%), a fact that can bring about large differences in per-
formance level and range for improvement, making difficult to compare
investigation results. The intervention methodologies and type of de-
signed utilized were as follows: 14 Hi-Hi (5 controlled), 4 Hi-Lo (3 con-
trolled), and 3 Hi-HiLo (1 controlled
Classical altitude training (Hi-Hi) controlled studies
Only 5 controlled Hi-Hi studies were identified (table 1). Using a
cross-over design, Adams et al. reported no potentiating effect of hard
endurance training at 2300 m over equivalent SL training on 2-mile per-
formance time or V̇ O2max in well-conditioned middle-distance runners
(23) The study by McLean et al. consisted on a Hi-Hi training camp involv-
ing a group of 21 Australian football players and 9 Lo-Lo controls. The
Hi-Hi group likely improved 2000-m time trial performance by 1.5% after
altitude with very large individual variability (90%CI: -3.3–6.3%) and low
individual responsiveness (0.8%). This change was paralleled by a very
likely increase in Hb mass (2.8%) (24). Levine & Stray-Gundersen failed
to find any effect on performance after 4 weeks of Hi-Hi intervention de-
spite an increase of V̇ O2max (3.4%) and red cell mass (10%) (9). Burtscher
et al. studied two groups of amateur runners and found no group dif-
ferences in cycling total work capacity between the Hi-Hi group and the
Lo-Lo controls: 3 and 16 days after the intervention the Lo-Lo group im-
proved 8% and 17% whereas the Hi-Hi group improved 0.3% and 8%,
respectively (25).

179
 sp or t science

Only Rodríguez et al. showed a significant improvement in swimming

performance after living and training at 2320 m during 3 or 4 weeks (e.g.,
3.1% and 3.4% in specific 100 m or 200-m time trial), but this change was
not significantly different from that experienced by the Lo-Lo control
group (3.7%) (26). Interestingly, two studies reporting increases in V̇ O2max
did not find a concomitant improvement in performance compared with
SL controls. In summary, only one study actually provided evidence of
superior improvement of Hi-Hi altitude training compared with SL (24).
However, the magnitude of these changes seems lower than can be ex-
pected because of a SL training camp, and placebo or nocebo effects
cannot be ruled out.
Hi-Lo and Hi-HiLo controlled studies
Four controlled studies were identified (table 2). Levine & Stray-Gun-
dersen were the first to use the Hi-Lo strategy in their classical study cited
above (9), in which they assigned collegiate and club amateur runners to
Hi-Hi, Hi-Lo, and Lo-Lo (control) groups. They reported an improvement
on 5000-m running performance (1.3%) in the Hi-Lo runners three weeks
after the training camp and attributed this to increased V̇ O2max (5%) and
red cell mass (10%), according to the “erythropoietic paradigm”. In a
study by Dehnert et al. two groups of subelite triathletes followed a Lo-
Lo or Hi-Lo intervention for 2 weeks and found no effects on cycling or
treadmill running performance despite a 7% increase in V̇ O2max and un-
changed Hb mass (27). In contrast, Wehrlin et al. (20) studied a group
of 10 elite orienteers using a Hi-Lo strategy for 24 days and comparing
them with 7 Lo-Lo cross country skiers, and reported an improvement
in 5000-m running performance (1.6%) in the Hi-Lo group, paralleled by
increased V̇ O2max (4.1%) and red cell mass (5.3%), which are comparable
to previous results in runners (9). Finally, Rodríguez et al. conducted the
only controlled study using the Hi-HiLo strategy in which athletes live at
altitude and train at the same and a lower altitude (26). Four groups of
international level swimmers were compared: Hi-Hi for 3 and 4 weeks
(previously cited), and Hi-HiLo and Lo-Lo controls for 4 weeks. Although
all groups improved after a well-controlled training camp, the Hi-HiLo
group of swimmers further improved 50-, 100- (sprinters) or 200- (non-
sprinters), and 400-m swimming performance (5.5%, 6.3% and 4.7%, re-
spectively) 2 to 4 weeks after the training camp. However, this substan-
tial improvement in performance could not be attributed to changes in
V̇ O2max, Hb mass or swimming economy and, therefore, to the “erythro-
poietic paradigm”.

180
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Classical altitude training (Hi-Hi) uncontrolled studies

Nine uncontrolled studies using the classical Hi-Hi strategy met the
inclusion criteria (table 3). Overall, only 2 studies showed some evidence
of beneficial effects on performance (28, 29) and 1 was uncertain (30).
Noteworthy, in the study by Roels et al. the modest increase in 2000-m
trial (1.9%) was only significant when the swimmers lived and trained
at 1200 m of altitude, but not at 1850 m (29). Moreover, another study
showed a decrease in performance after 21 to 27 days of Hi-Hi interven-
tion in four different groups (n = 97) of elite swimmers (31). The other 6
studies did not show significant changes in performance despite some
modest changes in V̇ O2max (32, 33), Hb mass (31, 34), blood red cell mark-
ers (30, 35) or without haematological changes (36).
Hi-Lo and Hi-HiLo uncontrolled studies
The systematic review included 3 uncontrolled studies examining the
Hi-Lo strategy and 3 studies using the Hi-HiLo strategy (table 4). These
studies were conducted by mostly the same group of researchers and
used similar designs and methodologies. In a retrospective study using
the Hi-Lo approach, 8 out of 12 collegiate runners were classified as re-
sponders and improved their time in a 5000-m time trial by 3.6% whereas
the non-responder decreased their performance time by 1.3%. The bet-
ter times of the responders was paralleled by a non-significant increase
in V̇ O2max (7.5%) (37). Coincident results were obtained by these authors
in a prospective study with elite distance runners (37). Also Stray-Gun-
dersen et al. reported modest but significant performance gains in 22
elite runners on a 3000-m time trial (1.1%) associated to a 3% increase in
V̇ O2max (18). To identify the optimal altitude for training using the Hi-HiLo
approach, Chapman et al. compared four groups of collegiate runners
living at four altitudes (from 1780 to 2800 m) and training at varying al-
titudes from 1250 to 3000 m. They found that only the middle altitudes
(i.e., 2084 and 2454 m) evoked significant gains in 3000-m time trial run-
ning performance (2.1 to 2.8%), associated to a 3% to 8% increase in
V̇ O2max (38). Similarly, Saugy et al. conducted a study with 13 well-trained
triathletes who lived at 220 m and trained at 1100-1200 m of altitude and
found an improvement in 3000-m time trial running performance (3.3%)
after 3 weeks upon return to sea level, with no changes 1 and 7 days after
the altitude training camp. V̇ O2max also increased by 5.2% (39).

181
 sp or t science

Table 1. Summary of CONTROLLED studies on natural altitude training using

the Hi-Hi strategy for sea level performance enhancement.

a
Lo-Lo, live low–train low; Hi-Hi, live high–train high.
b
F, females; M, males.
c
CON, controlled trial; UN, uncontrolled trial; R, randomised; NR,
non-randomised; X-over, cross-over
d
D#, testing post-intervention (day number).
e
TT, time trial in specific sport; CY, cycling test; TWC, total work capacity; TM,
treadmill test.
f
RCM, red cell mass; RCV, red cell volume.
↑, improvement/increase/benefit; <—>, no change; ↓, worsening/decrease/
harm; =, same as above
*, significantly different from values measured before training or compared
to sea-level controls (p<0.05); n.s., non-significant difference (p≥0.05); ?,
uncertain/not reported.

182
curren t a nd fu t ure t rends for perform a nce op t imiz at ion

183
 sp or t science

Table 2. Summary of CONTROLLED studies on natural altitude training using

the Hi-Lo or Hi-HiLo strategies for sea level performance enhancement.

a
Lo-Lo, living low, training low; Hi-Lo, live high–train low; Hi-HiLo,
live high–train high and low; L, living; T, training.
b
F, females; M, males. -over, crossover; R, randomised; NR, non-randomised.
d
D#, testing post-intervention (day number).
e
TT, time trial in specific sport; CY, cycling test; TM, treadmill test.
f
RCM, red cell mass; RCV, red cell volume.
↑, improvement/increase/benefit; <—>, no change; ↓, worsening/decrease/
harm; =, same as above
*, significantly different from values measured before training or compared
to sea-level controls (p<0.05); n.s., non-significant difference (p≥0.05); ?,
uncertain/not reported.

184
curren t a nd fu t ure t rends for perform a nce op t imiz at ion

185
 sp or t science

Table 3. Summary of UNCONTROLLED studies on natural altitude training us

a
Lo-Lo, living low, training low; Hi-Hi, live high–train high; L, living; T, training.
b
F, females; M, males.
c
CON, controlled trial (vs. sea level); UN, uncontrolled trial (vs. sea level);
X-over, crossover; R, randomised; NR, non-randomised.
d
D#, testing post-intervention (day number).
e
TT, time trial in specific sport; CY, cycling test; PPO, peak power output; TM,
treadmill test; Re, responders; NRe, non-responders; pts, FINA score points.

186
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

f
RCM, red cell mass; RCV, red cell volume.
↑, improvement/increase/benefit; <—>, no change; ↓, worsening/decrease/
harm; =, same as above
*, significantly different from values measured before training or compared
to sea-level controls (p<0.05);
<0.05);
0.05); n.s., non-significant diff
difference
erence (p≥0.05); ?, un-
certain/not reported.

187
 sp or t science

Table 4. Summary of UNCONTROLLED studies on natural altitude training us-

ing the Hi-Lo or Hi-HiLo strategies for sea level performance enhancement.

a
Lo-Lo, living low, training low; Hi-Hi, live high–train high; L, living; T, training.
b
F, females; M, males.
c
CON, controlled trial (vs. sea level); UN, uncontrolled trial (vs. sea level); X-
over, crossover; R, randomised; NR, non-randomised.
d
D#, testing post-intervention (day number).
e
TT, time trial in specific sport; CY, cycling test; PPO, peak power output;
TM, treadmill test; Re, responders; NR, non-responders; FINA pts, FINA score
points.
f
RCM, red cell mass; RCV, red cell volume.
↑, improvement/increase/benefit; <—>, no change; ↓, worsening/decrease/
harm; =, same as above

188
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

*, significantly different from values measured before training or compared

to sea-level controls (p<0.05);
<0.05);
0.05); n.s., non-significant diff
difference
erence (p≥0.05); ?, un-
certain/not reported.

189
 sp or t science

Discussion
This systematic review, which aimed at assessing the empirical evi-
dence sustaining the use of AT in athletes with focus on SL performance
enhancement, does not appear sufficiently robust to determine the ef-
ficacy and appropriate characteristics (duration, altitude and training
requirements) of an AT camp. Neither it can conclude which of the natu-
ral AT strategies is best for enhancing performance at SL. The reviewed
studies are difficult to compare with each other directly, given the many
differences in experimental design, type of participants, outcome meas-
ures and methodology. Notwithstanding, there seems to be a certain
consensus—perhaps lacking compelling evidence to support it—that
when athletes are exposed to a high enough altitude, for a long enough
amount of time, and are able to preserve fitness by training hard under
both hypoxic and normoxic conditions, the majority may improve physi-
cal performance (40).
We have reviewed publications from 1967 to 2017, a 50-year period.
A total of 20 studies have been appraised, but only 7 published articles
(33%) had a controlled design. The remaining 14 studies (67%) were un-
controlled and provide low quality evidence since performance changes
can be attributed to training alone, training camp effect or placebo/
nocebo effect. Concerning the 5 controlled Hi-Hi studies, only the one
published by MacLean et al. (24) provided limited evidence of superior
improvement compared with SL controls, although the high interindi-
vidual variability (-3.3–6.3%) argues against a real AT effect and placebo/
nocebo effects cannot be ruled out.
In the last decade the Hi-Lo approach has gained interest over the
classical Hi-Hi strategy in the scientific literature and among many en-
durance athletes (4). In our review, the results of the 4 controlled Hi-Lo or
Hi-HiLo studies seem somewhat more convincing compared with those
using the classical Hi-Hi approach. In their milestone study, Levine &
Stray-Gundersen showed that the Hi-Lo strategy evoked an increase in
5000-m time trial performance (1.3%) in collegiate and club runners (9),
despite some researchers argue that this modest improvement could be
attributed to a placebo or nocebo effect, as another Hi-Hi group showed
the same improvement in V̇ O2max and red cell mass without any change
in performance (1). Similar limitations can be attributed to the study by
Wehrlin et al. since the improved 5000-m running performance (1.6%)
lacked of concomitant performance measures in the control group (20).
A recent investigation was the first to show substantial performance im-
provements after a terrestrial Hi-HiLo intervention using a controlled

190
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

design (26). This investigation showed that SL swimming performance

of elite swimmers in 100- (sprinters) or 200-m (non-sprinters) time trials
was not altered, or in some cases impaired immediately, but improved
significantly by ~3.1–3.7% after 1 to 4 weeks of recovery following com-
pletion of a coach-prescribed training camp, whether it was conducted
at SL or at moderate altitude (2320 m). By including 2 weekly sessions
of high-intensity training at lower altitude (Hi-HiLo strategy) a greater
improvement in performance occurred 2 and 4 weeks after the train-
ing camp (5.3% and 6.3%, respectively). Similarly, further gains in 400-
and 50-m freestyle time trial performance was noted 2 weeks (4.2% and
5.2%, respectively) and 4 weeks (4.7% and 5.5%, respectively) following
return to SL. In addition, this study shows that the delayed performance
improvements are not linked to changes in V̇ O2max, oxygen kinetics or Hb
mass and hence cannot be attributed exclusively to an enhanced sys-
temic oxygen transport capacity.
Globally, these results are in line with estimations published in a meta-
analytic review by Bonetti & Hopkins, who concluded that “performance
changes in studies using the conventional Hi-Hi approach were unclear,
whereas changes using the terrestrial Hi-Lo strategy were considered
likely to be effective both for elite and subelite athletes (~4%), or a more
realistic 1.5% when performance was predicted from uncontrolled stud-
ies” (7). These estimations are in line with a recent review by Saunders
et al. (5) in which, by using a regression analysis of average performance
changes, it was estimated that a 3-week terrestrial AT camp would elicit
mean performance improvements of ~1.8% (Hi-Hi) and ~2.5% (Hi-Lo) (5).
Levine and Stray-Gundersen were the first to test the hypothesis
that acclimatization to moderate altitude (2500 m) plus training at low
altitude (1250 m) (Hi-Lo paradigm) improves SL performance in well-
trained runners more than in equivalent SL or Hi-Hi controls (9). They
also concluded that the correlation between the increase in V̇ O2max and
the improvement in 5000-m time after the field training camp argues
strongly that this is a key adaptation during altitude training and a nec-
essary mechanism for improving SL performance. Despite this, the per-
formance gain was only 1.3% from pre-training values. Notwithstanding,
this study changed the previous AT paradigm that was only focussed on
the classical Hi-Hi strategy.
Concerning the 9 uncontrolled Hi-Hi studies included in this review,
we can see a very similar picture as for the controlled studies, since only
two studies showed some beneficial effects on performance (28, 29), one
was uncertain (30), and 1 even showed impaired performance after in a

191
 sp or t science

very large number of elite swimmers (31).

Among the uncontrolled Hi-Lo or Hi-HiLo investigations, two (37)
should be analysed with caution as the participants, collegiate or elite
runners, were categorized post-facto into ‘responders’ and ‘non-re-
sponders’, as they aimed at investigating the individual variation in re-
sponse to AT. The other 2 studies showed performance changes similar
to the controlled groups, although the risk of placebo or nocebo effects
is exacerbated. Interestingly, moderate altitudes of ~2100-2500 m were
identified as optimal for benefits in 3000-m running performance (41).

Conclusions
There are several limitations in every study design using terrestrial
AT, including the impossibility of blinding the intervention, limitations in
recruiting large numbers of participants, difficulties in group randomisa-
tion, control of placebo and nocebo effects, large variability in the re-
sponse, etc. These barriers make difficult the comparison of the exist-
ing studies and the design of new investigations that can meet the high
standards of scientific research.
Contrary to common expectations, the systematic review of 20 arti-
cles published along 50 years (1967 to 2017) shows that the quality of
the empirical evidence about using natural altitude training in competi-
tive athletes with the main goal of improving sea level performance is
far from being compelling. However, the available evidence supports the
concept that the Hi-Lo and Hi-HiLo strategies offer the best potential for
performance benefits, as at least two controlled studies provided sound
evidence of positive effects on performance in collegiate/club runners
and elite swimmers, respectively. Uncontrolled studies also support this
concept despite the lower quality of the evidence.

References
1. Lundby C, Millet GP, Calbet JA, Bartsch P, Subudhi AW. Does ‘altitude
training’ increase exercise performance in elite athletes? Br J Sports Med.
2012;46(11):792-5.
2. Jacobs RA. Con: Live high-train low does not improve sea-level perfor-
mance beyond that achieved with the equivalent living and training at sea
level. High Alt Med Biol. 2013;14(4):328-30.
3. Hahn AG, Gore CJ, Martin DT, Ashenden MJ, Roberts AD, Logan PA. An
evaluation of the concept of living at moderate altitude and training at sea
level. Comp Biochem Physiol A Mol Integr Physiol. 2001;128(4):777-89.
4. Wilber RL. Application of altitude/hypoxic training by elite athletes. Med
Sci Sports Exerc. 2007;39(9):1610-24.
5. Saunders PU, Pyne DB, Gore CJ. Endurance training at altitude. High Alt

192
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Med Biol. 2009;10(2):135-48.

6. Rodríguez FA. Training at real and simulated altitude in swimming: too
high expectations? In: Kjendlie P-L, Stallman RK, Cabri J, editors. Biome-
chanics and Medicine in Swimming XI. Oslo: Norwegian School of Sport
Science; 2010. p. 30-2.
7. Armstrong LE, Johnson EC, Ganio MS, Judelson DA, Vingren JL, Kupchak
BR, et al. Effective body water and body mass changes during summer
ultra-endurance road cycling. J Sports Sci. 2015;33(2):125-35.
8. Levine BD, Stray-Gundersen J. Dose-response of altitude training: how
much altitude is enough? Adv Exp Med Biol. 2006;588:233-47. Epub
2006/11/09.
9. Levine BD, Stray-Gundersen J. “Living high-training low”: effect of moder-
ate-altitude acclimatization with low-altitude training on performance. J
Appl Physiol. 1997;83(1):102-12.
10. Levine BD, Stray-Gundersen J. Point: positive effects of intermittent hy-
poxia (live high:train low) on exercise performance are mediated primarily
by augmented red cell volume. J Appl Physiol. 2005;99(5):2053-5.
11. Gore CJ, Hopkins WG. Counterpoint: positive effects of intermittent hy-
poxia (live high:train low) on exercise performance are not mediated pri-
marily by augmented red cell volume. J Appl Physiol. 2005;99(5):2055-7;
discussion 7-8.
12. Gore CJ, Clark SA, Saunders PU. Nonhematological mechanisms of im-
proved sea-level performance after hypoxic exposure. Med Sci Sports Ex-
erc. 2007;39(9):1600-9.
13. Fukuda T, Maegawa T, Matsumoto A, Komatsu Y, Nakajima T, Nagai R, et al.
Effects of acute hypoxia at moderate altitude on stroke volume and car-
diac output during exercise. Int Heart J. 2010;51(3):170-5.
14. Mazzeo RS. Altitude, exercise and immune function. Exerc Immunol Rev.
2005;11:6-16.
15. Stray-Gundersen J, Alexander C, Hochstein A, Lemos D, Levine BD. Failure
of red cell volume to increase with altitude exposure in iron deficient run-
ners. Med Sci Sport Exerc. 1992;24(5):S90.
16. Young AJ. Energy substrate utilization during exercise in extreme environ-
ments. Exerc Sport Sci Rev. 1990;18:65-117.
17. McGinnis G, Kliszczewiscz B, Barberio M, Ballmann C, Peters B, Slivka D, et
al. Acute hypoxia and exercise-induced blood oxidative stress. Int J Sport
Nutr Exerc Metab. 2014;24(6):684-93.
18. Stray-Gundersen J, Chapman RF, Levine BD. “Living high-training low” al-
titude training improves sea level performance in male and female elite
runners. J Appl Physiol (1985). 2001;91(3):1113-20.
19. Stray-Gundersen J, Levine BD. Live high, train low at natural altitude.
Scand J Med Sci Sports. 2008;18 Suppl 1:21-8.
20. Wehrlin JP, Zuest P, Hallen J, Marti B. Live high-train low for 24 days in-
creases hemoglobin mass and red cell volume in elite endurance athletes.
J Appl Physiol. 2006;100(6):1938-45.
21. Jones AM, Haramizu S, Ranchordas M, Burke L, Stear S, Castell LM. A-Z
of nutritional supplements: dietary supplements, sports nutrition foods
and ergogenic aids for health and performance--Part 27. Br J Sports Med.

193
 sp or t science

2011;45(15):1246-8.
22. Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gotzsche PC, Ioannidis JP, et al.
The PRISMA statement for reporting systematic reviews and meta-anal-
yses of studies that evaluate health care interventions: explanation and
elaboration. PLoS medicine. 2009;6(7):e1000100.
23. Adams WC, Bernauer EM, Dill DB, Bomar JB, Jr. Effects of equivalent sea-
level and altitude training on VO2max and running performance. J Appl
Physiol. 1975;39(2):262-6.
24. McLean BD, Gore CJ, Kemp J. Application of ‘Live Low-Train High’ for En-
hancing Normoxic Exercise Performance in Team Sport Athletes. Sports
Med. 2014. Epub 2014/05/23.
25. Burtscher M, Nachbauer W, Baumgartl P, Philadelphy M. Benefits of train-
ing at moderate altitude versus sea level training in amateur runners. Eur J
Appl Physiol. 1996(74):558-63.
26. Rodriguez FA, Iglesias X, Feriche B, Calderon-Soto C, Chaverri D, Wachs-
muth NB, et al. Altitude training in elite swimmers for sea level perfor-
mance (Altitude Project). Med Sci Sports Exerc. 2015;47(9):1965-78.
27. Dehnert C, Hutler M, Liu Y, Menold E, Netzer C, Schick R, et al. Erythropoie-
sis and performance after two weeks of living high and training low in well
trained triathletes. Int J Sports Med. 2002;23(8):561-6.
28. Mizuno M, Juel C, Bro-Rasmussen T, Mygind E, Schibye B, Rasmussen B, et
al. Limb skeletal muscle adaptation in athletes after training at altitude. J
Appl Physiol (1985). 1990;68(2):496-502.
29. Roels B, Hellard P, Schmitt L, Robach P, Richalet JP, Millet GP. Is it more ef-
fective for highly trained swimmers to live and train at 1200 m than at 1850
m in terms of performance and haematological benefits? Br J Sports Med.
2006;40(2):e4.
30. Siewierski M, Słomiński P, Białecki R, Adamczyk J. Athletic performance of
swimmers after altitude training (2,300 m above sea level) in view of their
blood morphology changes. Biol Sport. 2012;29(2).
31. Wachsmuth NB, Volzke C, Prommer N, Schmidt-Trucksass A, Frese F, Spahl
O, et al. The effects of classic altitude training on hemoglobin mass in
swimmers. Eur J Appl Physiol. 2013;113(5):1199-211.
32. Faulkner JA, Kollias J, Favour CB, Buskirk ER, Balke B. Maximum aerobic ca-
pacity and running performance at altitude. J Appl Physiol. 1968;24(5):685-
91.
33. Schmitt L, Millet G, Robach P, Nicolet G, Brugniaux JV, Fouillot JP, et al. In-
fluence of “living high-training low” on aerobic performance and economy
of work in elite athletes. Eur J Appl Physiol. 2006;97(5):627-36.
34. Gough CE, Saunders PU, Fowlie J, Savage B, Pyne DB, Anson JM, et al. In-
fluence of altitude training modality on performance and total haemoglo-
bin mass in elite swimmers. Eur J Appl Physiol. 2012;112(9):3275-85. Epub
2012/01/12.
35. Faulkner JA, Daniels JT, Balke B. Effects of training at moderate altitude on
physical performance capacity. J Appl Physiol. 1967;23(1):85-9.
36. Hue O, Antoine-Jonville S, Sara F. The effect of 8 days of training in tropical
environment on performance in neutral climate in swimmers. Int J Sports
Med. 2007;28(1):48-52.

194
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

37. Chapman RF, Stray-Gundersen J, Levine BD. Individual variation in re-

sponse to altitude training. J Appl Physiol. 1998;85(4):1448-56.
38. Chapman RF, Karlsen T, Resaland GK, Ge RL, Harber MP, Witkowski S, et
al. Defining the “dose” of altitude training: how high to live for optimal sea
level performance enhancement. J Appl Physiol (1985). 2013.
39. Saugy JJ, Schmitt L, Cejuela R, Faiss R, Hauser A, Wehrlin JP, et al. Com-
parison of “live high-train low” in normobaric versus hypobaric hypoxia.
PLoS One. 2014;9(12):e114418.
40. Wilber RL. Introduction to altitude/hypoxic training symposium. Med Sci
Sports Exerc. 2007;39(9):1587-9.
41. Rehrer NJ, Brouns F, Beckers EJ, Frey WO, Villiger B, Riddoch CJ, et al.
Physiological changes and gastro-intestinal symptoms as a result of ultra-
endurance running. Eur J Appl Physiol Occup Physiol. 1992;64(1):1-8.

195
 sp or t science

Dietary nitrate effects in human athletic

performance
Maria Silva1 and Mónica Sousa2

Introduction
The field of sports nutrition has been studied for more than 50 years.
It keeps growing fast, bringing new strategies to enhance athletic per-
formance, to prepare for competition and to recover from exercise (1).
Indeed, there are some supplements and sports foods proved to boost
performance and recovery; nitrates (NO3-) seem to be one of them and
they are growing in popularity among athletes (1). These anions have
been studied for the past few years to understand their effects in athletic
performance.
According to Bailey SJ (1) and Lansley KE (2) NO3- supplementation,
either as beetroot juice or sodium nitrate, has benefits to cardiovascular
health, reducing blood pressure, enhancing blood flow and increasing
the driving pressure of oxygen to exercising tissue. Due to these reasons,
it is believed that NO3- could increase exercise performance (1). The ef-
fects of NO3- seem to be mediated via reduction to nitrite (NO2-) and then
to nitric oxide (3-5). In fact, NO has been linked to vasodilatation, an-
giogenesis, mitochondrial biological processes, glucose uptake and cal-
cium pump (6). All these properties appear to be very important in what
respects sports, namely exercise efficiency and exercise performance (6).
Until recently, it was believed that NO was generated solely from L-
arginine by an enzyme called NO synthase (7), resulting in endogenous
production of NO3- and NO2-. However, now it is known that there are
two pathways. So, NO can be produced from L-arginine and oxygen and,
subsequently, be oxidized to NO2- and NO3- or, alternatively, NO3- can
be reduced to NO2- by xanthine oxidase and facultative anaerobic bac-
teria in the oral cavity, and NO2- can be further reduced to NO and other
reactive nitrogen species (6). This process is described in Figure 1.

1. Faculty of Nutrition and Food Sciences, University of Porto, Portugal

2. Nutrition & Metabolism, NOVA Medical School, Faculdade de Ciências Médi-
cas, Universidade NOVA de Lisboa, Portugal

196
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Figure 1 - Adapted from Jones AM., Dietary nitrate supplementation and ex-
ercise performance. Nitrate (NO3-), nitrite (NO2-) and NOS (NO synthase).

Baseline levels of NO2- correlate positively with exercise performance

in highly trained athletes independent of endothelial function (8). In
fact, NO has been linked to vasodilatation, angiogenesis, mitochondrial
respiration, mitochondrial biogenesis, glucose uptake and sarcoplasmic
reticulum Ca2+ handling (6). All these properties are relevant to athletic
performance from fatigue resistance and exercise efficiency to exercise
performance (6). However, potential benefits related to NO3- supple-
mentation have been a controversial question. In fact, there are several
studies (9-11) which demonstrate that NO3- may not have any benefits
in exercise performance. Therefore, this review focuses on understand-
ing the evidence-base of nitrate as an ergogenic supplement; on which
mechanisms may be responsible for these ergogenic effects and to put
forward some practical guidelines for a useful and safe use of nitrates.

Nitrate food sources

Although NO3- exists in a range of vegetables (12) beetroot is the most
popular. It has been already demonstrated that the NO3- content present
in this vegetable, in fact, gives rise to NO which has been showed to have
benefits in exercise performance (6). Besides a high content of NO3-, it is
easy to deliver as a juice. Although it can be made at home, a commercial
option is often the most convenient and palatable option. There are al-
ready many commercial options of beetroot juice. “BEET it” is one of the
most known brand; BeetElite Neo shot, Love beets and Biotta also seem
to be good options, as it is showed in Table 1.

197
 sp or t science

James James white

white drinks beet Beet Biotta
Love
drinks it organic elite neo beetroot
beets
beet it beetroot shot juice
sport juice 1 L
Per serving 70 ml 250 ml 414 ml 10 g 250 ml
Energy (Kcal) 72 103 160 30 110
Protein 2.5 g 2.3 g 2g 0 2g
Total carbohydrates 16 g 23 g 41 g 8g 25 g
Of which sugars 16 g 22.5 g 30 g 7g 24 g
Total fat < 0,1 g < 0,1 g 0 0g 0g
Fibre < 0,5 g < 0,5 g 2g 0g 1g
Sodium < 0,1 g < 0,1 g 230 mg 15 mg 65 mg
Dietary nitrate 0.4 g 0.25 g N.M. 0.4 g N.M.

Table 1 - Beetroot juice: nutritional information

However, the consumption of foodstuffs with a reasonable content in

the compounds responsible for the increment of NO has not been direct-
ly linked to an enhancement of sports performance. Some of these com-
pounds might be L-arginine or L-citrulline which is a precursor of the first
one. For example, 300 ml per day of watermelon juice (3.4 g L-citrulline
per day) for 16 days have increased plasma concentration of L-arginine,
L-citrulline and NO2- (13). However, although systolic pressure was found
to be higher in the group consuming watermelon juice compared with
the control group, the time-to-exhaustion during a severe-intensity ex-
ercise test was not significantly different between groups (13). Nonethe-
less, the skeletal muscle oxygenation index during moderate-intensity
exercise was better among individuals supplemented with watermelon
juice (13). Moreover, other recent studies have shown that although
the plasma antioxidant capacity was found higher in watermelon juice
supplemented groups, there seems to be no influence on post-exercise
inflammation and changes in innate immune function (14). Therefore,
although watermelon juice supplementation can increase plasma NO2-
and improve muscle oxygenation during moderate-intensity exercise, it
does not seem to improve exercise endurance or even to lower blood
pressure in healthy adults (13).
Also, it is important to note that NO is synthetized from L-arginine by
the nitric oxide synthase. L-arginine is considered a semi-essential ami-
no acid because the body normally produces it in sufficient amounts.

198
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

According to Alvares et al. (3) physiological concentrations of L-arginine

in healthy individuals are enough to saturate nitric oxide synthase (3)
– L-arginine supplementation might not be useful. Meirelles and col-
laborators (15) also concluded that acute supplementation of L-arginine
neither affects strength performance nor NO production(15). Thus, if an
athlete intends to use nitric oxide as an ergogenic supplement, it might
be more efficient to use the anaerobic pathway: nitrates – nitrites – nitric
oxide. Approximately 80% of NO3- is derived from vegetable consump-
tion fertilizer (12).
Dietary NO3- is determined by the level of nitrate in the water sup-
ply, type of vegetables consumed, amount of vegetables consumed and
level of NO3- in the vegetables, including the NO3- content of fertilizer
(12). Nitrate is mainly found in cell vacuoles and is transported in the
xylem, which carries water and nutrients from the roots to the leaves.
This means that leaf crops such as cabbage, lettuce and spinach have
fairly large NO3- concentrations (16). The accumulation of NO3- is subject
of factors such as genotype, soil conditions, growth conditions and stor-
age and transport conditions. So, the average NO3- content of one type
of vegetable can vary substantially, as reported in Table 2.

Vegetable types and

Nitrite Nitrate
varieties
mg/100 g fresh weight mg/100 g fresh weight
Root vegetables
Carrot 0.002–0.023 92–195
Mustard leaf 0.012–0.064 70–95
Green vegetables
Lettuce 0.008–0.215 12.3–267.8
Spinach 0–0.073 23.9–387.2
Cabbage
Chinese cabbage 0–0.065 42.9–161.0
Bok choy 0.009–0.242 102.3–309.8
Cabbage 0–0.041 25.9–125.0
Cole 0.364–0.535 76.6–136.5
Melon
Wax gourd 0.001–0.006 35.8–68.0
Cucumber 0–0.011 1.2–14.3
Nightshade
Eggplant 0.007–0.049 25.0–42.4

199
 sp or t science

Table 2 - Adapted from Norman J Hord in “Food sources of nitrates and nitrites:
the physiologic context for potential health benefits”.

So, NO3- is consumed through the regular diet. In fact, median daily
intake of dietary NO3- in highly trained athletes is 106 mg (which is con-
sidered an ergogenic amount), although with a large inter-individual
variation (17). Thus, increasing the intake of nitrate-rich vegetables in
the diet, such as spinach, might be a great alternative strategy for NO3-
supplementation.

Nitrates Effects in Endurance Exercise

Following ingestion, NO3- is converted to NO2- by commensal faculta-
tive bacteria residing in crypts on the surface of the tongue (6). In condi-
tions of low oxygen availability, NO2- can be converted into NO, which is
known to play a number of important roles in vascular and metabolic
control (6). Dietary NO3- supplementation increases plasma NO2- con-
centration and reduces the oxygen cost of submaximal exercise and can,
in some circumstances, enhance exercise tolerance (18). It seems that a
single oral dose of 10mg/Kg of NO3- acutely (in an individual with 60 Kg
of body mass corresponds to ~9,5 mmol) reduces the VO2 peak without
compromising the maximal exercise performance (18).
Dietary NO3- supplementation with beetroot juice has been studied
over past years and there is still not many agreement regarding this prac-
tice. In 2011, the results of one study (2) suggested that acute dietary
NO3- supplementation with 6.2 mmol of NO3- ingested 2.5 h before the
event, improves cycling time trial performance in moderately trained
cyclists. One year later this theory was refuted with one study (9) that
showed that a single dose of beetroot juice with a higher content of NO3-
(140 ml – 8.7 mmol of NO3-) 2.5 h before the event did not have any effect
in male well-trained cyclists.
Longer-term use of NO3- supplementation has also been studied. In
one study (19) with dietary supplementation in the form of beetroot
juice, a group of 12 trained males ingested 140 ml of beetroot juice per
day (8 mmol/d NO3-) or placebo, for six days. After supplementation, on
day six, subjects performed 60 min of submaximal cycling followed by a
10-km time trial. Six days of NO3- supplementation reduced VO2 during
submaximal exercise and improved time-trial performance in moder-
ately trained cyclists. In the same year, it was published another study
(20) which concluded that a high dose (8.4 mmol NO3-) of beetroot juice,
but not a moderate dose (4.2 mmol NO3-), consumed 2 h before exer-

200
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

cise, may had improve a 2000-m rowing performance in highly trained

athletes.
The timing and dose of NO3- supplementation is highly variable
among studies. However, improvements in performance arising from an
acute dose of NO3- 75 – 150 min before exercise suggest that ergogenic
effects occur in a short time frame (21). Intake of beetroot juice should
be initiated within 90 min prior to exercise and at least 6 – 8 mmol of NO3-
is required in order to enhance endurance performance (22).

Nitrates Effects in Resistance Exercise

A variety of enzymes can catalyse the reduction of NO2- to NO and it is
known that this process is facilitated in conditions of ischemia, hypoxia
and low pH (6).
As far as NO3- effects in strength sports are concerned, there are fur-
ther less studies. However, it is known that during exercise the skeletal
muscle is under low partial pressure oxygen and pH conditions (6). It is
known that nitrate-nitrite-nitric oxide pathway is greatly enhanced in
hypoxia and acidosis (23). Therefore, it might be possible that the stimu-
lation of the nitrate-nitrite-nitric oxide pathway may influence muscle
function and exercise performance (6).
Notwithstanding, previous studies (24-28) have showed that NO can
lower force production, via depressing force output during isometric
contractions. In 2012, Fulford and collaborators (29) concluded that the
ingestion of 0.5 L/day of beetroot juice (10.2 mmol of NO3-) for fifteen
days resulted in a reduction in the mean phosphocreatine (PCr) cost
of exercise averaged. Nevertheless, increasing NO bioavailability by in-
gesting beetroot juice did not seem to be associated with a reduction of
force-generating capability in skeletal muscle. Also, these investigators
found that NO3- supplementation reduces PCr cost of force production.
In 2014, a research (30) showed that seven days of NO3- supplementation
(9.7 mmol per day) increased excitation-contraction coupling at low fre-
quencies of stimulation and enhanced evoked explosive force produc-
tion in untrained humans. However, voluntary force production was not
affected.
Another study (31) aimed to investigate the effects of NO3- supple-
mentation on performance of bench press resistance until failure. The
authors concluded that 70 ml of a shot of beetroot juice (BEET it Sport
nitrate) containing 6.4 mmol/L or 400 mg of NO3- improved the perfor-
mance in this resistance training session. Also in 2016, a study (32) con-
cluded that supplementation with a nitrate rich beetroot extract-based,

201
 sp or t science

which contained 3 g of concentrated beetroot extract and was fortified

with 35.2 mg of dietary NO3- (Advocare VO2 Prime Performance Nutrition
Bar), provided neuromuscular advantages during metabolically taxing
resistance exercise. These results were consistent with other trials: Cog-
gan and collaborators (33) concluded that acute dietary NO3- intake (11.2
mmol NO3- 2 h before the event) increased whole-body NO production
and muscle speed and power of knee extension by, on average, 11 and
6%, respectively, in healthy men and women. Although relatively small,
improvements of this magnitude might be significant in power-based
sports for an elite athlete, even a 1% increase in performance ability can
double chances of winning a particular event (34).
NO interacts with many different types of proteins, such as ryano-
dine receptors or guanylate cyclase (35). Ryanodine receptors are sarco-
plasmic reticulum calcium-release channels whereas guanylate cyclase
converts guanosine triphosphate to cyclic guanosine monophosphate
(cGMP) (35). The general consequence of cGMP-mediated effects of
NO is to improve mechanical and metabolic muscle power, similar to a
transformation of slow-twitch to fast-twitch muscle, an effect also sum-
marized as “slow-to-fast” shift (36). This may be the main reason for the
ergogenic effects of NO3- supplementation in resistance sports.
Also, in this type of exercise, supplementation protocols were highly
variable across studies. Although more research is needed, consuming
5 – 9 mmol of NO3- 2.5 – 3 h prior to an exercise bout seems to be a good
strategy (37).

Nitrates Effects in Intermittent Exercise

Intermittent exercise is becoming very popular among those who
want to lose body fat. Regular high intensity intermittent exercise in-
creases aerobic and anaerobic fitness and brings out significant muscle
adaptations (38). Also, it seems that intermittent exercise has strong
acute and chronic positive effects on insulin sensitivity and fat loss (39).
Fatigue development during this type of exercise is linked to the decline
of phosphocreatine concentration in muscle; since NO3- has been shown
to lower the PCr cost of force production during high-intensity inter-
mittent exercise (39). This finding suggests that NO3- supplementation
might improve performance during high-intensity interval training.
Wylie and collaborators (39) concluded that a 5-day supplementation
with 8.2 mmol NO3- 2.5 h prior to exercise improved performance during
repeated bouts of short-duration maximal-intensity interspersed with
short recovery periods, but not during longer duration intervals or when

202
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

a longer recovery duration was applied. It might be possible that NO3-

supplementation has more performance enhancement in some particu-
lar types of intermittent exercise (39). On the other hand, Aucouturier
and collaborators (40) concluded that NO3- supplementation was linked
to a better tolerance to supramaximal intensity intermittent exercise by
increasing total red blood cells concentration in the working muscle in
the absence of effects on contractile function.
Also in high-intensity interval training (HIIT), 490 ml of beetroot juice
(28 mmol NO3-) 30 h before the event improved the performance and
seemed to be a very useful ergogenic aid for team sports players (41).
Besides, during prolonged intermittent exercise, such as team sports,
research shows that 140 ml of beetroot juice per day (12.8 mmol NO3-
per day) for seven days enhanced repeated sprint performance and at-
tenuated the decline in cognitive function (reaction time, for example)
(42). In national-level male kayak athletes, 4.8 mmol NO3- improved
exercise economy predominantly reliant on the aerobic energy system
(43). Greater volumes of beetroot juice (9.6 mmol NO3-) also resulted in
enhancements of time trial performance in international-level female
kayak athletes. Nyakayiru and collaborators concluded that 140 mL of
beetroot juice (~800 mg nitrate/day) for six subsequent days improved
high-intensity intermittent type exercise performance in trained soccer
players (44). This 6-day beetroot juice protocol resulted in quantifiable
improvements in high-intensity intermittent running performance and
on the Yo-Yo IR1 test. During this test, mean heart rate was lower follow-
ing beetroot juice ingestion than following nitrate-depleted beetroot
juice treatment
CrossFit is a sport which includes a wide variety of functional high-
intensity exercises. Performance in this sport is highly dependent on
peak power and fatigue resistance (45). Therefore, NO3- could be a useful
ergogenic agent for CrossFit athletes. Kramer and colleagues concluded
that six days of dietary NO3- supplementation (8 mmol potassium NO3-
per day) in male CrossFit athletes improved the peak Wingate power 24
h following the final NO3- dose; however, there were no improvements
in 2-km rowing time-trials or CrossFit performance (46). Hence, further
research is needed to stablish the best supplementation protocol. Not-
withstanding, 8 mmol NO3- 2 – 4 h prior to exercise seems to be a rea-
sonable option to take advantage from to promote ergogenic effects by
NO3- .

203
 sp or t science

Metabolic Adaptations
Dietary NO3- has been linked to the reduction of oxygen cost of sub-
maximal exercise (6). Besides, NO3- supplementation seems to enhance
performance and improve exercise tolerance in moderately trained in-
dividuals (47). It is essential to be aware that there are some variables
that can affect the ergogenic benefits related to nitrate supplementation
in elite athletes. Being an elite athlete demands a very intense exercise-
training routine, which is connected with a much higher daily energy ex-
penditure and, of course, with a greater daily energy intake in compari-
son with a non-trained subject (6). According to Porcelli (47) the optimal
NO3- loading required to increase NO in plasma and, consequently, to
enhance performance in elite athletes is quite different from low-fit sub-
jects - elite athletes need a higher dose to improve sports performance.
This means that the individual fitness level seems to affect the benefits
induced by dietary NO3- supplementation. Additionally, if an athlete
maintains a healthy food pattern, dietary NO3- consumption is high (11).
Maybe for that reason, the plasma NO2- baseline is elevated in compari-
son with non-trained subjects. Alternatively, it has been concluded that
baseline plasma NO2- or NO3- is higher in endurance-trained athletes
compared to untrained individuals due to a greater nitric oxide synthase
activity (48). These differences might reduce the NO3- supplementation
potential, altering the physiological responses to exercise in endurance
athletes (11).
An elite athlete spends decades on intense exercise training.
This induces reconditioning responses in the skeletal muscle vasculari-
zation and mitochondrial efficiency (49). To what skeletal muscle vascu-
larization may concern, there is an increase in capillary density which is
likely to reduce the chance of developing a hypoxic environment in the
active muscle (49). Therefore, the pathway nitrate–nitrite–nitric oxide is
compromised. Therefore, this pathway seems to have more relevance in
sedentary individuals compared to elite athletes (11).
Another issue that is important to be aware is that an elite ath-
lete is an elite in a specific sport. It is expected that in that specific sport,
oxygen delivery does not restrict performance in a highly-trained athlete
who does that every single day (49). Also, NO3- supplementation seems
to reach better results if exercise activities combine a higher relative
oxygen consumption (%VO2 maximum) with higher absolute workout
(Kg/W). A greater dependency on fast-twitch muscle fibre recruitment
and anaerobic energy supply can improve contractile function and in-
crease performance (50). This suggests that endurance athletes, who

204
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

typically have a low proportion of these fibres in their body, might not
benefit of the physiological response to NO3- supplementation (6).

Acute Supplementation versus Chronic Supplementation

Acute and chronic NO3- supplementation have been used in plenty
of studies with controversial conclusions. It is not yet known whether
shorter or longer periods of supplementation have similar or different
results. Webb et al. (51) assessed the effects of a single dose of beetroot
juice (22.5 mmol of NO3-) on plasma nitrite concentration and blood
pressure over 24 h. Plasma NO2- concentration reached its peak 3 h post-
ingestion and remained close to this value for more 2 h. It returned to
baseline after 24 h. The systolic and diastolic blood pressure decreased
2.5 – 3 h after the beetroot juice intake (51).
Larsen FJ et al. (52) used two periods of tree days separated by a
washout interval of ten days. The daily dose of 0.1 mmol of NO3-/Kg was
divided and ingested three times daily. The results showed a lower oxy-
gen demand during submaximal work and an increase in lactate concen-
tration, indicating that energy production had become more efficient.
In 2010, Vanhatalo and collaborators showed that low (3 mmol) and
moderate (6 mmol) amounts of NO3- supplementation increased plasma
NO2- in a dose-dependent manner after 2 h, seven days and four weeks
of supplementation. The same did not happen with the submaximal ex-
ercise oxygen uptake – a low dose did not have any effects either after 2
h, seven days or four weeks of supplementation. However, a moderate
dose lowered this marker in all groups. Generally speaking, a reduction
in submaximal exercise oxygen uptake can be achieved by ingesting 6
mmol NO3- per day chronically with and without the acute ingestion of
nitrates.
In athletes, acute beetroot consumption (6.2 mmol of NO3-) seems to
increase the mean power output and the power output to VO2 ratio and
mitochondrial efficiency (53) and reduces blood pressure (54). These ef-
fects are maintained if supplementation is continued. Long-term NO3-
supplementation modulates muscle function through its role in the
regulation of blood flow, glucose uptake, calcium homeostasis, mito-
chondrial respiration and biogenesis (6).
Although 15 days of NO3- supplementation has been shown to im-
prove exercise performance (37), most of recent investigations suggest
that chronic NO3- supplementation over seven days also has positive er-
gogenic effects (54).
For now, it seems that NO3- supplementation may enhance exercise

205
 sp or t science

performance either after 2 – 3 h of supplementation or after seven or

more days.

Are nitrates safe?

NO3- and NO2- can form N-nitrosamines which are potential carcino-
gens (37). In fact, in 2011 dietary N-compounds were associated with a
higher gastrointestinal cancer incidence, specifically rectal cancer (55).
Indeed, the Food and Drugs Administration (FDA) put restrictions to reg-
ulate nitrates and nitrites levels in both food and water within the US.
However, in 2015, a study (56) showed that the presence of NO2- in food
is free of danger and a diet high in NO3- had benefits for health. In turn,
the European Food Safety Authority (EFSA) defends that no new data
were identified that would require a revision of Adequate Dietary Intake
(41) values – 0 – 3.7 mg NO3-/Kg body weight and 0 – 0,07 mg NO2-/Kg
body weight (57).
While direct ingestion of NO2- is potentially hazardous, the use of NO3-
from natural vegetable sources, such as beetroot juice, is of much less
concern in terms of acute toxicity. In this case, only a minor part of the
NO3- is converted in vivo (58). The average increase of NO2- concentra-
tion in saliva is 20 ppm/100 mg NO3- ingested (59). Also, any possible
harmful effect of nitrosation is inhibited by antioxidants which are also
in those vegetables (6). Still, a potential risk of a deleterious effect exists
if there is an inappropriately storage (58). Contamination of food or bev-
erage by nitrate-reducing bacteria leads to NO2- accumulation over time
(58). However, the expert consensus refers that NO3- supplementation
with BJ is very unlikely to be harmful (6). Therefore, current data does
not provide strong evidence supporting NO3- restriction (37). Athletes
should be advised to use vegetables sources if they intend to use dietary
NO3- supplementation as an ergogenic aid. Also, as far as kidney failure
is concerned, 450 mg of potassium nitrate per day for seven days did not
induce any detrimental effect in the kidney (60). Even so, it is not known
if longer-term intake of high-nitrate vegetable products may be danger-
ous for health and further long-term studies are necessary.

Practical considerations for nitrate supplementation

Another question that it is important to be aware is that NO3- is con-
verted to NO2- by commensal bacteria on the surface of the tongue (6)
and that there are some factors that can affect these bacteria. Rising the
mouth with the antibacterial mouthwash prior to the NO3- supplemen-
tation has no effect on NO3- accumulation in saliva or plasma but sup-

206
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

presses its conversion to NO2- in saliva and attenuates the NO2- plasma
rise (61). This happens because the antibacterial mouthwash removes
the commensal bacteria and constricts the NO-dependent biological ef-
fects of dietary NO3-. This theory was confirmed by a study (62) which
concluded that a twice daily use of an antibacterial mouthwash for three
days decreased oral NO3- reduction with a concomitant increase in sys-
tolic blood pressure. Regardless the fact that this might assure the safety
of NO3- supplementation, because nitrate in not converted to nitrite, it
will inhibit any ergogenic potential of nitrate.
It is important to note that none of these studies were performed with
children or adolescents. Further research in this population is needed to
assess potential hazards linked to nitrate supplementation and possi-
ble ergogenic effects in young athletes. Also, the International Olympic
Committee advises to encourage a dietary education, which should em-
phasize good eating patterns to support health and sports participation
demands: a balanced intake of nutrition-dense carbohydrates, high-
quality protein and sufficient dietary calcium, iron and vitamin D (63).
Another consideration should be taken in what respects the carbohy-
drate amounts in beetroot juice used in studies. When placebo refers to
beetroot juice nitrate-depleted, conclusions are more trustworthy since
the amount of carbohydrates (and other components, except for NO3-)
is the same in both groups (placebo and intervention). However, when
placebo is water (64), the carbohydrate content of whole beetroot juice
might induce a slightly more pronounced ergogenic effect (65). This car-
bohydrate amount should be considered in the pre-event meal carbohy-
drates calculation.

Conclusion
There is not yet a completely defined protocol for NO3- amounts and
intake times, but 6 – 8 mmol of NO3- per day, ingested chronically (seven
days) is a good option to enhance exercise performance. Also, the same
amount ingested acutely, 2 h before the event, seems to result in an
equivalent improvement of athletic performance.
Further research is necessary to assess potential hazards linked to
NO3- supplementation protocols longer than seven days. It is impor-
tant to refer that the NO3- ergogenic potential can, actually, be achieved
through foodstuffs instead of supplements – increasing the intake of ni-
trate-rich vegetables, such as bok choy, rocket and cabbage (17). Dietary
NO3- supplementation through a diet rich in vegetables seems to result,
in a similar way to supplementation, in a lower oxygen demand and in a

207
 sp or t science

greater efficiency of energy production (52).

Real conditions in which NO3- may have some ergogenic proprieties
are still unknown and further research is required. Notwithstanding, it
is already known that dietary NO3- supplementation effects depend on
the subject, fitness status, food patterns, intensity and duration of the
exercise, dose and time of intake. Moreover, athletes should be advised
not to rise the mouth with the antibacterial mouthwash prior to the NO3-
supplementation.

References
1. Bailey SJ FJ. Dietary nitrate supplementation enhances muscle contractile
efficiency during knee-extensor exercise in humans. J Appl Physiol (1985).
2010.
2. Lansley K, Winyard P, Bailey S, Vanhatalo A, Wilkerson D, Blackwell J, et al.
Acute dietary nitrate supplementation improves cycling time trial perfor-
mance. Medicine and Science in Sports and Exercise. 2011.
3. Alvares TS, Conte-Junior CA, Silva JT, Paschoalin VMF. Acute L-Arginine
supplementation does not increase nitric oxide production in healthy sub-
jects. Nutrition and Metabolism. 2012.
4. Ferreira L, Behnke B. A toast to health and performance! Beetroot juice
lowers blood pressure and the O2 cost of exercise. Journal of Applied Phys-
iology (1985). 2011.
5. Jones AM, Vanhatalo A, Bailey SJ. Influence of dietary nitrate supplemen-
tation on exercise tolerance and performance. Nestle Nutrition Institute
workshop series. 2013;75:27-40.
6. Jones A. Dietary nitrate supplementation and exercise performance.
Sports Medicine. 2014.
7. Moncada S, Higgs A. The L-arginine-nitric oxide pathway. The New England
Journal of Medicine. 1993.
8. Totzeck M, Hendgen-Cotta U, Rammos C, Frommke L, Knackstedt C, Predel
H, et al. Higher endogenous nitrite levels are associated with superior exer-
cise capacity in highly trained athletes. Nitric Oxide. 2012.
9. Cermak N, Res P, Stinkens R, Lundberg J, Gibala M, LJ. VL. No improvement
in endurance performance after a single dose of beetroot juice. Interna-
tional Journal of Sports Nutrition and Exercise Metabolism. 2012.
10. Boorsma R, Whitfield J, Spriet L. Beetroot juice supplementation does not
improve performance of elite 1500-m runners. Medicine Science Sports Ex-
ercise. 2014.
11. Wilkerson D, Hayward G, Bailey S, Vanhatalo A, Blackwell J, Jones A. Influ-
ence of acute dietary nitrate supplementation on 50 mile time trial per-
formance in well-trained cyclists. European Journal of Applied Physiology.
2012.
12. Hord NG, Tang Y, Bryan NS. Food sources of nitrates and nitrites: the physi-
ologic context for potential health benefits. The American Journal of Clini-
cal Nutrition. 2009.
13. Bailey S, Blackwell J, Williams E, Vanhatalo A, Wylie L, Winyard P, et al. Two

208
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

weeks of watermelon juice supplementation improves nitric oxide bio-

availability but not endurance exercise performance in humans. Nitric Ox-
ide. 2016.
14. Shanely R, Nieman D, Perkins-Veazie P, Henson D, Meaney M, Knab A, et
al. Comparison of Watermelon and Carbohydrate Beverage on Exercise-
Induced Alterations in Systemic Inflammation, Immune Dysfunction, and
Plasma Antioxidant Capacity. Nutrients. 2016.
15. Meirelles C, Matsuura C. Acute supplementation of L-arginine neither af-
fects strength performance nor nitric oxide production. The Journal of
Sports Medicine and Physical Fitness. 2016.
16. Chowdhury A, Das A. Nitrate Accumulation and Vegetable Quality. Interna-
tional Journal of Science and Research (IJSR). 2013.
17. Jonvik K, Nyakayiru J, Van Dijk J, Wardenaar F, van Loon L, Verdijk L. Habit-
ual Dietary Nitrate Intake in Highly Trained Athletes. International Journal
of Sports Nutition and Exercise Metabolism. 2016.
18. Bescos R, Rodriguez F, Iglesias X, Ferrer M, Iborra E, Pons A. Acute adminis-
tration of inorganic nitrate reduces VO(2peak) in endurance athletes. Medi-
cine and Science in Sports and Exercise. 2011.
19. Cermak N, Gibala M, van Loon L. Nitrate supplementation’s improvement
of 10-km time-trial performance in trained cyclists. International Journal
of Sports Nutrition and Exercise Metabolism. 2012.
20. Hoon M, Jones A, Johnson N, Blackwell J, Broad E, Lundy B, et al. The ef-
fect of variable doses of inorganic nitrate-rich beetroot juice on simulated
2,000-m rowing performance in trained athletes. International Journal of
Sports Physiology and Performance. 2014.
21. Hoon M, Johnson N, Chapman P, Burke L. The Effect of Nitrate Supple-
mentation on Exercise Performance in Healthy Individuals: A Systematic
Review and Meta-Analysis. International Journal of Sport Nutrition and Ex-
ercise Metabolism. 2013.
22. Domínguez R, Cuenca E, Maté-Muñoz JL, García-Fernández P, Serra-Paya
N, Estevan MCL, et al. Effects of Beetroot Juice Supplementation on Cardi-
orespiratory Endurance in Athletes. A Systematic Review. Nutrients. 2017.
23. Lundberg JO, Weitzberg E, Gladwin MT. The nitrate-nitrite-nitric oxide path-
way in physiology and therapeutics. Nat Rev Drug Discov. 2008;7(2):156-67.
24. Bisnett T, Anzueto A, Andrade F, Rodney JG, Napier W, Levine S, et al. Effect
of Nitric Oxide Synthase Inhibitor on Diaphragmatic Function after Resis-
tive Loading. Comparative Biochemistry and Physiology Part A: Molecular
& Integrative Physiology. 1998.
25. Galler S, Hilber K, Gobesberger A. Effects of nitric oxide on force-generating
proteins of skeletal muscle. Pflugers Arch: European Jounal of Physiology.
1997.
26. Kobzik L, Reid M, Bredt D, Stamler J. Nitric oxide in skeletal muscle. Nature.
1994.
27. Reid M. Role of nitric oxide in skeletal muscle: synthesis, distribution and
functional importance. Acta Physiologica Scandinavica. 1998.
28. Reid M. Nitric oxide, reactive oxygen species, and skeletal muscle contrac-
tion. Medicine and Science in Sports and Exercise. 2001.
29. Fulford J, Winyard P, Vanhatalo A, Bailey S, Blackwell J, Jones A. Influence

209
 sp or t science

of dietary nitrate supplementation on human skeletal muscle metabolism

and force production during maximum voluntary contractions. Pflugers
Arch: European Journal of Physiology. 2013.
30. Haider G, Folland J. Nitrate supplementation enhances the contractile
properties of human skeletal muscle. Medicine and Science of Sports and
Exercise. 2014.
31. Mosher S, Sparks S, Williams E, Bentley D, Mc Naughton L. Ingestion of a
nitric oxide enhancing supplement improves resistance exercise perfor-
mance. The Journal of Strength and Conditioning Research. 2016.
32. Flanagan S, Looney D, Miller M, DuPont W, Pryor L, Creighton B, et al. The Ef-
fects of Nitrate-Rich Supplementation on Neuromuscular Efficiency during
Heavy Resistance Exercise. The Journal of the American College of Nutri-
tion. 2016.
33. Coggan A, Leibowitz J, Kadkhodayan A, Thomas D, Ramamurthy S, Spearie
C, et al. Effect of acute dietary nitrate intake on maximal knee extensor
speed and power in healthy men and women. Nitric Oxide. 2015.
34. Hopkins W, Hawley J, Burke L. Design and analysis of research on sport
performance enhancement. Medicine and Science of Sports and Exercise.
1999.
35. Hodges RR, Shatos MA, Tarko RS, Vrouvlianis J, Gu J, Dartt DA. Nitric oxide
and cGMP mediate α1D-adregenic receptor-stimulated protein secretion
and p42/p44 MAPK activation in rat lacrimal gland. Investigative Ophtal-
mology and Visual Science. 2006.
36. Maréchal G, Gailly P. Effects of nitric oxide on the contraction of skeletal
muscle. Cellular and Molecular Life Sciences. 1999.
37. Clements W, Lee S, Bloomer R. Nitrate ingestion: a review of the health and
physical performance effects. Nutrients. 2014.
38. Boutcher S. High-Intensity Intermittent Exercise and Fat Loss. Journal of
obesity. 2011.
39. Wylie LJ, Bailey SJ, Kelly J, Blackwell JR, Vanhatalo A, Jones AM. Influence
of beetroot juice supplementation on intermittent exercise performance.
European Journal of Applied Physiology. 2016.
40. Aucouturiera J, Boissièrea J, Pawlak-Chaoucha M, Cuvelierb G, Gamelina
F-X. Effect of dietary nitrate supplementation on tolerance to supramaxi-
mal intensity intermittent exercise. Nitric Oxide. 2015.
41. Wylie L, Mohr M, Krustrup P, Jackman S, Ermiotadis G, Kelly J, et al. Dietary
nitrate supplementation improves team sport-specific intense intermit-
tent exercise performance. European Journal of Applied Physiology. 2013.
42. Thompson C, Wylie L, Fulford J, Kelly J, Black M, ST M, et al. Dietary nitrate
improves sprint performance and cognitive function during prolonged in-
termittent exercise. European Journal of Applied Physiology. 2015.
43. Peeling P, Cox G, Bullock N, Burke L. Beetroot Juice Improves On-Water 500
M Time-Trial Performance, and Laboratory-Based Paddling Economy in
National and International-Level Kayak Athletes. International Journal of
Sports Nutrition and Exercise Metabolism. 2015.
44. Nyakayiru J, Jonvik KL, Trommelen J, Pinckaers PJM, Senden JM, van Loon
LJ, et al. Beetroot Juice Supplementation Improves High-Intensity Inter-
mittent Type Exercise Performance in Trained Soccer Players. Nutrients.

210
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

2017.
45. Bellar D, Hatchett A, Judge L, Breaux M, Marcus L. The relationship of aer-
obic capacity, anaerobic peak power and experience to performance in
CrossFit exercise. Biology of Sport. 2015.
46. Kramer SJ, Baur DA, Spicer MT, Vukovich MD, Ormsbee MJ. The effect of six
days of dietary nitrate supplementation on performance in trained Cross-
Fit athletes. Journal of the International Society of Sports Nutrition. 2015.
47. Porcelli S, Ramaglia M, Bellistri G, Pavei G, Pugliese L, Montorsi M, et al.
Aerobic Fitness Affects the Exercise Performance Responses to Nitrate
Supplementation. Medicine and Science of Sports and Exercise. 2015.
48. McConell G, Bradley S, Stephens T, Canny B, Kingwell B, Lee-Young R. Skel-
etal muscle nNOS mu protein content is increased by exercise training in
humans. American Journal of Physiology: Regulatory, Integrative and
Comparative Physiology. 2007.
49. Jonvik K, Nyakayiru J, van Loon L, Verdijk L. Can elite athletes benefit from
dietary nitrate supplementation? Journal of Applied Physiology (1985).
2015.
50. Jones AM, Ferguson SK, Bailey SJ, Vanhatalo A, Poole DC. Fiber Type-
Specific Effects of Dietary Nitrate. Exercise and Sport Sciences Reviews.
2016;44(2):53-60.
51. Webb AJ, Patel N, Loukogeorgakis S, Okorie M, Aboud Z, Misra S, et al.
Acute blood pressure lowering, vasoprotective and anti-platelet proper-
ties of dietary nitrate via bioconversion to nitrite. Hypertension. 2008.
52. Larsen F, Weitzberg E, Lundberg J, Ekblom B. Effects of dietary nitrate on
oxygen cost during exercise. Acta of Physiology (Oxf). 2007.
53. Zafeiridis A. The effects of dietary nitrate (beetroot juice) supplementation
on exercise performance: A review American Journal of Sports Science.
2014;2(4):97 - 110.
54. Vanhatalo A, Bailey SJ, Blackwell JR, DiMenna FJ, Pavey TG, Wilkerson
DP, et al. Acute and chronic effects of dietary nitrate supplementation on
blood pressure and the physiological responses to moderate-intensity and
incremental exercise. American journal of physiology Regulatory, integra-
tive and comparative physiology. 2010;299(4):R1121-31.
55. Loh Y, Jakszyn P, Luben R, Mulligan A, Mitrou P, Khaw K. N-Nitroso com-
pounds and cancer incidence: the European Prospective Investigation into
Cancer and Nutrition (EPIC)-Norfolk Study. American Journal of Clinical
Nutrition. 2011.
56. Butler A. Nitrites and nitrates in the human diet: Carcinogens or beneficial
hypotensive agents? Journal of Ethnopharmacology. 2015.
57. Alexander J, Cockburn A, Cravedi J-P, Dogliotti E, Domenico AD, Fink-Grem-
mels J, et al. Opinion of the Scientific Panel on Contaminants in the Food
chain on a request from the European Commission to perform a scientific
risk assessment on nitrate in vegetables. The EFSA Journal. 2008.
58. Lundberg J, Larsen F, Weitzberg E. Supplementation with nitrate and ni-
trite salts in exercise: a word of caution. Journal of Applied Physiology
(1985). 2011.
59. Spiegelhalder B, Eisenbrand G, Preussmann R. Influence of dietary nitrate
on nitrite content of human saliva: Possible relevance to in vivo formation

211
 sp or t science

of N-nitroso compounds. Food Cosmet Toxicol. 1976.

60. Carpentier A, Stragier S, Brejeon C, Poortmans J. Nitrate Supplementation,
Exercise, and Kidney Function: Are There Detrimental Effects? Medicine
and Science of Sports and Exercise. 2015.
61. Govoni M, Jansson E, Weitzberg E, Lundberg J. The increase in plasma ni-
trite after a dietary nitrate load is markedly attenuated by an antibacterial
mouthwash. Nitric Oxide. 2008.
62. Bondonno C, Liu A, Croft K, Considine M, Puddey I, Woodman R, et al. An-
tibacterial mouthwash blunts oral nitrate reduction and increases blood
pressure in treated hypertensive men and women. American Journal of
Hypertension. 2015.
63. Bergeron MF, Mountjoy M, Armstrong N, Chia M, Côté J, Emery CA, et al.
International Olympic Committee consensus statement on youth athletic
development. British Journal of Sports Medicine. 2015.
64. Hernández A, Schiffer T, Ivarsson N, Cheng A, Bruton J, Lundberg J, et al. Di-
etary nitrate increases tetanic [Ca2+]i and contractile force in mouse fast-
twitch muscle. The journal of physiology. 2012.
65. Poortmans JR, Gualano B, Carpentier A. Nitrate supplementation and hu-
man exercise performance: too much of a good thing? Current opinion in
clinical nutrition and metabolic care. 2015.

212
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Exercise periodization and Taleb’s antifragility

José Afonso1, Ivo Hilvoorde2, Niek Pot3, Alexandre Medeiros4 and
Len Almond5

In Memoriam
To our dearest Len Almond, who unfortunately has passed away af-
ter a long battle against cancer. This work would not have been possible
without you. More than everything, thank you for your ever-stimulating
presence and for constantly challenging us to bring about newer under-
standings of life and of sports sciences.

Introduction
Exercise periodization is a foundational concept of Training Theory
and Methodology, being used extensively in performance and health set-
tings (1, 2). A plethora of periodized models abound, but all share a set
of core principles: (i) establishing dates for the main competitions and/
or other relevant milestones; (ii) planning time slots for periods extend-
ing from a week to years; and (iii) elaborating load management with the
aim of achieving peak performance at the intended moments, while at
the same time avoiding states of overtraining (1-3). In sum, periodized
programs focus on managing load variation rationally in an attempt to
predict directions, magnitudes, and timings of intended adaptations.
However, performance is loosely defined and multidimensional in
many sports, which complicates the act of quantifying load (4). Moreo-
ver, sports form is constantly changing, making its evaluation and pre-
diction a complex matter (5). Many athletes perform very well in training
sessions and evaluations, yet fail astonishingly in competition (6). And
empirical research in highly-controlled performance settings such as
strength training is not consensual in showing that periodized programs
are superior to non-periodized programs (7). Overall, research has pro-
vided some support for varied training programs, but not necessarily for

1. Centre for Research, Formation, Innovation and Intervention in Sport. Fac-

ulty of Sport, University of Porto, Portugal
2. Faculteit Bewegingswetenschappen, Vrije Universiteit Amsterdam, Nether-
lands
3. Windesheim University of Applied Sciences, Netherlands
4. Institute of Physical Education and Sport, Federal University of Ceará, Brazil
5. School of Education, St Mary’s University, United Kingdom

213
 sp or t science

periodized training programs (8). A systematic review with meta-analy-

sis comparing different periodized programs suggested that novelty or
training variety, and not periodization per se, are important for stimulat-
ing further development of strength (9). Depending on the type of sport,
this is an important indicator for performance.
A recent systematic review (10) focused on conceptual and methodo-
logical features of research on periodized training programs. Forty-two
randomized or randomized controlled trials on periodized programs
with human subjects were included, and the results were disappointing.
First and foremost, the concepts of periodization and variation are being
using interchangeably, as non-periodized programs are being wrongly
equated with constant programs, i.e., programs that remain unchanged
in time (e.g., same exercises, number of sets, number of repetitions, ex-
ercise sequence, and so on). Furthermore, constant programs might use
progressive load (i.e., keep all variables constant, but increase load as
one progresses), possibly improving performance. Secondly, no predic-
tions concerning direction, magnitude, or timing of adaptations were
tested. An additional number of methodological problems, such as poor
control of confounding factors (e.g., nutrition, supplementation and
medication), severely limits the claims of research. A common argument
is that periodization might be more useful in individual sports, especially
those of more closed nature (i.e., less contextual interference, such as
swimming). However, this systematic review has shown that research
has not provided support for the utilization of periodized programs,
even in tightly-controlled scenarios as strength-training programs.
In this vein, Taleb’s concept of antifragility (11) presents a possible
alternate framework, a line of reasoning that deserves some testing. The
author defined three main types of systems: (i) fragile systems, whose
dynamics break down in the face of errors or unexpected events; (ii)
robust systems, whose dynamics are largely immune to errors or unex-
pected events; and (iii) antifragile systems, evolving entities that actual-
ly use errors and perturbations has a developmental factor in improving
their dynamics. Admittedly, dynamic systems theory has embraced the
concept of perturbation and its importance for systemic behaviour (12,
13). Notwithstanding, systems react very differently to perturbations; it
is only when errors and perturbations are used by the system to evolve
(i.e., to change) that a system becomes antifragile (11). As such, antifrag-
ile systems are a specific subset of complex systems.
Therefore, our aim is to provide a critique of sports periodization
making use of the concept of antifragility. Although this analysis is the-

214
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

oretically-driven, we hope to pinpoint novel ideas for research on the

topic. This paper does not aim at establishing any doctrine, instead pro-
poses that acritically accepted concepts are revisited and reanalysed us-
ing different lenses.

Antifragility and sports periodization

The majority of systems is moulded through dynamic exchang-
es with the environment in a permanent state of non-equilibrium; con-
sequently, systemic behaviour evolves over time (14, 15), and the more
complex a system, the less able we are of predicting its future behaviour
(16). Highly stable (i.e., robust) systems are very difficult to perturb, so
transitioning to a new behavioural state may be challenging (15). For ex-
ample, literature on motor learning has alerted practitioners to the risks
of highly automated and stereotyped movements, especially when the
environment where they perform is variable (17). Antifragile systems
benefit from and grow with errors, which become information-rich op-
portunities for development (11). In the language of motor learning, such
systems benefit from a certain degree of contextual interference (17).
Antifragility presupposes a delicate interplay between systemic be-
haviours and environment. Periodized approaches, however, attempt to
predetermine a sequential process beforehand (1-3). Even if linear rela-
tionships were predominant in response to training stimuli, the success
of a given athlete in a certain competition using a prearranged perio-
dized program would not be valid as an argument; in the end, we should
ask whether alternative methods and means could have produced even
better results (8). In fact, many athletes deliver great performances de-
spite deviating from the training plans, while others don’t achieve suc-
cess despite following the training plans (18). And the numerous cases
of unsuccess seldom make their way to widespread knowledge, yet they
must be acknowledged to verify the real effectiveness of any periodized
program (8).
Therefore, classical paradigms of stimulus-response are misapplied
to dynamic systems (19), as dose-response relationships and develop-
ment are non-linear in nature (11, 20). Non-linear relationships bring
about emergent behaviour (21), meaning that the end-result cannot be
predicted through a sum of the parts (22). Emergent behaviour implies
self-organized novelty (23), and novelty is not (fully) predictable from
previous information, although some aspects of it might be probabil-
istic stipulated (e.g., hysteresis) (12). Indeed, self-organization may be
defined as the system organizing itself, but without there being a ‘self’,

215
 sp or t science

no agent intentionally doing the organizing (13). In addition, as the be-

haviour of complex systems is exquisitely sensitive to initial conditions,
the smallest changes can result in ever-large changes over time (24), re-
sulting in an amplification of fluctuations with time (22, 23). And, as the
initial conditions of any given system can never be stipulated accurately
enough (15, 24), valid and accurate baseline measurements of any fea-
ture of an athlete are utopic, so reliability of predictions should be ques-
tioned (6). This is extended to ongoing conditions.
There have been some so-called ‘non-linear’ periodized approaches
to training (25). Contrarily to what their proponents state, these propos-
als are actually linear. Linear systems have some well-established math-
ematical properties: (i) the sum of the functions of each component is
equal to the function of the sum of the components [f(x) + f(y) = f(x + y)];
and (ii) multiplying the function of x by a scalar must equal the function
of multiplying x by a scalar [αf(x) = f(αx)]. When at least one of these con-
ditions is not respected, a function is termed non-linear. Indeed, linear
functions are but a specific case of non-linear functions (26). ‘Non-linear’
periodized approaches, however, expect that a given sum of processes
will result in a prearranged outcome in a predetermined period.
While systems operating far from equilibrium give rise to singu-
lar events, subverting statistical generalizations (27), empirical research
on periodization focus solely on average values (10). Non-linear func-
tions render the average irrelevant; the important is the stability sur-
rounding the average, i.e., the degree of dispersion or variation (11). But
even when the average is an appropriate factor, it is still not relevant nor
informative for each individual case (8, 28). Moreover, population distri-
butions can never be empirically known, so the form and the specifics of
each distribution are, in themselves, estimates (27). Taken together, this
invites scientists and practitioners to question the accuracy of predic-
tions based on momentary and necessarily limited evaluations.
Life is convoluted, yet our brains elaborate a coherent narrative,
smoothing out inconsistencies in a posteriori rationalization (29). Train-
ing is no exception, with the predictive ideal of science clashing with the
constraints and idiosyncrasies of daily practice. Postdictions create the
illusion that we comprehend the world (11) and restructure the past to fit
present beliefs, in an act of historical revisionism (30), and perhaps this
illusion of prediction satiates the human need for living a coherent nar-
rative. In this scope, and despite individual learning rates varying dras-
tically (31), periodized training plans establish timelines for rhythms of
learning (6). As paradigms are usually abandoned only a new paradigm

216
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

is available (32), here we will attempt to elaborate an alternative para-

digm to current periodized training programs.

Towards a new paradigm

The lack of scientific evidence in favour of periodized programs
(10) may be temporary. Better conceptually and methodologically de-
signed empirical research might justify its application. And even if the
lack of current evidence is coupled with theoretical reasons for doubting
the efficacy of periodization, this approach should not be discarded alto-
gether. Nonetheless, this does suggest that distinct approaches should
be experimented with. Indeed, alternatives to periodization should be
tested, expanding our knowledge and options instead of focusing in
a single paradigm. Thus, in what follows we will present two possible
alternate frameworks, theoretical proposals that will require empirical
testing.
A conservative line of action would follow a minimalist (i.e., periodize
to the less extent possible and to the smaller devisable time lengths) and
negative approach to periodization. The negative approach focuses on
enumerating what should not be done (11), and draws an important par-
allel with the work of Sir Karl Popper (33) in socio-politics. In this mini-
malist and negative approach, periodized programs would be reduced
to the basics, such as avoid having many consecutive weeks of training
with extremely high volume and intensity, attempting to escape over-
training (34). Perhaps they could also establish strategic rest periods
during the year, usually associated with pauses in the competition and/
or culturally-specific holidays (e.g., the 4th of July is the Independence
Day in the U.S.A., but in many countries, it is merely a regular day), and
perhaps delivering effects similar to those associated with tapering (35).
This minimalist and negative approach would still be a form of periodi-
zation, albeit a ‘reduced’ form.
Even so, the specifics of load management and arrangement would
remain open to ongoing evaluation of the process, fully recognizing that
two persons will respond differently to the same stimulus, and that the
same person will respond differently to the same stimulus in different
moments (34). What is more, these inter- and intra-individual variations
in response cannot be predicted in advance (22). Therefore, perhaps a
more drastic confrontation with the paradigm of periodized training pro-
grams would be to abandon them altogether. The hard truth is that the
world is too random and unpredictable to establish precise plans based
on foreseeing the future (11, 24). The evolution of athletes is a non-linear

217
 sp or t science

process, filled with surprises and discontinuities (36). And random and
unexpected incidents must come to be expected in sports performance
(16).
Still, periodized programs provide a (false) sense of stability and con-
trol over the process (8). Perhaps such illusion is necessary; it may be a
necessary condition to thrive in complex, uncertain environments, even
if through self-fulfilling prophecy (37). However, it is when we believe to
be experiencing a relatively stable period that insidious vulnerabilities
tend to accumulate and generate problems (11). So perhaps we should
focus on exposition to failure, making prediction irrelevant. Practicing
alternative gameplays, for example, might provide a team with more
resources to deliver a solid performance in varying contexts and con-
ditions. Even non-competitive settings may appeal to such logic. Con-
temporary dance also promotes dealing with the unexpected instead of
anticipating actions, exploring momentary constraints and affordances
and incorporating them into the dancers’ movements (38). In fact, elite
performers vary widely in their executions, and the same person per-
forms differently as time passes (39).
Therefore, the determination of the contents that will be taught or
coached should not be defined a priori, but elaborated based on contin-
uous observation (40). Contrarily to periodized approaches, non-perio-
dized (but varied) approaches procrastinate. Therefore, when decisions
are effected, they are temporally closer to the action. Thus, by not having
such a rigid, pre-stipulated framework, we are better prepared to grasp
and explore emerging opportunities (8, 11). Postponing our planning de-
cisions is more compatible with promoting the emergence of innovative
processes and solutions in sports performance, as functional solutions
can only emerge following exploration and discovery (41). Although we
cannot predict which events will impact our coaching processes, it is
possible to establish options, which largely reduces the need to rely on
anticipatory evaluations (11). The focus would be on preparing players
for unpredictability (42), as learning to improve our capacity to correct
course might be more relevant and decisive than predicting the course.
Even in sports such as weightlifting or cycling, significant modifications
of biomechanics in function of how an athlete will respond to training
are warranted, and some such alterations cannot be determined in ad-
vance.
In this vein, athletes should develop a degree of redundancy, a funda-
mental risk-managing property in natural systems (11). Adaption to exer-
cise actualizes the principle of redundancy, usually known as overcom-

218
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

pensation (34): after bouts of acute exercise, the body surpasses its initial
state, thereby being better prepared for future aggressions. Conversely,
detraining produces the opposite effect. Thus, the body manages the de-
gree of redundancy required depending on the regularity and intensity
of the aggression it has to cope with. By developing redundancy (e.g.,
learning additional technical resources; exploring alternative tactical
structures) the system becomes better prepared to face a more varied
array of possible events. Degeneracy (i.e., the ability of structurally dif-
ferent elements in performing the same function) should be incorporat-
ed in our mindset, acknowledging the existence of several performance
solutions (41, 43), recognizing that multiple degrees of freedom enable
performers to complete tasks in various ways (15). This renders an opti-
mal, idealized solution a utopic golem (8, 44).

Final considerations
Frequently, depictions of coaching and training highlight the
planned, coordinated, rationalized part of the activity (36, 45). Periodi-
zation is a very human creation that attempts to divide a sports season
into supposedly rational phases and cycles (18). The belief that it is pos-
sible to delineate time periods and manage responses to pre-stipulated
load curves (2) provides a sense of power and control over the process,
and the sheer repetition of an idea has a remarkable ability to get us
to accept it (46). But analyses concerning peak performance in selected
sports (including Track & Field) have shown that periodized approaches
present very low rates of effectiveness (3). The works of Carpinelli and
colleagues (7), Kiely (8) and, more recently, Afonso and colleagues (10)
raised further doubts concerning the trustworthiness and scope of re-
search on periodized training programs.
Compartmentalisation of human performance into basic and spe-
cific capacities and regulation of overload and supercompensation in
specific timings of the training cycle are not grounded on physiological
evidence (3, 8). Furthermore, our inability to foresee the future, to deal
with probabilities, to calculate chain events, secondary effects and unin-
tended consequences (26, 30), can result in an iatrogenic effect, provok-
ing more harm than good (11). A very interesting discussion of the prob-
lems with predicting performance when applied to talent detection and
selection is presented by Ackerman (31). Paradoxically, by attempting to
pre-establish the process to a large extent and make predictions based
on theoretical models, periodization may increase a system’s fragility,
increasing the risk of exposition to unexpected events.

219
 sp or t science

A broad, loosely-defined planning might be best suited to a dynamic,

evolving process, in comparison to a periodized approach. This would
allow effectively respecting the principle of individualization. Inter- and
intra-individual variation in response to external loads find strong sup-
port in the literature (4, 34, 47). What suits a given athlete might be
prejudicial to another, and the same subject will respond distinctively
in different moments (5, 8, 22, 31). There are non-responders to different
types of resistance training (48), altitude training (49), cardiac resynchro-
nization therapy (50), pulmonary rehabilitation (51), among others.
Pre-established programs are always delusive, especially when lit-
erature on expertise and career planning have shown that many elite
athletes have failed talent detection programmes and performed poorly
on standardized physical batteries (17, 52). Commenting on Long-Term
Athlete Development (LTAD), the National Strength and Conditioning As-
sociation stated that LTAD pathways should accommodate for the highly
individualized and non-linear nature of development, recognizing the
existence of significant interindividual variance regarding magnitude of
change, timing (onset of change), and tempo (rate of change) of biologi-
cal processes, therefore impacting profoundly on each athlete’s train-
ability (53). Overall, generic methodologies are likely to fail (8).
Mainstream thinking, however, is still hostage to the ideal of Quéte-
let’s homme moyen and, in sports, to the ideal of the optimal movement
pattern (42, 47). Worldviews are frameworks that present a vision of the
nature of the world (32), and the practices motivated by such world-
views tend to crystalize and become discursive ‘truths’ (45). Periodiza-
tion relies on the worldview of a predictable world, where complexity
and unpredictability can somehow be encapsulated into pre-planned
frameworks (8). This worldview is common to economic models, despite
evidence repeatedly showing these models fail amazingly (26, 54). Peer
pressure and cultural inertia weigh considerably, and novel ideas, espe-
cially if breaking away with established paradigms, threaten our deeply
engrained beliefs (41). Notwithstanding, it is only when individuals
break with current thinking in a given field that new ideas and concepts
can flourish and bring about a new status quo (55).
In this context, we believe that Taleb’s concept of an antifragile sys-
tem and its implications for developmental processes should be incor-
porated into training theory and methodology. Perhaps it is time to
acknowledge the full implications of the ambiguous and unpredictable
nature of coaching (56), namely that performance and learning environ-
ments can rarely be planned and prescribed in advance (21). We should

220
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

shift the focus from prediction to preparation (26), using open-ended

and highly-flexible approaches to planning (6, 57), i.e., use simple, sug-
gestive and flexible plans and modify them according to the athlete’s ac-
tual progression (58). Planning should be a generic perspective, a letter
of intentions without venturing into details, an outline of basic starting
points, checkpoints and endpoints with only a sparse skeleton (8). The
coach then orchestrates the process as it unfolds (57). The training ses-
sion or the class unit are the fundamental entities of the process (58).

References
1. Issurin VB. Block periodization versus traditional training theory: A review.
. Journal of Sports Medicine and Physical Fitness. 2008;48:65-75.
2. Naclerio F, Moody J, Chapman M. Applied periodization: a methodological
approach. Journal of Human Sport and Exercise. 2013;8:350-66.
3. Loturco I, Nakamura F. Training periodisation. An obsolete methodology?
Aspetar Sports Medicine Journal. 2016;5:110-5.
4. Mujika I. Challenges of team-sport research. International Journal of
Sports Physiology and Performance. 2007;2:221-2.
5. Lames M. Computer science for top-level team sports. International Jour-
nal of Computer Science in Sport. 2003;2:57-72.
6. Denison J, Mills J. Planning for distance running: coaching with Foucault.
Sports Coaching Review. 2014;3:1-16.
7. Carpinelli RN, Otto RM, Winett RA. A critical analysis of the ACSM Position
Stand on resistance training: insufficient evidence to support recommend-
ed training protocols. Journal of Exercise Physiology Online. 2004;7:1-60.
8. Kiely J. Periodization paradigms in the 21st century: evidence-led or tradi-
tion-driven? International Journal of Sports Physiology and Performance.
2012;7:242-50.
9. Harries S, Lubans D, Callister R. Systematic review and meta-analysis of
linear and undulating periodized resistance training programs on muscu-
lar strength. Journal of Strength and Conditioning Research. 2015;29:1113-
25.
10. Afonso J, Nikolaidis PT, Sousa P, Mesquita I. Is empirical research on perio-
dization trustworthy? A comprehensive review of conceptual and meth-
odological issues. Journal of Sports Science and Medicine. 2017;16:27-34.
11. Taleb NN. Antifragile: Things that gain from disorder: Nassim Nicholas Ta-
leb; 2012.
12. Haken H. Synergetics: Nonequilibrium phase transitions in Physics, Chem-
istry and Biology. Berlin: Springer; 1977/83.
13. Kelso JAS. Dynamic patterns. The self-organization of brain and behavior:
Mit Press; 1995.
14. McGarry T, Anderson D, Wallace S, Hughes MD, Franks IM. Sport compe-
tition as a dynamical self-organizing system. Journal of Sports Sciences.
2002;20:771-81.
15. Van Emmerik REA, Ducharme SW, Amado AC, Hamill J. Comparing dynami-
cal systems concepts and techniques for biomechanical analysis. Journal

221
 sp or t science

of Sport and Health Science. 2016;5:3-13.

16. Aicinena S. The impact of chaos, complexity, and luck on coaching suc-
cess. International Journal of Society, Science & Education. 2013;3:551-65.
17. Schmidt R, Wrisberg C. Motor learning and performance. A situation-based
learning approach (4th Ed.). Champaign, Illinois: Human Kinetics; 2008.
18. Denison J, Scott-Thomas D. Michel Foucault: Power and discourse: the
‘loaded’ language of coaching. In: Jones R, Potrac P, Cushion C, Ronglan
LT, editors. The sociology of sports coaching. New York: Routledge; 2011.
p. 27-39.
19. McGarry T, Franks IM. System approach to games and competitive playing:
reply to Lebed (2006). European Journal of Sport Science. 2007;7:47-53.
20. Overton W. Taking conceptual analyses seriously. Research in Human De-
velopment. 2015;12:163-71.
21. Davids K. Learning design for nonlinear dynamical movement systems.
The Open Sports Sciences Journal. 2012;5:9-16.
22. Rowland T. The athlete’s clock. How biology and time affect sport perfor-
mance. Champaign, Illinois: Human Kinetics; 2011.
23. Latash M. Neurophysiological basis of movement (2nd Ed.). Champaign, Il-
linois: Human Kinetics; 2008.
24. Saltzstein P. Chaos & An Unpredictable Tomorrow. Philosophy Now
2016;114.
25. Hasegawa H, Dziados J, Newton R, Fry A, Kraemer W, Häkkinen K. Perio-
dized training programs for athletes. In: Kraemer W, Häkkinen K, editors.
Strength Training for Sport: Wiley-Blackwell; 2002. p. 69-134.
26. Taleb NN. The Black Swan: The impact of the highly improbable: Random
House; 2006.
27. Hendry D, Mizon G. Unpredictability in economic analysis, econometric
modeling and forecasting. Journal of Econometrics. 2014;182:186-95.
28. Gigerenzer G, Brighton H. Homo Heuristicus: why biased minds make bet-
ter inferences. Topics in Cognitive Science. 2009;1:107-43.
29. Damásio A. Descartes’ Error: Emotion, reason, and the human brain: Put-
nam Publishing; 1994.
30. Schön D. Educating the reflective practitioner. Toward a new design for
teaching and learning in the professions. San Francisco: Jossey-Bass Pub-
lishers; 1987.
31. Ackerman P. Nonsense, common sense, and science of expert perfor-
mance: Talent and individual differences. Intelligence. 2014;45:6-17.
32. Overton W. The process-relational paradigm and relational-developmen-
tal-systems metamodel as context. Research in Human Development.
2014;11:323-31.
33. Popper K. The open society and its enemies, Vol. 1: The spell of Plato (5th
Rev Edition: Princeton University Press; 1971.
34. Kenney WL, Wilmore JH, Costill DL. Physiology of Sport and Exercise (5th
Ed.). Champaign, Illinois: Human Kinetics; 2012.
35. Pyne D, Mujika I, Reilly T. Peaking for optimal performance. Research limi-
tations and future directions. Journal of Sports Sciences. 2009;27:195-202.
36. Bowes I, Jones R. Working at the edge of chaos: understanding coaching as
a complex, interpersonal system. The Sport Psychologist. 2006;20:235-45.

222
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

37. Merton RK. The self-fulfilling prophecy. Antioch Review. 1948;8:193-210.

38. Louppe L. Poetic of contemporary dance. [Portuguese original title: Poé-
tica da dança contemporânea]. Lisboa: Orfeu Negro; 2012.
39. Bailey R, Pickard A. Body learning: examining the processes of skill learn-
ing in dance. Sport, Education & Society. 2010;15:367-82.
40. Gréhaigne J-F, Wallian N, Godbout P. Tactical-decision learning model and
students’ practices. Physical Education and Sport Pedagogy. 2005;10:255-
69.
41. Renshaw I, Araújo D, Button C, Chow JY, Davids K, Moy B. Why the Con-
straints-Led Approach is not Teaching Games for Understanding: A clarifi-
cation. Physical Education and Sport Pedagogy. 2016;21:459-80.
42. Passos P, Araújo D, Davids K, Shuttleworth R. Manipulating constraints to
train decision making in Rugby Union. International Journal of Sports Sci-
ence & Coaching. 2008;3:125-40.
43. Glazier P, Davids K. Constraints on the complete optimization of human
motion. Sports Medicine. 2009;39:15-28.
44. Lederman E. The fall of the postural-structural-biomechanical model in
manual and physical therapies: Exemplified by lower back pain. Journal of
Bodywork and Movement Therapies. 2011;15:131-8.
45. Denison J. Planning, practice and performance: the discursive formation
of coaches’ knowledge. Sport, Education & Society. 2010;15:461-78.
46. Varki A, Brower D. Denial. Self-deception, false beliefs, and the origins of
the human mind. New York: Twelve; 2013.
47. Davids K, Button C, Bennett S. Dynamics of skill acquisition. A constraints-
led approach. Champaign, Illinois: Human Kinetics; 2008.
48. Jones N, Kiely J, Suraci B, Collins DJ, de Lorenzo D, Pickering C, et al. A
genetic-based algorithm for personalized resistance training. Biology of
Sport. 2016;33:117-26.
49. Hamlin M, Manimmanakorn A, Sandercock G, Ross J, Creasy R, Hellemans
J. Heart rate variability in responders and non-responders to live-moder-
ate, train-low altitude training. World Academy of Science, Engineering
and Technology. 2011;77:936-40.
50. Auricchio A, Prinzen F. Non-responders to cardiac resynchronization ther-
apy. The magnitude of the problem and the issues. Circulation Journal.
2011;75:521-7.
51. Stoilkova-Hartmann A, Janssen D, Franssen F, Wouters EF. Differences in
change in coping styles between good responders, moderate responders
and non-responders to pulmonary rehabilitation. Respiratory Medicine.
2015;109:1540-5.
52. Pinder R, Renshaw I, Davids K. The role of a representative design in tal-
ent development: a comment on ‘Talent identification and promotion pro-
grammes of Olympic athletes’. Journal of Sports Sciences. 2013;31:803-6.
53. Lloyd RS, Cronin JB, Faigenbaum AD, Haff GG, Howard R, Kraemer WJ, et
al. National Strength and Conditioning Association position statement on
long-term athletic development. Journal of Strength and Conditioning Re-
search. 2016;30:1491-509.
54. Krugman P. The return of depression economics and the crisis of 2008: W.
W. Norton & Company; 2009.

223
 sp or t science

55. Olszewski-Kubilius P, Subotnik R, Worrell F. Antecedent and concurrent

psychosocial skills that support high levels of achievement within talent
domains. High Ability Studies. 2015;26:195-210.
56. Barker E, Bailey J. There’s more to coaching than the context: a Bourdieu-
sian account of an embodied athlete. Sports Coaching Review. 2015;4:41-7.
57. Jones R, Wallace M. Another bad day at the training ground: coping with
ambiguity in the coaching context. Sport, Education & Society. 2005;10:119-
34.
58. Bompa T. Periodization. Theory and methodology of training (4th Ed.).
Champaign, Illinois: Human Kinetics; 1999.

224
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Effects of different rest intervals

between potentiation exercises on sprint
performance in trained soccer players
Vasiliki Manou1, Vasilis Minos1, Athanasios A. Dalamitros1

Introduction
A variety of different training methods and techniques have been
used by strength and conditioning practitioners and sports scientists to
enhance sports performance. Towards this direction, the post-activation
potentiation phenomenon (PAP) (Robbins & Docherty, 2005) has drawn
increasing attention as a relatively new concept. To exploit this phenom-
enon during sports training, the complex method is used (Docherty et al.,
2004). This training method refers to the condition where the execution
of a previous high-intensity activity may induce short-term performance
during explosive athletic movements such as jumps, sprints and throws.
The proposed mechanisms responsible for PAP have previously been
suggested (Tillin & Bishop, 2009), while the response magnitude of PAP
is depended on multiple variables such as gender and chronological age
(Arabatzi et al., 2014), genetics (Hamada et al., 2000), as well as train-
ing characteristics and methodological issues (exercise volume and in-
tensity, recovery interval, movement and muscle contraction type) (Es-
formes & Bambouras, 2011; Wilson et al., 2013). Training status has been
also denoted as an important parameter affecting PAP responses, while
stronger or power trained individuals seem to be benefitting the most
(Seitz et al., 2014). In addition, PAP effect is larger for sprint performance,
compared to jumping and throwing activities (Seitz & Haff, 2015). Based
on the above observations, PAP is presented as an individualized and
multifaceted phenomenon.
The efficacy of PAP has been examined in numerous sports such as
rowing (Doma et al., 2016), swimming (Sarramian et al., 2015), in track
and field athletes (Ountzoudi et al., 2014), basketball (Tsimachidis et al.,
2013) and rugby (Bevan et al., 2009). Soccer is an activity where PAP has
a practical application, as explosive activities (e.g. sprints) are of great
importance. High-level soccer players spend approximately 11% of their
game in sprinting activities, while the short duration sprints (< 5 s) ac-

1. Faculty of Physical Education & Sport Sciences, School of Physical Education

& Sport Sciences, Aristotle University of Thessaloniki, Greece

225
 sp or t science

count to 90% of the overall sprints (Andrzejewski et al., 2013). Thus, it

is not surprising that acute (Low et al., 2015; McBride et al., 2005; Rah-
man, 2007) and longitudinal studies (Brito et al., 2014; Maio Alves et al.,
2010) have been focused on the PAP, or complex training method effect
on sprint performance in distances varying from 5 to 40 m in soccer play-
ers of different training levels demonstrating improved running times.
In all these cases, the preceding activity was the heavy load back squat
exercise with intensities equal or above 85% of the athletes’ one repeti-
tion maximum (1RM).
As earlier mentioned, recovery interval times consist a major deter-
minant for PAP responses. A previous study that examined six different
recovery times (15s, 4, 6, 8, 12 and 20 minutes) during a potentiation
protocol consisting of a 3RM squat exercise, failed to provide a specific
guideline regarding the optimal recovery period for vertical jumping per-
formance (peak power and jump height) in professional soccer players
(Mola et al., 2014). However, most of the responders demonstrated an
optimal PAP “window” at the 4 min recovery time. The same interval pe-
riod (4 minutes) was also prescribed in the study of Rahimi (2014), pre-
senting improved 40 m sprint times in elite soccer players after a loaded
(85% of 1RM) squat exercise. From a practical point of view when work-
ing in a team setting, long recovery times between sets (e.g., 6 to 20 min-
utes) are difficult to be applied, since during a conditioning program the
total exercise time generally does not exceed a 20 min time frame.
Interestingly, when analysing the methodological approaches regard-
ing the potentiation protocols applied in studies using soccer players, a
differentiation in the number of exercises preceding the “goal” activity
is acknowledged. More analytically, a combination of strength (loaded
squat exercise) and plyometric exercises (unloaded vertical jumps) was
used in the studies examining the effects of a training program, while a
single activity was implemented (loaded squat or loaded vertical jump-
ing exercise) during the acute-type studies.
Nevertheless, the training adaptations previously reported could be
solely attributed to the complex training intervention program applied
in these studies. Hence, the necessity for examining the acute effects
of a potentiation protocol including both strength and plyometric ex-
ercises has raised, since the possible fatigue manifestation is a critical
parameter for the subsequent activity (DeRenne, 2010). Moreover, to
the author’s knowledge, no study to date has investigated the effect of
different rest interval approaches during a potentiation activity and a
plyometric type exercise on the subsequent running sprint performance

226
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

of various distances. Therefore, the aim of this study was to analyse the
acute effect of a potentiation protocol including a heavy load back squat
exercise followed by an unloaded countermovement exercise, executed
with three different rest interval strategies on the subsequent accelera-
tion time (5 m), and sprint performance (20 - 30 m) in semi-professional
male soccer players.

Methods
Participants
Forty-five adult male soccer players (age 22.2 ± 4.4 yrs, height 175.7
± 6.9 cm, body mass 69.9 ± 6.6 kg, training experience 9.4 ± 2.6 yrs, 1RM
back squat 97.2 ± 9.6 kg) from 2 different clubs were recruited during
their competition phase and volunteered to participate in this study.
They were all engaged in a training program 5 times per week, including
1 session of resistance training and had a minimum 2 years of experience
in performing back squats. In addition, they were familiar with vertical
jumping exercises. All athletes had no current injuries, while the exclu-
sion criteria included lower body myoskeletal pain within the preceding
two-week time frame. In addition, they were requested to abstain from
alcohol consumption for 24 hours, and to avoid any strenuous exercise
the day before testing. All participants provided informed consent with
all procedures approved by the Institutional Human Research Ethics
Committee and in accordance with the Declaration of Helsinki.
Procedures and measures
Participants completed a 3-day testing procedure. Before the ini-
tiation of this period, a baseline sprint test including the assessment of
the 5 m acceleration time (S5), the maximal 20 m (S20) and 30 m sprint
times (S30) was performed (BASE). Two trials were allowed for each
athlete, while the single highest score for the S30 was used for further
analysis. During the same occasion, a familiarization session was com-
pleted. The 1RM determination was conducted according to the NSCA
guidelines (Baechle & Earle, 2008) on a different occasion. As the depth
of the squat exercise may influence the potentiation and the subsequent
performance (Esformes & Babouras, 2013), a member of the research
personnel was responsible for controlling a 90º knee angle and ensur-
ing the proper technique. During the countermovement activity (CMJ)
participants executed 8 maximal vertical jumps at a self-selected depth,
starting from an upright position, with their hands on the waist. Sprint
performance was evaluated in a straight 30 m line, in an artificial turf
ground. Four pairs of photoelectric cells (STC3/ACCO58) with an elec-

227
 sp or t science

tronic timer (Microgate Race Time 2, Italy) were placed 70 cm from the
ground level on the starting line, at 5, 20 and 30 m. Each athlete began
with the front foot placed 30 cm behind the starting line with a stand-
ing start (Ramírez-Campillo et al., 2016). The three testing sessions were
performed at the same time of the day (early afternoon) with one week
apart and were randomized using a counterbalanced design. Finally, en-
vironmental conditions were almost similar during the testing sessions.
After an identical warm-up (10 min including low intensity run-
ning, dynamic stretches, slow bodyweight back-squats, fast CMJ’s and
sprints), and following a 2 min rest, the athletes executed one of the
three different potentiation protocols. During the first protocol (EXP1),
athletes executed the 6RM back squat exercise (85% of their 1RM) and
the CMJ exercise, while a 4 min rest was allowed between the two activi-
ties. In this case, the subsequent 30 m sprint test was executed exactly
after the plyometric exercise. The second potentiation strategy (EXP2) in-
cluded the same exercise order, with no rest interval between the heavy
load activity and the plyometric exercise. After an equal rest interval (4
min), the 30 m sprint test was performed. Finally, during the third pro-
tocol (EXP3), two passive rest intervals of equal duration (2 minutes)
were placed after the back-squat activity and the CMJ exercise, before
the sprint assessment. Strong verbal encouragement was given during
all tests, while participants were instructed to wear the same shoes dur-
ing the 3 testing occasions. The experimental design is demonstrated in
Figure 1.

Figure. 1 Experimental design of the study.

228
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Statistical Analysis
A one-way repeated measures analysis of variance (ANOVA) was used
to examine differences between the baseline test and the potentiation
protocols. Paired samples t-tests with a Bonferroni’s correction were
used to analyse between conditions differences. Cohen’s d effect sizes
(d = difference between means¸ pooled SD) were calculated for the dif-
ference between means. The small, medium, and large effects were re-
flected in values greater than 0.20, 0.50, 0.80, respectively (Cohen, 1988).
All statistical analyses were performed using SPSS 22.0 (IBM, New York,
NY, USA). Statistical significance was set at p ≤ 0.05. Data are expressed
as means ± SD values.

Results
ANOVA with repeated measures revealed a significant condition effect
at S5 (F3,132 = 8.39, p ≤ 0.001), S20 (F3,132 = 22.31, p ≤ 0.001) and S30 times
(F3,132 = 31.27, p ≤ 0.001). Values were faster during EXP3 protocol, whereas
all sprint tests were slower during EXP1 protocol (Table 1).

BASE
BASE BASE vs.
BASE EXP1 EXP2 EXP3 vs.
vs. EXP1 EXP2
EXP3
S5 1.008 ± 1.034 ± 1.007 ± 0.991 ± ES = ES = 0.00 ES =
0.072 0.078 0.641 0.064 0.35 -0.24
S20 3.001 ± 3.033 ± 2.977 ± 2.945 ± ES = ES = ES =
0.126 0.137 0.124 0.115 0.24 -0.19 -0.41
S30 4.192 ± 4.228 ± 4.168 ± 4.119 ± ES = ES = ES =
0.160 0.175 0.149 0.156 0.21 -0.16 -0.48

Table 1. Mean ± SD values of the acceleration (5 m) and maximal 20 and 30 m

sprint times (s) during the baseline test and after the three different poten-
tiation protocols. Effect size (ES) is also presented (n=45).

BASE: Best 30 m sprint test during the familiarization session;

EXP1: 4 min rest between Squat and CMJ; EXP2: 4 min rest after CMJ;
EXP3: 2 min rest after Squat and CMJ; S5 = Acceleration time 5 m; S20
= Sprint time at 20 m; S30 = Sprint time at 30 m.

Paired samples t-tests revealed a significant statistical difference be-

tween almost all tested conditions (p = 0.000 to 0.036), while a small to
medium effect size was observed between testing conditions (≤ 0.50). As
an, exception BASE and EXP2 values for S5 test were almost identical (p =

229
 sp or t science

0.991) (Figures 2, 3 and 4).

Figure 2. Results for acceleration time for all testing conditions; *Significantly
different from BASE at p value; # Significantly different from EXP1 at p val-
ue; ^ Significantly different from EXP2 at p value; ‡ Significantly different
from EXP3 at p value; BASE: Best 30 m sprint test during the familiarization
session; EXP1: 4 min rest between Squat and CMJ; EXP2: 4 min rest after
CMJ; EXP3: 2 min rest after Squat and CMJ; S5 = Acceleration time 5 m; S20
= Sprint time at 20 m; S30 = Sprint time at 30 m.

Figure 3. Results for maximal 20 sprint time for all testing conditions; *Signifi-
cantly different from BASE at p value; # Significantly different from EXP1 at
p value; ^ Significantly different from EXP2 at p value; ‡ Significantly differ-
ent from EXP3 at p value; BASE: Best 30 m sprint test during the familiari-
zation session; EXP1: 4 min rest between Squat and CMJ; EXP2: 4 min rest
after CMJ; EXP3: 2 min rest after Squat and CMJ. S5 = Acceleration time 5 m;
S20 = Sprint time at 20 m; S30 = Sprint time at 30 m, as these info are miss-
ing from this note (Check Figures 1, 2 and 4).

230
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Figure 4. Results for maximal 30 sprint time for all testing conditions; *Signifi-
cantly different from BASE at p value; # Significantly different from EXP1 at
p value; ^ Significantly different from EXP2 at p value; ‡ Significantly differ-
ent from EXP3 at p value; BASE: Best 30 m sprint test during the familiari-
zation session; EXP1: 4 min rest between Squat and CMJ; EXP2: 4 min rest
after CMJ; EXP3: 2 min rest after Squat and CMJ. S5 = Acceleration time 5 m;
S20 = Sprint time at 20 m; S30 = Sprint time at 30 m.

Discussion
Post-activation potentiation phenomenon has gained considera-
ble research interest, whereas its practical application is commonly used
to optimize performance in various sports activities. This study aimed to
analyze the effect of three different rest interval approaches during a po-
tentiation protocol, including both strength and plyometric activities, on
the subsequent acceleration and maximal sprint time in trained soccer
players. According to the results, EXP3 was presented as the most effec-
tive strategy as an acute significant reduction in S5, S20 and S30 perfor-
mance times was observed, compared to BASE test, as well as the rest
of the potentiation protocols applied. S20 and S30 m times were also
significantly improved after executing EXP2 strategy, compared to BASE
test. Interestingly, EXP1 potentiation protocol concluded to significantly
slower times during all distances evaluated compared to the rest of the
potentiation protocols and BASE test. Finally, S5 times were almost iden-
tical between BASE and EXP2 approaches.
Despite not being a biomechanically similar movement pattern, load-
ed back squat strength values are closely related to sprint performance
during 10 and 30 m distances in soccer players (Wisløff et al., 2004), since

231
 sp or t science

the same muscle groups contributing to running sprint are activated

(Newman et al., 2004). This exercise is regularly used as a potentiation
activity to acutely enhance sprint performance in soccer demonstrating
a positive effect in sprint distances up to 40 m (Low et al., 2015; McBride
et al., 2005; Rahimi, 2007). 1RM mean squat value in our study (97.2 kg)
was somewhat similar to the squat load denoted by Mola et al. (2014)
(~100 kg), but also significantly lower compared to a previous report
(171.7 kg) (Wisløff et al., 2004) in a sample of professional soccer players.
According to Jo and colleagues (2010), the respective values for recrea-
tional soccer players were 83.6 kg. Considering these values, we could
describe our participants’ strength level as “moderate”.
The rest interval implemented in the current study (4 min) could not
be characterized as “optimal”, according to previous reports examining
PAP manifestation on the subsequent sprint performance (Crewther et
al., 2011; Mola et al., 2014). In fact, a recent meta-analysis proposed an
adequate rest interval period of 7 to 10 minutes (Wilson et al., 2013).
Even so, the 4 min rest interval has been applied in numerous studies
(for a review, see Healy et al., 2017). However, we chose to apply this
methodology, as one of our main objectives was to closely simulate an
actual training program occur during a conditioning session in soccer.
During game analysis, 10 to 15 m sprint distances are usually per-
formed every 90 seconds by professional soccer players (Withers et al.,
1982). Moreover, up to 36 jumps are executed by a soccer player during
a game situation (Mohr et al., 2003), usually preceding a sprint activity.
This representation of training and game conditions could partially ex-
plain the fact that the shorter rest interval approach (2 min) concluded to
the most noticeable results among the different strategies applied in this
study. Another suggestion for this outcome could be that the CMJ prob-
ably impacted as an accumulated potentiation activity, even though it
was executed in an unloaded manner. Indeed, a greater PAP response
realized during a shorter time interval (0.3 - 4 min) has been observed
after the completion of an unloaded plyometric exercise, compared to
activities characterized by a high-intensity stimulus since less fatigue is
produced (Seitz & Haff, 2015). Regarding the slower sprint times in S5,
S20 and S30 observed during the EXP1 protocol compared to the base-
line test, we could hypothesize that the power exercise performed with
no rest interval prior to the sprint test may cause a reduction in the par-
ticipants’ ability for force production, resulting in decreased sprint per-
formance. Nevertheless, in the absence of a neuromuscular activation
test (such as EMG), this hypothesis can only be assumed.

232
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Complex training, including the combination of heavy resistance

and plyometric exercises executed in the same session, is presented as
a time-efficient method for the concurrent improvement of both mus-
cular strength and power abilities (Ebben, 2002), especially during the
in-season period where the time for strength and conditioning programs
is limited. The development of these abilities is considered highly cru-
cial during game situations, as they are associated with more effective
jumps, kicks and tackles (Stolen et al., 2005) and improved repeated
sprint performance (Bogdanis et al., 2011). Therefore, the optimization
of complex training programs is of interest in soccer for improving ath-
letes’ fitness level as well as transferring these gains to specific technical
skills (Brito et a.l, 2014). Despite the existence of only trivial to small ef-
fect size, the results presented in this study are likely to have a practical
application and are expected to facilitate soccer coaches and strength
and conditioning practitioners towards the improvement of athletes’
fitness level, supplementary to the acceleration and maximal running
sprint enhancement.
The current study has several strengths (i.e., large number of trained
soccer players, testing during actual training conditions, counter-bal-
anced design). However, some limitations need to be addressed. First,
given that maximum potentiation effects vary among athletes with dif-
ferent strength status, an inclusion criterion related to the 1RM back
squat exercise should have been inserted. However, the relatively small
SD value for the 1RM back squat observed (± 9.6kg) indicates a small
variability within the participants. Second, exercise intensity during the
CMJ execution was not controlled. Even though this exercise was not the
potentiation activity in our study, different levels of effort may influence
the outcome. Nevertheless, strong verbal encouragement was given
throughout all testing sessions. Finally, the evaluation of the players’
rate of perceived exertion might have assisted in more detailed results.
Future studies should aim to address these issues.

Conclusions
The results of this study clearly present the effectiveness of a PAP
strategy including a heavy loaded back squat exercise with a 2-min pas-
sive rest interval and an unloaded vertical jumping exercise following
the same rest period, for enhancing the subsequent acceleration (5 m)
and maximal sprint times (20 & 30 m) in semi-professional male soccer
players. This potentiation strategy was also found superior in improving
the above-mentioned variables, compared to the rest of the protocols

233
 sp or t science

applied and structured with different rest intervals. Finally, the data pre-
sented here indicate that by eliminating the rest interval between a plyo-
metric exercise and a sprint test may conclude to the opposite effects.

References
Andrzejewski, M., Chmura, J., Pluta, B., Strzelczyk, R., Kasprzak, A.J. (2013).
Analysis of sprinting activities of professional soccer players. Journal of
Strength & Conditioning Research, 27(8): 2134-2140.
Arabatzi, F., Patikas, D., Zafeiridis, A., Giavroudis, K., Kannas, T., Gourgoulis,
V., Kotzamanidis, C.M. (2014). The post-activation potentiation effect on
squat jump performance: age and sex effect. Pediatric Exercise Science,
26(2): 187-94.
Baechle, T.R., Earle, R.W. (2008). Essentials of strength and conditioning (2nd
ed.). Champaign, IL: Human Kinetics.
Bevan, H.R., Cunningham, D.J., Tooley, E.P., Owen, N.J., Cook, CJ., Kilduff, L.P.
(2010). Influence of postactivation potentiation on sprinting performance
in professional rugby players. Journal of Strength & Conditioning Re-
search, 24(3):701-5.
Bogdanis, G.C., Papaspyrou, A., Souglis, A.G., Theos, A., Sotiropoulos,
A., Maridaki, M. (2011). Effects of two different half-squat training programs
on fatigue during repeated cycling sprints in soccer players. Journal of
Strength & Conditioning Research, 25(7):1849-1856.
Brito, J., Vasconcellos, F., Oliveira, J., Krustrup, P., Rebelo, A. (2014). Short-
term performance effects of three different low-volume strength-training
programmes in college male soccer players. Journal of Human Kinetics,
(40): 121-128.
Cohen, J. (1988). Statistical Power Analysis for the Behavioral Sciences (2nd ed.)
New Jersey, NJ: Lawrence Erlbaum.
Crewther, B.T., Kilduff, L.P., Cook, C.J., Middleton, M.K., Bunce, P.J., Yang, G.Z
(2011). The acute potentiating effects of back squats on athlete perfor-
mance. Journal of Strength & Conditioning Research, 25(12): 3319-3325.
DeRenne, C. (2010). Effects of postactivation potentiation warm-up in male and
female sport performances: a brief review. Strength & Conditioning Jour-
nal, 32: 58–64.
Docherty, D., Robbins, D., Hodgson, M. (2004). Complex training revisited: A re-
view of its current status as a viable training approach. Strength & Condi-
tioning Journal, 26(6): 52-57.
Doma, K., Sinclair, W.H., Hervert, S.R., Leicht, A.S. (2016). Postactivation po-
tentiation of dynamic conditioning contractions on rowing sprint perfor-
mance. Journal of Science and Medicine in Sport, 19(11): 951-956.
Ebben, W.P. (2002). Complex training: A brief review. Journal of Sports Science
and Medicine, 1(2): 42-46.
Esformes, J.I., Bampouras, T.M. (2011). Effect of back squat depth on lower-body
postactivation potentiation. Journal of Strength & Conditioning Research,
27(11): 2997–3000.
Hamada, T., Sale, D.G., MacDougall, J. D., Tarnopolsky, M.A. (2000). Postactiva-
tion potentiation, fiber type, and twitch contraction time in human knee

234
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

extensor muscles. Journal of Applied Physiology, 88(6): 2131-2137.

Healy, R., Comyns, T.M. (2017). The Application of Postactivation Potentiation
Methods to Improve Sprint Speed. Strength & Conditioning Journal, 39(1):
1-9.
Jo, E., Judelson, D.A., Brown, L.E., Coburn, J.W., Dabbs, N.C. (2010). Influence
of recovery duration after a potentiating stimulus on muscular power in
recreationally trained individuals. Journal of Strength & Conditioning Re-
search, 24(2): 343–347.
Low, D., Harsley, P., Shaw, M., Peart, D. (2015). The effect of heavy resistance ex-
ercise on repeated sprint performance in youth athletes. Journal of Sports
Sciences, 33(10):1028-34.
McBride, J.M., Nimphius, S., Erickson, T.M. (2005). The acute effects of heavy-
load squats and loaded countermovement jumps on sprint performance.
Journal of Strength & Conditioning Research, 19(4): 893-897.
Maio Alves, J.M., Rebelo, A.N., Abrantes, C., Sampaio, J. (2010). Short-term ef-
fects of complex and contrast training in soccer players’ vertical jump,
sprint, and agility abilities. Journal of Strength & Conditioning Re-
search, 24(4): 936-41.
Mola, J.N., Bruce-Low, S.S., Burnet, S.J. (2014). Optimal recovery time for
postactivation potentiation in professional soccer players. Journal of
Strength & Conditioning Research, 28(6): 1529–1537.
Mohr, M., Krustrup, P., Bangsbo, J. (2003). Match performance of high-standard
soccer players with special reference to development of fatigue. Journal of
Sports Sciences, 21(7): 519-528.
Newman, M.A. Tarpenning, K.M., Marino, F.E. (2004). Relationship between
isokinetic knee strength, single-sprint performance, and repeated-sprint
ability in football players. Journal of Strength and Conditioning Research,
18(4): 867-872.
Ountzoudi, T.H., Manou, V., Dalamitros, A., Vavritsas, G., Kellis, S. (2014). The
Effects of a Complex Training Protocol, with Strength and Power Preactiva-
tion Stimulus, on Sprint Running in Adolescent Sprinters. Inquiries in Sport
& Physical Education, 1(3): 1-7.
Rahimi, R. (2007). The acute effects of heavy versus light-load squats in sprint
performance. Physical Education & Sport, 5: 163-169.
Ramírez-Campillo, R., Vergara-Pedreros, M., Henríquez-Olguín, C., Martínez-
Salazar, C., Alvarez, C., … Izquierdo, M. (2016). Effects of plyometric training
on maximal-intensity exercise and endurance in male and female soccer
players. Journal of Sports Science, 34(8): 687-93.
Robbins, D.W., Docherty, D. (2005). Effect of loading on enhancement of power
performance over three consecutive trials. Journal of Strength & Condi-
tioning Research, 19(4): 898-902.
Sarramian, V.G., Turner, A.N., Greenhalgh, A.K. (2015). Effect of postactivation
potentiation on fifty-meter freestyle in national swimmers. Journal of
Strength & Conditioning Research, 29(4): 1003–1009.
Seitz, L.B., Haff, G.G. (2015). Factors Modulating Post-Activation Potentiation of
Jump, Sprint, Throw, and Upper-Body Ballistic Performances: A System-
atic Review with Meta-Analysis. Sports Medicine, 46(2): 231-240.
Seitz, L.B., Villarreal, E.S., Haff, G.G. (2014). The temporal profile of postactiva-

235
 sp or t science

tion potentiation is related to strength level. Journal of Strength & Condi-

tioning Research, 28(3): 706–715.
Tillin, N.A., Bishop, D. (2009). Factors modulating post-activation potentiation
and its effect on performance of subsequent explosive activities. Sports
Medicine, 39(2): 147-166.
Tsimachidis, C., Patikas, D., Galazoulas, C., Bassa, E., Kotzamanidis, C. (2013).
The post-activation potentiation effect on sprint performance after com-
bined resistance/sprint training in junior basketball players. Journal of
Sports Science, 31(10): 1117-24.
Wilson, J.M., Duncan, N.M., Marin, P.J., Brown, L.E., Loenneke, J.P., Wilson, S.M.,
Jo, E., Lowery, R.P., Ugrinowitsch, C. (2013). Meta-Analysis of post activa-
tion potentiation and power: Effects of conditioning activity, volume, gen-
der, rest periods, and training status. Journal of Strength & Conditioning
Research, 27(3): 854-859.
Wisløff, U., Castagna, C., Helgerud, J., Jones, R., Hoff, J. (2004). Strong correla-
tion of maximal squat strength with sprint performance and vertical jump
height in elite soccer players. British Journal of Sports Medicine, 38(3): 285-
288.
Withers, R.T., Maricic, Z., Wasilewski, S., Kelly, L. (1982). Match analysis of Aus-
tralian professional soccer players. Journal of Human Movement Studies,
8: 159-176.

236
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Swedish tennis test: a field test to estimate

maximal oxygen uptake in tennis players
Pierre-Marie Leprêtre1, Philippe Lopes2,3 and Pascal Sirvent4

Introduction
Tennis is a very complex sport, involving strength, power, speed, agil-
ity and explosiveness, as well as endurance components [Filipčič and
Filipčič
čič
ičč 2006, Baiget et al. 2015]. Particularly with respect to the dura-
tion of the tournaments and the length of the matches, a good aero-
bic capacity could positively influence the outcome of games [König et
al. 2001]. Game analysis suggested that tennis matches intensity would
correspond to a high intensity intermittent exercise [Mendez-Villanueva
et al. 2007, Smekal et al. 2001] and data from most studies examining
power recovery and maximal oxygen uptake ( O2max) suggest that a
higher value of this variable results in improved recovery when perform-
ing repeated bouts of high intensity intermittent exercise [Tomlin and
Wenger 2001]. Finally, based on the postulate that phosphocreatine re-
synthesis is controlled by aerobic metabolism, it is suggested that the
faster its recovery rate might be attributed to a greater aerobic capac-
ity [Yoshida T. 2002]. These findings support the idea that high O2max
values would constitute an important factor for tennis player’s success.
Classically, a ramp test performed on motorized treadmill or on field
track test could be used to directly or indirectly determine O2max. A
more specific test, the tennis specific fitness test, has been proposed to
estimate aerobic fitness in tennis players [Girard et al. 2006]. Based on
a personal game analysis, these authors proposed an intermittent pro-
gressive shuttle test with feigned strokes and randomized frontal and
sagittal displacements spaced by passive recovery. When comparing the
physiological responses after a progressive treadmill exercise test and

1. Laboratoire de recherche Adaptations Physiologiques à l’Exercice et Réadap-

tation à l’Effort, EA-3300, UFR-STAPS, Université de Picardie Jules Verne,
Amiens, France.
2. UFR Sciences Fondamentales Appliquées, Département STAPS, Université
d’Evry-Val-d’Essonne, Evry, France.
3. INSERM, UMR-S 1124, Sorbonne Paris Cité, Faculté des Sciences Fondamen-
tales et Biomédicales, Université Paris Descartes, Paris, France.
4. Laboratoire de Biologie des Activités Physiques et Sportives, EA-3533, UFR-
STAPS, Université Blaise Pascal – Clermont-Ferrand, Aubière, France.

237
 sp or t science

following a tennis specific fitness test, O2max was found higher with the
tennis specific fitness test suggesting that the motorized treadmill test
failed to discriminate the respective part of artificial strokes and the abil-
ity to change direction. Finally, authors did not conclude about the in-
terchangeability of tennis specific fitness and motorized treadmill tests.
In the past, validity of maximal multistage 20 m shuttle run test was
established to predict O2max in adults [Léger et al. 1988, Ahmaidi et al.
1992, Berthoin et al. 1996]. The comparison of lower limbs and upper-
lower limbs tests, both with changes of direction, might better advance
the influence of upper limbs activity on the whole exercise O2 response.
Fargeas-Gluck and Léger (2012) and Brechbuhl et al. (2016) compared
the cardiovascular responses measured during the maximal multistage
20 m shuttle run test with those obtained during a tennis specific test.
However, as in the protocol proposed by Baiget et al. (2014), these two
studies are only based on lateral movements from baseline, without the
back and forth or diagonal trajectories normally present in an ecologi-
cal playing situation [Filipčič and Filipčič 2006]. In addition, these tests
[Fargeas-Gluck and Léger, 2012, Baiget et al. 2014, Brechbuhl et al. 2016],
with real ball strikes with designated return areas, would alter both the
cardiorespiratory responses [Ando et al. 2016] and the appropriate ex-
ecution [Brechbuhl et al. 2017].
As increasing the complexity of the task by introducing real strikes
would also be detrimental to a practical evaluation that should be quick
and easy to implement in the field (one experimenter and one half-field)
and to the reliability of measurement depending on player’s level of ex-
pertise [Wengaard et al. 2017], we aimed to compare O2 responses dur-
ing the maximal multistage 20 m shuttle run test and the Swedish tennis
tests, the latter being a modified version of tennis specific fitness test.
We supposed that O2max is not markedly elevated during combined
upper-lower limbs (Swedish tennis test) compared to lower limbs exer-
cise (maximal multistage 20 m shuttle run test).

Methods
Subjects
Eight recreational male tennis players (third level French national
tennis ranking) participated in the current study and all were healthy
and free of cardiac and pulmonary disease. Their physical characteris-
tics (mean ± SD) were: 26.0 ± 5.7 years, 184.4 ± 6.7 cm and 81.7 ± 10.7 kg.
Each subject was familiarised to the experimental procedures and was
informed of the risks and stresses associated with the protocol. Subjects

238
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

gave their written informed consent in accordance with the guidelines of

the University of Clermont-Ferrand.

Protocol
On separated days, all subjects performed two incremental exercise
tests on an indoor hard tennis court (20.3°C of temperature, 35.0% of
relative humidity and 762 mmHg of atmospheric pressure) 2-3 h after
breakfast.
Regarding the maximal multistage 20 m shuttle run test, after a 2 min
warm-up at 8.0 km.h-1, each subject ran back and forth continuously be-
tween two points separated by 20 m apart. The runs were synchronized
with beeps emitted from a pre-recorded tape. The subjects were forced
to increase their speed since the interval between each successive beep
decreased over the course of the test (0.5 km.h-1 speed increase every 1
min until exhaustion for a total exercise duration not exceeding 25 min;
Léger and Gadoury, 1989).
The Swedish tennis test is a continuous progressive exercise test per-
formed with a tennis racquet in which subjects repeat displacements
replicating the game of tennis (back and forth, and sideways). Each
stage consists of shuttle runs, performed from a central base to one of
six targets located around the court [Filipčič and Filipčič 2006] (figure 1).
The sets of displacements included two forward (offensive), two lateral
(neutral) and two backward (defensive) courses. When subjects arrived
at the target, they were instructed to realize a powerful ball strike with
their racquet (three forehands and three backhands, respectively) be-
fore moving back to baseline after each set. The initial running work rate
was 1.67 m.s-1 (i.e. 6.0 km.h-1) lasting 1 min, then run shuttles progres-
sively increased by 0.07 m.s-1 (i.e. 0.25 km.h-1) every 1 min until volun-
tary exhaustion or the incapacity of subject to attain target in time or
to perform strokes with an acceptable technique. Movement velocities
and directions were controlled by visual and sound feedback from a PC.
Velocities and displacements were calculated from data collected during
official competitions [Botton et al. 2011].

Data collection procedures

Breathing frequency (RF), tidal volume (VT), minute ventilation ( ), in-
spired O2 and CO2 fractions (FeO2 and FeCO2, respectively) were meas-
ured at rest and throughout exercise using Cosmed K4b2 ambulatory
systems [Duffield et al. 2004]. Then, O2 and CO2 values were estimated
using the Haldane Equation. K4b2 was calibrated according to the manu-

239
 sp or t science

facturer’s instruction manual, particularly using ambient air (20.9% O2

and 0.03% CO2) and calibration gas (16.0% O2 and 5.0% CO2). The cali-
bration of the turbine flow meters was performed with a 3 L syringe. Ex-
pired gases were measured breath by breath and the Cosmed software
was used to automatically eliminate ectopic values and average the data
every 15 s and O2max was defined as previously proposed (Lepretre et al.
2008). In addition, heart rate (HR) was recorded beat-to-beat, averaged
over each breath (S810, Polar, Kempele, Finland) and, finally, averaged
every 15 s.

Statistical Methods
Descriptive statistics are expressed as means and standard deviations
(SD). After the normality of the data was verified (Skewness and Kurtosis
measures, Fisher-Snedecor F-test), the non-parametric Wilcoxon signed-
rank test for paired data was used to compare the basal and maximal
physiological values. The significance alpha level was set at p<0.05. The
relationships between cardiopulmonary variables obtained during both
tests were assessed using a spreadsheet [Hopkins 2000], particularly us-
ing linear regression to assess the agreement for both raw and log trans-
formed data, providing measures of bias and its 95% confidence limits,
as well as the Pearson’s correlation coefficient (r) for measured variables.

Figure 1. Court zones of the Swedish tennis test protocol.

240
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Results
Time to exhaustion was longer during specific tennis test compared
to the maximal multistage 20 m shuttle run test (15.5 ± 1.4 vs. 10.5 ± 0.5
min, p=0.043) with a large size effect (Cohen’s d=0.93, effect-size correla-
tion = 0.42). Table 1 presents the physiological variables measured at rest
and during the maximal multistage 20 m shuttle run and Swedish tennis
tests. Basal and maximal HR values were not different, with a moderate
size effect (Cohen’s d=0.56, effect-size correlation = 0.27). Absolute and
relative O2, RF and E did not differ at rest, with a small size effect for
RF (Cohen’s d=0.15, effect-size correlation = 0.08) and large one for other
variables (Cohen’s d ranged from 1.45 to 1.65 and effect-size correlation
ranged from 0.59 to 0.65). VT baseline and maximal values were differ-
ent when comparing tests, with a large size effect (Cohen’s d were 2.55
and 2.22, and effect-size correlations were 0.79 and 0.74, respectively).
Consequently, maximal E was higher during the maximal multistage 20
m shuttle run test comparing to the Swedish tennis test (with a large size
effect (Cohen’s d=2.33 and effect-size correlation = 0.76). Absolute and
relative O2max obtained during maximal multistage 20 m shuttle run
test was also higher compared to the Swedish tennis test with a large
size effect (Cohen’s d=2.4 and effect-size correlation = 0.77).
Pearson correlation analysis indicated relevant relationships between
physiological variables between tests: HR (r=0.94, p<0.05), absolute and
relative O2 (r=0.98, p<0.01), E (r=0.98, p<0.01), RF (r=0.79, p<0.05) and
T
(r=0.89; p<0.05) (figure 2). Mean, upper and lower limits of bias were:
absolute O2, 0.3 (equivalent to 3.8%) ranging from 0.1 to 1.0 L.min-1;
relative O2, 4.1 (i.e. 3.3%), ranging from 2.2 to 12.7 mL.min-1.kg-1; VE ,
8.3 L.min-1 (i.e. 3.6 %), ranging from 1.3 to 19.9 L.min-1; RF, 0.9 breaths.
min-1 (i.e. 2.2 %), ranging from 3.1 to 5.8 breaths.min-1 and T, 0.2 L (i.e.
1.3 %) ranging from 0.3 to 1.6 L. For a mean bias of 3.3%, the true error
of O2 measurement was likely to be substantially greater for maximal
multistage 20 m shuttle run test compared to Swedish tennis test.

241
 sp or t science

Resting values Maximal values

UM-TT Swedish tennis test UM-TT Swedish tennis test
O2 (L.min-1) 0.7 ± 0.1 0.9 ± 0.1 4.7 ± 0.3 4.0 ± 0.3*
O2 (mL.min . -1
9.3 ± 1.5 11.7 ± 1.8 59.9 ± 3.7 51.1 ± 2.8*
kg-1)
VE (L.min-1) 21.3 ± 2.2 25.2 ± 2.7 142.9 ± 126.0 ± 8.1*
6.3
RF (breaths. 24.2 ± 1.6 23.9 ± 1.6 48.8 ± 3.9 50.9 ± 2.5
min-1)
V T (L) 0.9 ± 0.1 ± 0.1* 4.1 ± 0.4 2.5 ± 0.1*
HR (bpm) 91.6 ± 6.9 82.0 ± 12.3 181.1 ± 4.8 182.6 ± 4.0

Table 1. Basal and maximal physiological values measured before and during
both 20 m shuttle run and Swedish tennis test.

Note: UM-TT = maximal multistage 20 m shuttle run test, O2 = oxygen up-

take, E = ventilation, RF = respiratory frequency, VT = tidal volume, HR = heart
rate. * p<0.05.

242
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Figure 2. Regression diagrams regarding physiological variables in-between

tests.

Note: UM-TT and SST corresponded to the maximal multistage 20 m shuttle

run test and Swedish Tennis Test, respectively.

Discussion
Lasting from 1 to 5 h, a tennis match consists of repeated short and
intense muscular actions, including fast running intercepted by longer
periods of moderate and low intensity activity (Fernandez et al. 2006). It
was observed that the aerobic energy expenditure was quite low over the
game (49.4 ± 4.8% O2max) and that the maximal power during points
was up to two or three times the O2max (Botton et al. 2011). Thus, a
tennis match is an intermittent anaerobic sport with an aerobic recovery
phase (Fernandez et al. 2006). Using video analysis and indirect calo-
rimetry, Botton et al. (2011) considered the anaerobic metabolism to be
32% of the total energy expenditure across the game, 67% for points and
95% for hitting the ball. However, the aerobic system plays a significant
role in the maintenance of intensity level during a game, which is charac-
terized by short activities bursts [Tomlin and Wenger 2001, Meckel et al.
2009]. Literature suggests that aerobic fitness enhances recovery from
high intensity intermittent exercise through increased aerobic response,
improve lactate removal and enhanced phosphocreatine regeneration
[Tomlin and Wenger 2001]. The aim of this study was to compare O2
responses measured during maximal multistage 20 m shuttle run and
Swedish tennis tests, with the main results showing good correlation
in-between tests in the maximal HR and O2 values. This is consistent
with previous observations that indicated that maximal values of cardio-
respiratory variables were closely correlated when comparing specific
tennis test and incremental treadmill protocols [Ferrauti et al. 2011] or
maximal multistage 20 m shuttle run test [Fargeas-Gluck et Leger. 2012].
However, in the current study O2max was markedly lower for the Swed-

243
 sp or t science

ish tennis test (combined upper-lower limbs exercise test) compared to

the values obtained in the maximal multistage 20 m shuttle run test (low-
er limbs exercise only).
Previous studies showed that field tests significantly under-predicted
the O2max compared to treadmill testing (Smekal et al. 1995, St Clair
Gibson et al. 1998, Girard et al. 2006). A significant correlation between
O2max determined by the maximal multistage 20-m shuttle run test
and the highest O2 measured at the end of a previous tennis protocol
was also shown (r=0.61; P<0.05) by St Clair Gibson et al. (1998). However,
these authors attributed the less robust correlation between the shuttle
run and motorized treadmill O2max tests to the tennis specific training
(St Clair Gibson et al. 1998). In agreement with these previous studies,
the highest O2 measured during Swedish tennis test was less com-
pared to O2max. We also showed a good correlation between O2max
measured during both the maximal multistage 20 m shuttle run test and
Swedish tennis test. In the present study, a difference in minute ventila-
tion was also found. Smekal et al. (2000) had already reported higher
minute ventilation values in motorized treadmill test compared to field
tests. However, recent studies did not show any difference in O2max or
HR when comparing the specific tennis and maximal multistage 20 m
shuttle run tests (Fargeas-Gluck et Leger. 2012, Brechbuhl et al. 2016),
evidencing that specific tennis tests also allow reaching O2max and
are useful for assessing aerobic fitness in this sport. Some methodologi-
cal points (like expertise level and age) may explain this dissimilarity. In
fact, Brechbuhl et al. (2016) were interested in assessing cardiorespira-
tory responses in highly trained competitive tennis players who were
accustomed to changes of direction. The regular repetitions of changes
of direction repetitions result in improved endothelial function and arte-
rial blood vessel stiffness (Kruse Scheuermann, 2017) and, therefore, the
oxygen muscle delivery in trained subjects. As a result, trained subjects
can reach their O2max during tennis specific exercises involving chang-
es in direction (Born et al. 2017). In contrast, acute changes of direction
induce excessive over-stretching of the quadriceps and calf muscles
(Kruse Scheuermann, 2017), which alter blood flow and subsequently
oxygen availability and muscle utilization in our recreational players.
Secondly, our subjects are older than the tennis players studied before
(Fargeas-Gluck et Leger, 2012). Children rely more on oxidative metabo-
lism and less on creatine kinase reaction to meet energy demand during
exercise (Tomson et al., 2010), which is explained by a greater oxidative
capacity, probably linked to a higher relative content in slow-twitch fib-

244
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

ers before puberty. Hence, children solicited high O2 values during in-
tense and exhaustive exercises, whatever the sport. Finally, differences
in the displacements quality may have induced some specific cardiores-
piratory responses, as oxygen transport to working skeletal muscles is
challenged during whole-body exercise (Reybrouck et al. 1975). Muscle
mass solicitation may explain the difference between O2max meas-
ured during the maximal multistage 20 m shuttle run and Swedish ten-
nis tests. In fact, Secher and Volianitis (2006) reported that the combined
upper and lower limbs exercise induced a 10% decrease in lower limbs
blood flow at a given work rate compared to lower limbs exercise alone.
It has been reported that physical activities requiring repeated phases
of high-force isometric and eccentric contractions resulted in restricted
blood flow to the working muscle, thereby reducing oxygen delivery and
increasing metabolite accumulation (Ferguson 2010). In consequence,
there will be skeletal muscle fatigue, through both central and periph-
eral mechanisms, and a potential loss of motor control ultimately lim-
iting performance. During Swedish tennis test, stretch-shortening cycle
(Grezios et al. 2006) and eccentric wrist extensor muscular actions dur-
ing simulated tennis stroke and displacements could potentially induce
peripheral vasoconstriction and, therefore, compromise the ability to
increase cardiac output and O2.
On the other hand, previous studies reported a protocol duration ef-
fect on the O2max attainment. Buchfuhrer et al. (1983) have demon-
strated that 8 to 17 min tests elicited higher values of O2 than shorter
(<8 min) treadmill and cycle ergometer tests. In the current study, time
to perform was longer to 8 min for both tests but the Swedish tennis test
lasted longer than maximal multistage 20 m shuttle run test. Yoon et al.
(2007) showed that healthy, moderately and highly trained men attained
a higher O2max when they were engaged in 8 rather than 16 min in-
cremental test. Earlier, McCole et al. (2001) have reported that a 12 min
incremental test elicited O2max with a lower cardiac output compared
with a 6 min one. Stage duration protocol may affect the stroke volume
responses due to a different muscle mass recruitment (Lepretre et al.
2004) but also because of different cutaneous vasodilatation. McCole et
al. (2001) supposed that twice longer duration test (12 vs. 6 min) would
be responsible for the increase in core temperature eliciting cutaneous
vasodilatation. Increased competition for the distribution of cardiac
output can be made to the detriment of the highest O2 value achievement.
Hence, Yoon et al. (2007) suggested that protocol duration of tests to O2max
should be between 8 and 10 min for moderately trained subjects.

245
 sp or t science

Together, these data suggest that protocol duration might also explain a
part of significant difference in higher O2 value due to a different car-
diac output challenge.
To conclude, our results indicate a strong relationship between the
highest O2 values measured during the Swedish tennis test and the 20
m shuttle run. However, the Swedish tennis test underestimated the
O2max due to the specific motor action. Tennis training regimes should
be adapted to the specific demands imposed by match-play. Sport-spe-
cific actions interact with the achievement of highest O2 values, there-
fore, the Swedish tennis test rather than 20 m shuttle run should be used
to evaluate the specific aerobic capacities of recreational male tennis
players.

References
1. Ahmaidi S, Collomp K, Caillaud C, Préfaut C. Maximal and functional aero-
bic capacity as assessed by two graduated field methods in comparison
to laboratory exercise testing in moderately trained subjects. Int J Sports
Med. 1992 Apr;13(3):243-8.
2. Ando S, Komiyama T, Kokubu M, Sudo M, Kiyonaga A, Tanaka H, Higaki Y.
Slowed response to peripheral visual stimuli during strenuous exercise.
Physiol Behav. 2016 Jul 1;161:33-37.
3. Aziz AR, Chia MY, Teh KC. Measured maximal oxygen uptake in a multi-stage
shuttle test and treadmill-run test in trained athletes. J Sports Med Phys
Fitness. 2005 Sep;45(3):306-14.
4. Baiget E, Fernández-Fernández J, Iglesias X, Rodríguez FA. Tennis Play In-
tensity Distribution and Relation with Aerobic Fitness in Competitive Play-
ers. PLoS One. 2015 Jun 22;10(6):e0131304.
5. Baiget E, Fernández-Fernández J, Iglesias X, Vallejo L, Rodríguez FA. On-
court endurance and performance testing in competitive male tennis play-
ers. J Strength Cond Res. 2014 Jan;28(1):256-64.
6. Berthoin S, Pelayo P, Lensel-Corbeil G, Robin H, Gerbeaux M. Comparison
of maximal aerobic speed as assessed with laboratory and field measure-
ments in moderately trained subjects. Int J Sports Med. 1996 Oct;17(7):525-
9.
7. Bland JM, Altman DG. Statistical methods for assessing agreement between
two methods of clinical measurement. Lancet. 1986 Feb 8;1(8476):307-10.
8. Born DP, Kunz P, Sperlich B. Reliability and validity of an agility-like incre-
mental exercise test with multidirectional change-of-direction movements
in response to a visual stimulus. Physiol Rep. 2017 May;5(9). pii: e13275.
9. Botton F, Hautier C, Eclache JP. Energy expenditure during tennis play: a
preliminary video analysis and metabolic model approach. J Strength
Cond Res. 2011 Nov;25(11):3022-8.)
10. Brechbuhl C, Girard O, Millet GP, Schmitt L. On the Use of a Test to Exhaus-
tion Specific to Tennis (TEST) with Ball Hitting by Elite Players. PLoS One.
2016 Apr 1;11(4):e0152389.

246
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

11. Brechbuhl C, Girard O, Millet GP, Schmitt L. Technical Alterations during an

Incremental Field Test in Elite Male Tennis Players. Med Sci Sports Exerc.
2017 Sep;49(9):1917-1926.
12. Christmass MA, Richmond SE, Cable NT, Arthur PG, Hartmann PE. Exer-
cise intensity and metabolic response in singles tennis. J Sports Sci. 1998
Nov;16(8):739-47.
13. Duffield R, Dawson B, Pinnington HC, Wong P. Accuracy and reliability
of a Cosmed K4b2 portable gas analysis system. J Sci Med Sport. 2004
Mar;7(1):11-22..
14. Fargeas-Gluck MA, Léger LA. Comparison of two aerobic field tests in young
tennis players. J Strength Cond Res. 2012 Nov;26(11):3036-42.
15. Ferguson RA. Limitations to performance during alpine skiing. Exp Physiol.
2010 Mar;95(3):404-10. Epub 2009 Nov 6.
16. Fernandez-Fernandez J, Mendez-Villanueva A, Fernandez-Garcia B, Terra-
dos N. Match activity and physiological responses during a junior female
singles tennis tournament. Br J Sports Med. 2007 Nov;41(11):711-6. Epub
2007 Jun 11.
17. Fernandez J, Mendez-Villanueva A, Pluim BM. Intensity of tennis match
play. Br J Sports Med. 2006 May;40(5):387-91; discussion 391.
18. Ferrauti A, Kinner V, Fernandez-Fernandez J. The Hit & Turn Tennis Test: an
acoustically controlled endurance test for tennis players. J Sports Sci. 2011
Mar;29(5):485-94.
19. Filipčič A, Filipčič T. Analysis of time and game characteristics in top profile
tennis Acta Universitatis Carolinae. Kinanthropologica, 2006 Jan;42(1), 41-
53.
20. Flouris AD, Metsios GS, Famisis K, Geladas N, Koutedakis Y. Prediction of
VO2max from a new field test based on portable indirect calorimetry. J Sci
Med Sport. 2010 Jan;13(1):70-3. Epub 2009 Jun 27.
21. Girard O, Chevalier R, Leveque F, Micallef JP, Millet GP. Specific incremental
field test for aerobic fitness in tennis. Br J Sports Med. 2006 Sep;40(9):791-
6. Epub 2006 Jul 19.
22. Katić R, Milat S, Zagorac N, Durović N. Impact of game elements on tennis
match outcome in Wimbledon and Roland Garros 2009. Coll Antropol. 2011
Jun;35(2):341-6.
23. König D, Huonker M, Schmid A, Halle M, Berg A, Keul J. Cardiovascular, met-
abolic, and hormonal parameters in professional tennis players. Med Sci
Sports Exerc. 2001 Apr;33(4):654-8.
24. Kruse NT, Scheuermann BW. Cardiovascular Responses to Skeletal Muscle
Stretching: “Stretching” the Truth or a New Exercise Paradigm for Cardio-
vascular Medicine? Sports Med. 2017 Dec;47(12):2507-2520.
25. Léger L, Boucher R. An indirect continuous running multistage field test:
the Université de Montréal track test. Can J Appl Sport Sci. 1980 Jun;5(2):77-
84.
26. Léger L, Gadoury C. Validity of the 20 m shuttle run test with 1 min stages to
predict VO2max in adults. Can J Sport Sci. 1989 Mar;14(1):21-6.
27. Léger LA, Lambert J. A maximal multistage 20-m shuttle run test to predict
VO2max. Eur J Appl Physiol Occup Physiol. 1982;49(1):1-12.
28. Lepretre PM, Lopes P, Koralsztein JP, Billat V. Fatigue responses in exercise

247
 sp or t science

under control of VO2. Int J Sports Med. 2008 Mar;29(3):199-205. Epub 2007
Sep 18.
29. Lepretre PM, Koralsztein JP, Billat VL. Effect of exercise intensity on rela-
tionship between VO2max and cardiac output. Med Sci Sports Exerc. 2004
Aug;36(8):1357-63.
30. Meckel Y, Machnai O, Eliakim A. Relationship among repeated sprint tests,
aerobic fitness, and anaerobic fitness in elite adolescent soccer players. J
Strength Cond Res. 2009 Jan;23(1):163-9.
31. Mendez-Villanueva A, Fernandez-Fernandez J, Bishop D, Fernandez-Garcia
B, Terrados N. Activity patterns, blood lactate concentrations and ratings
of perceived exertion during a professional singles tennis tournament. Br J
Sports Med. 2007 May;41(5):296-300; discussion 300. Epub 2007 Jan 19.
32. Mendez-Villanueva A, Fernandez-Fernández J, Bishop D, Fernandez-Garcia
B. Ratings of perceived exertion-lactate association during actual singles
tennis match play. J Strength Cond Res. 2010 Jan;24(1):165-70.
33. Novas AM, Rowbottom DG, Jenkins DG. A practical method of estimating
energy expenditure during tennis play. J Sci Med Sport. 2003 Mar;6(1):40-
50.
34. Reid M, Crespo M, Lay B, Berry J. Skill acquisition in tennis: research and
current practice. J Sci Med Sport. 2007 Feb;10(1):1-10. Epub 2006 Jun 30.
35. Reybrouck T, Heigenhauser GF, Faulkner JA. Limitations to maximum oxy-
gen uptake in arms, leg, and combined arm-leg ergometry. J Appl Physiol.
1975 May;38(5):774-9.
36. Secher NH, Volianitis S. Are the arms and legs in competition for cardiac
output? Med Sci Sports Exerc. 2006 Oct;38(10):1797-803. Review.
37. Smekal G, Baron R, Pokan R, Dirninger K, Bachl N. Metabolic and cardiores-
piratory reactions in tennis-players in laboratory testing and under sport-
specific conditions. Wien Med Wochenschr. 1995;145(22):611-5.
38. Smekal G, Pokan R, von Duvillard SP, Baron R, Tschan H, Bachl N. Compari-
son of laboratory and “on-court” endurance testing in tennis. Int J Sports
Med. 2000 May;21(4):242-9.
39. Smekal G, von Duvillard SP, Rihacek C, Pokan R, Hofmann P, Baron R,
Tschan H, Bachl N. A physiological profile of tennis match play. Med Sci
Sports Exerc. 2001 Jun;33(6):999-1005.
40. Smekal G, von Duvillard SP, Rihacek C, Pokan R, Hofmann P, Baron R,
Tschan H, Bachl N. A physiological profile of tennis match play. Med Sci
Sports Exerc. 2001 Jun;33(6):999-1005.
41. St Clair Gibson A, Broomhead S, Lambert MI, Hawley JA. Prediction of maxi-
mal oxygen uptake from a 20-m shuttle run as measured directly in runners
and squash players. J Sports Sci. 1998 May;16(4):331-5.
42. Tomlin DL, Wenger HA. The relationship between aerobic fitness and re-
covery from high intensity intermittent exercise. Sports Med. 2001;31(1):1-
11.
43. Yoon BK, Kravitz L, Robergs R. VO2max, protocol duration, and the VO2 pla-
teau. Med Sci Sports Exerc. 2007 Jul;39(7):1186-92.
44. Yoshida T. The rate of phosphocreatine hydrolysis and resynthesis in exer-
cising muscle in humans using 31P-MRS. J Physiol Anthropol Appl Human
Sci. 2002 Sep;21(5):247-55.

248
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

45. Hopkins WG (2000). Analysis of validity by linear regression (Excel spread-

sheet). In: A new view of statistics. sportsci.org: Internet Society for Sport
Science, sportsci.org/resource/stats/xvalid.xls.
46. Tomlin DL, Wenger HA. The relationship between aerobic fitness and re-
covery from high intensity intermittent exercise. Sports Med. 2001;31(1):1-
11.
47. Tonson A, Ratel S, Le Fur Y, Vilmen C, Cozzone PJ, Bendahan D. Muscle en-
ergetics changes throughout maturation: a quantitative 31P-MRS analysis.
J Appl Physiol (1985) 2010 Dec; 109(6): 1769–1778.
48. Wengaard E, Kristoffersen M, Harris A, Gundersen H. Cardiorespiratory Fit-
ness Is Associated with Selective Attention in Healthy Male High-School
Students. Front Hum Neurosci. 2017 Jun 28;11:330.

249
 sp or t science

How Down syndrome impacts on the athletes’

sports performance?
Tânia Bastos1,2, Ana Querido3,4, Ariel Villagra5, José Irineu Gorla6,
Ricardo J. Fernandes1,7, Rui Corredeira8

Introduction
Sport is meaningful in the lives of many people, including those with
Down syndrome (DS). Like any others, individuals with DS participate
in sport for different reasons and, for some of them, the transition from
recreational sport to intensive training and competition is a natural pro-
gression for testing personal limits and pursuing athletic dreams and
goals.(1) Nevertheless, trained individuals with DS are still scarcely stud-
ied.(2,3) In fact, little is known about the impact of training and sports
performance in this population.
Therefore, in this review paper we go beyond the literature about in-
tellectual disability and, specifically target individuals with DS. We in-
tend to provide an overview on the evolution of sports participation for
athletes with DS, from a recreational and rehabilitation perspective to a
competitive performance perspective. To do so we will present a retro-
spective examination of the most up to dated literature about the impact
of exercise and sport in this population. In addition, we seek to demon-
strate which proper assessment instruments are available to support the
training-coaching process and consequently, to reach the performance
goals in athletes with DS. This book chapter reviews the performance
factors implicated on competitive standards for athletes with DS and the

1. Centre of Research, Education, Innovation and Intervention in Sport (CIFI2D),

Faculty of Sport, University of Porto, Porto, Portugal.
2. Research Center in Sports Sciences, Health Sciences and Human Develop-
ment, CIDESD, University Institute of Maia, ISMAI, Maia, Portugal.
3. Associação de Natação do Norte de Portugal (ANNP), Porto, Portugal.
4. Clube de Propaganda da Natação (CPN), Ermesinde, Portugal.
5. Department of Physical Education, Sport and Human Movement, Autono-
mous University of Madrid, Madrid, Spain.
6. Laboratory of Physical Evaluation in Adapted Sport and Exercise, Depart-
ment of Studies Adapted Physical Education, Faculty of Physical Educa-
tion, University of Campinas, Campinas, Brazil.
7. Porto Biomechanics Laboratory, University of Porto, Porto, Portugal.
8. Research Centre in Physical Activity, Health and Leisure (CIAFEL), Faculty of
Sport, University of Porto, Porto, Portugal.

250
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

context of sports organizations in this field. Specifically, we want to bring

to discussion the inclusion of athletes with DS in Paralympic Games. We
also provide practical recommendations for coaches focusing on the en-
hancement of sports performance of athletes with DS.

Down syndrome definition and characteristics

Down syndrome also called Trisomy 21 is the most frequent chromo-
some condition with an estimated incidence ranging between 1 in 1,000
to 1 in 1,100 live births worldwide.(4) Generally, this genetic disorder
is caused by a fully or partial supplementary 21 chromosome and can
appear in three forms, namely Simple Trisomy 21, Translocation, and
Mosaicism.(5) Due to this genetic information, individuals with DS have
a specific phenotype gathering a unique combination of intellectual,
physical and physiological characteristics. Individuals with DS have in-
tellectual disability ranging from mild to severe and 70 to 75% of them
exhibit an IQ of between 25 to 50.(6)
Individuals with DS have a classic body shape and morphology, de-
scribed as shorter in stature and shorter limbs-to-torso ratio, muscle hy-
potonia and atlanto-axial instability,(7) scoliosis, several foot deformities
(e.g., flatfoot, hallux valgus, overriding fifth toe),(8) flat face, eyes slant-
ing up, irregular shaped ears and large tongue relative to the mouth.(4)
It is also known that they frequently exhibit congenital heart conditions,
hearing and ophthalmological impairments, gastrointestinal disorders,
obstructive sleep apnea (related with smaller upper airways, midfacial
hypoplasia, adenoid hypertrophy, lingual tonsillar hypertrophy), thyroid
disease,(9) leukaemia, epilepsy, diabetes and Alzheimer’s disease.(4)
Consequently, individuals with DS are different from their peers with
intellectual disabilities but without DS at physical and physiological lev-
els. Individuals with DS have low peak oxygen consumption (VO2peak)
and consequently low cardiorespiratory fitness in comparison with indi-
viduals with intellectual disabilities but without DS.(10) Autonomic dys-
function, reduced ventilatory capacity and metabolic dysfunction are
the three factors pointed to explain the poor aerobic capacities in this
population.(8) Pitetti, Climstein, Mays, and Barrett (11) demonstrated
that individuals with DS also have lower levels of isokinetic muscular
strength compared with their peers with intellectual disability but with-
out DS. They have lower bone density compared with those who have an
intellectual disability.(12) Finally, body composition in this population is
characterized by higher levels of fat mass, lower levels of fat-free mass,
and a different fat and lean distribution.(13) Consequently, individuals

251
 sp or t science

with DS have an increased risk of obesity compared with individuals

with intellectual disabilities without DS.(14)

State of art about exercise and sport in individuals with down

syndrome
Although it is well known that exercise and sport have several ben-
efits for those who engage in a regular and continuous practice, indi-
viduals with DS are generally classified as sedentary.(15) To understand
the impact of exercise and sports training in this population, we provide
an updated state of art about physical fitness and sport-specific skills
proficiency of individuals participating in exercise training. Full papers
published in English were searched from six major electronic databases,
namely Scopus, Ebsco, Sportdiscus, Web of Science, Pubmed and B-on
over the last 7 years (2010-2017).

Physical fitness programs

Regarding physical fitness programs, Boer and Moss (16) developed
a comparative analysis about the effectiveness of continuous aero-
bic training (continuous cycling and walking at an intensity of 70-80%
VO2max) and interval training (10-30s all out sprints with 90s of low
intensity walking or cycling) in anthropometrical, physiological and
functional parameters of adults with DS. The previously mentioned au-
thors demonstrated that both types of training have positive effects on
various assessed parameters, but interval training was more effective in
the improvement of body weight and aerobic capacity. Aerobic training
was also proven to be effective in the reduction of pro-inflammatory cy-
tokines and acute phase proteins associated with metabolic syndrome
and insulin resistance in obese women with DS.(17)
Specifically, walking activities programs provided improvements on
grip strength,(18) muscle strength (i.e., hip extensor, hip flexor, knee
extensor, knee flexors, hip abductors and ankle plantarflexor), agility
performance (19) and moderate positive effects on waist circumference,
weight and walking endurance.(20) Assisted cycling exercise improved
fine manual dexterity (i.e., dominant and bimanual hands) in individu-
als with DS.(21) Hussein (22) found that strength training of lower limb
muscles including hip, knee and ankle effectively enhances forced vital
capacity and peak expiratory flow rate in children with DS. In the same
line of research, improvements on the pulmonary functions (i.e., vital ca-
pacity, forced vital capacity, forced expiratory volume after 1 second and
peak expiratory flow rate) of children with DS were also demonstrated

252
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

after a rowing aerobic exercise regimen.(23)

Different authors (24,25) have shown that progressive resistance
training can improve upper and lower limb muscle strength in young
individuals with DS. Isokinetic training is also an effective training meth-
od to improve muscular strength and postural balance in children with
DS.(26) Finally, a 21-week conditioning training programme combined
with plyometric jumps had positive effects on the acquisition of bone
mass (i.e., total and hip bone mineral content, and total lean mass) in
young individuals with DS.(27) Similar results were reported after a
1-year physical training program in the bone mineral content of adoles-
cents.(28)
Overall, literature has showed that individuals with DS can be en-
gaged in physical exercise programs in a safe and effective way. There is
a research trend regarding the assessment of fitness outcomes focusing
in cardiovascular fitness, muscular strength, body composition and bal-
ance. Research includes individuals with DS with a broad age range (i.e.,
from early childhood to adulthood) and from both genders. Moreover,
researchers discuss their findings considering the importance of exer-
cise programs as a way to improve physical health status and the ability
to perform daily activities. The impact of physical fitness outcomes in
the lives of individuals with DS is related with the decrease of long-term
effects on the pathophysiological consequences of this syndrome, with
the improvement of work tasks performance (e.g., box stacking, pail car-
ry), and the adoption of an active lifestyle. It is possible to conclude that
research about the impact of physical fitness parameters in the sports
performance of athletes with DS is almost none-existent.

Sports training programs

In a sport-specific approach, research also provides insights on how
involvement in sports training programs can affect individuals with DS.
In a mix approach to sport-games (i.e., athletics, handball, football, bas-
ketball or volleyball) and swimming at aerobic levels, young individuals
with DS were able to improve their cardiorespiratory fitness (i.e., VO-
2peak, ventilation and maximum slope).(29) Voznesenskiy, Rivera-Qui-
natoa, Bonilla-Yacelga, and Cedeno-Zamora (30) explored the impact of
horse riding and reported improvements in gross motor function meas-
ures. Water exercises and swimming are also very popular interventions
among individuals with DS. Casey and Emes (31) reported significant
improvements on the respiratory aspects of speech in a group of ado-
lescents enrolled in swim training. Systematic practice of swimming can

253
 sp or t science

also have beneficial effects in stabilizing posture and in achieving sym-

metrical static ability during movements in children with DS.(32)
Carter and Horvat (33) showed that a 10-weeks taekwondo training
successfully improved lower body strength of individuals with DS. A
dance program had positive effects on the static balance with improve-
ments related with the use of visual input associated to centre of pres-
sure (closed and open eyes) being reported.(34) In a combined interven-
tion based on dance and gymnastic exercises, children with DS improved
flexibility and muscle elasticity, force and balance.(35)
In general, the analysis of sports training programs for individuals
with DS clearly emphasizes the lack of research aiming to understand
the specific-sports potential of athletes. Several studies were developed
within a rehabilitation or physical therapy perspective meeting rec-
ommended guidelines for health benefits. Limited sports training and
competition interventions assessed physical, technical or tactical sport-
specific skills needed in a sport performance context. To our knowledge,
only a small group of researchers in the field of swimming contrast with
this research trend. Next, we provide detail information about the ad-
vances on research about performance factors in swimmers with DS.

Performance factors in swimming

One important issue in swimming research with athletes with DS is
the characterization of upper limbs coordination factors implied in the
different swimming techniques. Querido et al. (36) characterized the
index of arm coordination (i.e., time between propulsive phases of the
right and left upper limbs) in front crawl and backstroke and discussed
the possible impact of DS features (i.e., lower muscular strength, higher
body fat mass and anthropometric traits) in the swimmers inter upper
limbs coordination. The relative duration of each upper limbs cycle
phase, in addition with relative duration of propulsive and non-propul-
sive phases were also assessed. Results demonstrated that in both alter-
nated techniques, all swimmers exhibited a catch-up mode coordination
and negative index of coordination values. Hand lag time scores were far
above comparing with those of elite swimmers without disabilities.
In front crawl significant positive correlations were found between
this variable and the push phase as well as the propulsive phase. An
inverse correlation was found between IdC and the non-propulsive
phase. In backstroke, there was an inverse relationship between IdC and
velocity. Querido et al. (36) argued that although swimmers with inter-
national experience were assessed in this study, athletes were not able

254
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

to adapt their upper limbs coordination as usually occurs in swimmers

with higher level of proficiency. This may suggest the presence of techni-
cal faults or physical shortcomings (e.g., higher hand lag time). Finally,
the authors recommended the use of the index of coordination as an
important tool to improve the coaches’ knowledge on underwater cycle
phases (i.e., augmenting the propulsive phases of the upper limbs and,
with this, diminishing the lag time) and to identify technical mistakes.
In an even more detailed analysis on front crawl swimming tech-
nique, Querido et al. (37) assessed velocity, intra-cyclic velocity fluctua-
tion, index of coordination, propelling efficiency and general stroking
parameters in a sample of six international level athletes with DS. The
authors reported a poorer performance on velocity, stroke rate, stroke
length, index of coordination and propelling efficiency of the partici-
pants comparing to the reference scores available in literature for swim-
mers without disabilities. Moreover, it was reported a negative index of
coordination (i.e., catch-up coordination mode) and higher intra-cyclic
velocity fluctuation comparing with swimmers with DS. In contrast,
swimmers with DS exhibited a direct relationship between velocity and
stroke length and between index of coordination and stroke rate, which
is in accordance with the literature for swimmers without disabilities.
Overall results support the perspective that swimmers with DS struggle
to maintain continuous propulsive actions, and lower coordinative de-
velopment and technical efficiency.
In a complementary analysis, Marques-Aleixo et al. (38) examined the
differences in intracycle velocity variation and upper limbs coordina-
tion in front crawl in swimmers with DS in three breathing conditions
(i.e., without breathing, breathing to the preferred side, and breathing to
the nonpreferred side). The main results showed that swimming veloc-
ity was higher without breathing and intracyclic velocity variation was
higher while breathing. Swimmers tended to a catch up upper limbs
coordination mode for both breathing conditions and a superposition
mode when not breathing. Once again, research emphasis difficulties
in upper limbs coordination that negatively impacts swimming perfor-
mance in athletes with DS. Ulate and Campos (39) explored the effects
of a 10-weeks weight resistance training on strength in swimmers with
DS and reported significantly improvements in the maximum strength
of pectoral muscles, dorsal muscles, femoral biceps, quadriceps, and tri-
ceps. After a 2-weeks follow-up, these improvements were maintained
with no significant changes.
Lastly, the importance of swimmers with DS being able to follow a

255
 sp or t science

race strategy has also been approached in the literature.(40) It was

suggested that visual pacer is an interesting instrument for swimming
training providing swimmers with visual feedback about the cadence
techniques (pacing) and allowing a self-evaluation of the performance.
The authors reported that swimmers were not able to take advantage of
short term visual feedback to maintain a pacing strategy. Initially, swim-
mers were not able to repeat the pace when feedback was removed and
as target speed increased they moved from over estimation (i.e., too fast)
to underestimation (i.e., too slow). This indicates that the swimmers
were able to use the feedback but needed more practice. It is possible
to conclude that coaches should systematically use pacer lights system
in training and competition settings to improve the race performance in
swimmers with DS.(40)
All the above-mentioned studies represent an innovative and origi-
nal approach to deepen the knowledge about the performance factors in
swimmers with DS. This because these evidences are the first attempts
to describe the swimming technique of athletes with DS using several bi-
omechanical related parameters. Overall, it is possible to conclude that
elite swimmers with DS with extensive training and competition might
experience disadvantaged in several performance parameters (e.g.,
swimming speed, propulsion, resistance, hydrodynamic drag).

Proper instruments to assess individuals with down syndrome

The existence of valid and reliable instruments to assess the perfor-
mance factors of athletes with DS is an important issue for research and
practice. Coaches, athletes and practitioners in general can make sub-
stantial improvements in the intervention methodologies both in train-
ing and competition settings and, therefore, enhance the sports-skills
ability and the athletic performance. Nevertheless, the lack of appro-
priate instruments for measuring the different performance factors of
athletes with DS remains a problem. This because it is well-known that
many instruments used with the general population are not valid and
reliable to apply to individuals with disabilities.
Therefore, to conduct quality research and overcome the above-
mentioned issue, researchers cannot forget to explore the feasibility,
validity and reliability of the tools they wish to apply on individuals with
DS. Hereinafter, we present a critical analysis about the instruments that
have been properly developed and available to be used in this popula-
tion. Full papers published in English were searched from six major elec-
tronic databases, namely Scopus, Ebsco, Sportdiscus, Web of Science,

256
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Pubmed, and B-on over the last 7 years (2010-2017). Manual search of
journals (e.g., Adapted Physical Activity Quarterly, European Journal of
Adapted Physical Activity, Journal of Intellectual and Developmental Dis-
ability, Journal of Intellectual Disability Research) and scanning of refer-
ence lists to pick up any further studies missed by the electronic search
were also conducted. These sources were chosen, because together,
they provide a complete overview of research concerning DS, sports/
physical activity and exercise. The combination of search terms em-
ployed included: Down syndrome/Trisomy 21, Intellectual Impairment/
Intellectual Disability/Mental Retardation, Physical Activity, Sports, Per-
formance, Training, Competition, Athletes, validation/validity, reliability
and feasibility. Papers were excluded if they met the following exclusion
criteria: i) not related to the scope; ii) papers included in the systematic
review papers already selected and iii) mix sample of individuals with
intellectual disabilities and DS.

Physical fitness
In a broader perspective about the validity and reliability of physical
fitness field-based tests, it was conducted a systematic review to identify
the available protocols to be use in the population with DS.(41) Matching
with the inclusion criteria, several studies were identity related with aer-
obic resistance field test assessments (n=7), strength tests (n=5), balance
tests (n=3), and agility tests (n=1). The authors stated that the half-mile
run-walk (i.e., aerobic endurance), hand-held dynamometer (i.e., mus-
cular strength) and the four-square balance tests (i.e., balance) seem to
be the most valid tests. These findings are very helpful because these
instruments are easy to administrate to a large number of participants
with minimal costs.
Also contributing for the validation of fitness testes in adults with
DS, Boer and Moss (42) explored the test-retest reliability and minimal
detectable change in a group of balance (static balance – standing in
one leg; dynamic balance – walking on a balance bean), flexibility (sit-
and-reach; shoulder stretch), muscular strength and endurance (chair
stand test; handgrip strength; modified curl-up-test; trunk lift; isometric
push-up), aerobic (16-m modified shuttle-run test; 6-min walk test) and
functional task (8-foot get-up-and-go test) tests. All tests demonstrated
adequate feasibility and relative and absolute test–retest reliability in
adults with DS. The authors highlighted the importance of their findings
to monitor performance alterations over time and success of training in-
terventions.

257
 sp or t science

In a specific sub-group of cardiorespiratory fitness tests, the reli-

ability and convergent validity of the 6 minutes run test was assessed.
(43) A group of young adults with DS performed the 6-min run test and
the 16-min shuttle run test twice with a one week-interval between test
and retest. The results of this study indicate that the 6 minutes run test
shows high test/retest reliability and moderate to moderately high con-
vergent validity when performed by adults with DS. Boer and Moss (44)
established criterion-related validity of the 16-metre PACER and six-
minute walk distance tests to VO2peak as well as predictors of VO2peak
in adults with DS. The authors concluded that both 16-metre PACER and
six-minute walk distance are valid field tests for predicting aerobic ca-
pacity, showing moderate-to-strong coefficients of determination and
correlation coefficients in adults with DS.
Aranha, Samuel, and Saxena (45) discussed the importance of having
valid tests to assess balance (static and dynamic) in children with DS. In
this sense, the authors estimate the reliability and sensitivity to change
of the timed standing balance test (i.e., standing on a floor and standing
on a foam pad) using the time required to maintain in four conditions
(i.e., eyes open static, eyes closed static, eyes open dynamic, and eyes
closed dynamic) and concluded that the timed standing balance test is
an easy to administer test and sensitive to change with strong absolute
and relative reliabilities. Therefore, this instrument is very useful to as-
sess balance in children with DS in clinical practice but also in commu-
nity-based interventions such as exercise programs. Similarly, Bandong,
Madriaga, and Gorgon (46) proved the reliability and validity of the Four
Square Step Test (i.e., ability to step in different directions) in this popu-
lation.

Anthropometric measures
The development of an accurate method to assess body composition,
in particular fat percentage (BF%), is an important research topic and
substantial research has been conducted. Bandini, Fleming, Scampi-
ni, Gleason, and Must (47) determined the validity of body mass index
(BMI) to identify excess fatness in youth with DS based on dual-energy
X-ray absorptiometry (DXA) and Freedman’s cut-offs. Overall, the obesity
(≥95th percentile) cut-off performs better than the overweight cut-off
(85–94th percentile) in identifying elevated fatness in adolescents with
DS. The authors suggest that more research is needed to confirm the
cut-off points on a larger population of individuals and to explore the
association with adverse health outcomes.

258
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Similarly, it was estimated the BMI cut-off point for the diagnosis of
obesity in adolescents with DS according to different references for BMI
classification in relation to the BF% measured by DXA.(48) The authors
reported that the criteria that use the BMI for estimating obesity, as well
as references based on specific curves for the population with DS are
associated with BF% estimated by the DXA and concluded that the cut-
off point of z-score above of 2.14 of the World Health Organization pre-
sented better specificity.
In the same line of research, Nascimento et al. (49) determined a pre-
dictive equation, calculated by DXA analysis, for BF% based on BMI for
adolescents with DS. The authors demonstrated that BMI is an effective
measure and provided an accurate equation to estimate BF% developed
from BMI values. Likewise, Freire, Costa, and Gorla (50) also verified that
the BMI exhibited a strong correlation with the BF% in children and ado-
lescents with Down syndrome. However, in both previously mentioned
studies it is highlighted the need to conduct future studies to confirm the
reproducibility of the findings.
Using a similar research design, it was explored the agreement of
BMI-based equations (tested 4 equations) and DXA in determining BF%
but in a sample of adults with DS. (51) The results showed a substan-
tial amount of inter method discrepancy and wide limits of agreement
exist between BMI-BF% equations and DXA. Consequently, the authors
suggested that BMI-based on BF% equations for estimating the level of
BF% should not be used in individuals with DS. Although BMI is the best
know and worldwide used method to determine body fat, it is possible
to verify that findings of different studies in individuals with DS are not
consensual.
In this sense, the validity of the body adiposity index for measuring
BF% in adults with DS has also been explored.(52) The criterion for BF%
was determined by DXA and predicted BF% was estimated by the body
adiposity index method. The authors concluded that the use of the body
adiposity index does not appear to be accurate for estimating BF% in
individuals with DS. Loveday, Thompson, and Mitchell (53) also payed
attention to the best method to determine the BF% in children with DS.
However, the previously mentioned authors explored the accuracy of
bioelectrical impedance using DXA validation and aimed to identify the
existing algorithm best predicts BF% in the population with DS. The au-
thors concluded that bioelectrical impedance can be used to accurately
measure adiposity and the recommended the use of Schaeffer’s algo-
rithm for calculation of BF% in this population.

259
 sp or t science

More recently, it was recommended the use of a predictive equa-

tion to estimate BF% from skinfolds (biceps, triceps, subscapular, su-
praspinale, abdominal, front thigh, and medial calf) in adolescents with
DS.(54) However, Usera, Foley, and Yun (55) demonstrated that skinfolds
and anthropometric girth measurements are not a feasible way to assess
body composition and highlight the need to develop new equations spe-
cifically for in individuals with DS.

Physiological measures
Boer (56) explored the accuracy of the Fernhall’s prediction equation
for the determination of maximum heart rate in adults with DS using a
maximal aerobic test and concluded that it was not accurate for the age
range of his sample. It was suggested that more research is needed to
develop different prediction equations for more specific age and body
mass index categories for individuals with DS. On the other hand, Chen,
Ringenbach, Snow, and Hunt (57) explored the validity of a pictorial
rate of perceived exertion scale, based on heart rate index in a walking
protocol, to monitor exercise intensity in young adults with DS. Results
indicated significant positive relations between heart rate and rate of
perceived exertion in most of the sample. The participants were able to
perceive and report a subjective estimation of physical exertion, which
was reflective of the change of heart rate. Therefore, the authors con-
cluded that the rate of perceived exertion scale is useful to monitor exer-
cise performance in adults with DS.
Agiovlasitis et al. (58) critically analysed the use of the American
College of Sport Medicine equation to predict gross-VO2 during over-
ground walking in individuals with DS. The results demonstrated that
the American College of Sport Medicine formula under-estimated gross-
VO2 across speeds showing altered curvilinear gross-VO2 to speed rela-
tionship. Consequently, the previously mentioned authors tested a new
equation, specifically developed for individuals with DS, and demon-
strated that this equation was accurate and could be used for prescrib-
ing over-ground walking intensities. These findings are important to
establish appropriate exercise intensities, according to the participants’
needs that will improve the quality of the exercise programs offered to
individuals with DS.
Finally, Seron and Greguol (59) conducted a systematic review to
determine the validity and reliability of protocols to assess VO2max in
young individuals with DS. Matching with the inclusion criteria, the au-

260
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

thors selected 18 tests, 14 measured the VO2peak in a laboratory and 4

were field tests. Concerning the tests performed in a laboratory, only 8
used validated protocols (i.e., treadmill and rowing) for individuals with
DS. Of the field tests, only one was not validated. The authors concluded
that most of the selected studies used maximal tests. A large number
of studies used tests which had not been specifically validated for the
evaluated population and a few number of studies used field tests to
evaluate VO2.

Physical activity levels

To proper assess the physical activity levels of individuals with DS,
researchers explored the accuracy of pedometers and accelerometers in
this population. Agiovlasitis, Beets, Lamberth, Pitetti, and Fernhall (60)
examined the accuracy of a pedometer with a tri-axial accelerometer
mechanism - Walk4life MVPa pedometer, in measuring steps at different
walking speeds comparing individuals with and without DS. The authors
demonstrated that this instrument measures steps with high accuracy
at the preferred walking speed and at speeds ≥1.0 m s-1 in both groups
of individuals. However, accuracy is compromised at slower speeds in
the group with DS. The authors suggest that professionals can use with
confidence the Walk4life MVPa pedometer for monitoring the steps at
walking speeds ≥1.0 m s-1 in individuals with DS. Below 1.0 m s-1 walk-
ing speeds, the pedometer use is compromised.
A cross-validation and reliability study comparing two triaxial ac-
celerometers (the SenseWear and RT3) to a criterion measure in young
adults with DS concluded that both instruments can be used to monitor
physical activity.(61) However, limitations were observed regarding the
use of the accelerometers in energy expenditure estimates.
After an extensive and detailed analysis of the literature about the ex-
istence of valid and reliable instruments to assess individuals with DS it
is possible to conclude that: i) this topic is in the agenda of the research-
ers and the vast majority of the studies were published very recently;
ii) there is a research trend towards the exploration of physical fitness
and anthropometric tests that require simple equipment, time-saving,
and low cost administration; iii) most studies highlight the importance
of valid and reliable tools to improve the design of lifestyle interventions
(e.g., exercise), skills for daily life, and clinical practice; and iv) future re-
search needs to replicate and confirm the present findings because in
some cases outcomes from different studies are contrary.
Although most of the instruments previously mentioned can be used

261
 sp or t science

in a competitive sport context to assess athletes with DS, we could not

identify any study developing validity or reliability procedures about
instruments related with specific physical, technical or technical sport
skills.

Sports organizations for athletes with down syndrome

From “Special Olympics” to “Sports Union for Athletes with Down syn-
drome”
Traditionally, individuals with DS had been included in sports organi-
zations for individual with intellectual disabilities. In this context, Spe-
cial Olympics International is a global non-profit sporting organization
and the largest one (over 4.7 million athletes) serving individuals with
developmental disabilities, including individuals with intellectual dis-
abilities and DS.(62) The Special Olympics was originally founded in the
United States of America in 1968 and is represented in 169 countries with
a range of 35 individual and team sports. Special Olympics is well-known
by the diversity of programs offered such as Motor Activity Training Pro-
gram, Youth Athletes Program, Unified Sports, Sports Partnerships and
Partner Clubs.
According to the Special Olympics,(62) its mission is to provide a
year-round sports training and athletic competition in a variety of sports
for children and adults with intellectual disabilities. The goal is to offer
sport opportunities to develop physical fitness, demonstrate courage,
experience joy and participate in a sharing of gifts, skills and friendship
with their families, other athletes and the community. Benefits regard-
ing the participation in Special Olympics programs are associated to the
improvement self-esteem and self-confidence, social support and social
acceptance, as well as, positive self-perceptions.(63) As such, elite per-
formance is not the main focus of Special Olympic organization.
Since 1986, the International Federation for Intellectual Disability
Sport (INAS) assumed the purpose of managing elite sport competition
for athletes with intellectual disabilities worldwide, allowing athletes to
achieve excellence in sport and high-level competition. INAS is repre-
sented in 80 countries with more than 300,000 athletes with intellectual
disability participating in 14 sports. INAS is responsible for the Global
Games (2004-2019), the world largest elite sport event for athletes with
intellectual disabilities participating in athletics, swimming, table ten-
nis, rowing, basketball, futsal, tennis, cycling and taekwondo.(64) This
Federations is a member of the International Paralympic Committee
(IPC) and in the last decades developed intensive collaborative work

262
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

with the IPC in the field of the classification system to allow the re-entry
of athletes with intellectual disability in London Paralympic Games.(65)
Over the edge of IPC-INAS, athletes with intellectual disabilities par-
ticipated in the Atlanta (1996), Sydney (2000), London (2012) and Rio
(2016) Paralympic Games. In Rio, athletes with intellectual disabilities
participated in athletics, swimming and table tennis competitions.(64)
Athletes with DS are eligible to compete in both INAS (i.e., Global Games)
and IPC (i.e., Paralympic Games) competitions in the same category as
the athletes with intellectual disability. The IPC advocates that athletes
with DS must be included in a single disability category meeting the
sport-specific criteria for athletes with intellectual disability. This be-
cause Paralympic classification system does not include categories for
specific type of disability but only three major areas, namely physical,
visual and intellectual impairment.(65)
Although classification rules include all athletes with an intellectual
disability (athletes with DS included), as we have previously stated there
are research evidences that athletes with DS are not able to reach the
same performance standards that athletes with intellectual disabilities
in an equal elite sports context. The specific combination of intellectual,
physical and physiological features of DS seems to have a negative im-
pact on high-level performance when compared with athletes with intel-
lectual disabilities. For example, Smedley (66) argued that atlanto-axial
instability and diving are crucial concerns for many swimmers struggling
to meet the regulation for competitive swimming.
In addition, practical evidences also support this fact. For instance,
we developed a detailed analysis of the swimming events results on the
Global Games (2004, 2009, 2011 and 2015) and Paralympic Games (1996,
2000, 2012 and 2016) and we could not identify any swimmer with DS
reaching a final phase, even less, being medalled. It seems that athletes
with Down syndrome have been outclassed by athletes with intellec-
tual disabilities without other associated comorbidities. Despite of, at
the request of the 2013 INAS General Assembly and the membership to
re-consider the traditional single ‘intellectual disability’, INAS has been
considering the feasibility and benefits of introducing additional eligibil-
ity groups into competition programme, specifically for DS. In this sense,
INAS is developing a pilot project, included in INAS Strategic Plan 2017-
2020, to monitor and evaluate these new eligibility groups in next 2017
INAS World Table Tennis and Swimming Championships.(64)
In face of the previously mentioned challenges for athletes with DS, in
2012 a new sports Union was formed to support athletes with DS aiming

263
 sp or t science

to achieve the highest levels of sporting excellence, named Sports Union

for Athletes with Down syndrome (SU-DW).(67) The SU-DW represents
and co-ordinates 7 sports and organizations for individuals with DS,
namely the International Athletics Association for athletes with Down
Syndrome, the Down Syndrome International Swimming Organization,
the Football International Federation for players with Down Syndrome,
the Down Syndrome International Gymnastics Organization, the Inter-
national Table Tennis Association for players with Down Syndrome, the
Judo for Down Syndrome players and the Skiing for people with Down
Syndrome. In 2016, Italy, occurred the first World Trisome Games for all
the sports represented by the SU-DW. The current mission of this organi-
zation is the recognition of a new class, specific for athletes with DS and
apart from the intellectual disability classification, in Tokyo’s Paralympic
Games in 2020 by the Paralympic Committee.(67)

Practical recommendations for coaches and other practitioners

Having the above-mentioned rational in consideration, we suggest
a group of practical recommendations to help coaches to enhance the
performance of athletes with DS. The race or game analysis has been
widely accepted in the scientific and coaching community to understand
the athlete’s behaviour in competition. However, little has been done
at international events for DS athletes. Coaches should look up for this
kind of approach and, although methodological issues make it difficult
to compare different studies, videotaping their athletes in competition
allows the coaches to, at least, analyse and observe behaviour and per-
formance in different competitions.
Athletes with DS seem to improve their physical fitness, however,
their anthropometric characteristics, lack of strength and reduced car-
diovascular capacity, when compared to athletes without disability or
athletes with intellectual disability, seems to impact on their sports per-
formance. Nevertheless, we emphasize the idea that competitive sports
practice may help individuals attaining a better physical fitness profile.
Therefore, coaches should enable physical fitness training and promote
activities directed towards performance.
Coaches must find training strategies to promote their understanding
of the technical gesture as well as the race or game strategy (e.g., videos,
manipulation, filming and after analysis). It is also important to establish
realistic goals in accordance with the athletes (i.e., make them part of
the process). As part of a team, coaches should also promote opportuni-
ties for athletes with DS to interact with all the persons from the Club

264
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

(e.g., teammates, other coaches, directors), with or without disability,

creating some tasks or games together. It is also our belief that, at some
points of the season, coaches should enable their athletes to have differ-
ent experiences and do other things together such as try other sports,
have lunch or dinner together and go to the movies.

Conclusions
In general, this review putted in evidence the need to improve knowl-
edge about physical fitness, technical and technical sport-specific skills
influencing the performance of athletes with DS. Exception was made
for the swimming area where it was possible to identify a small but con-
sistent body of research about the biomechanical factors influencing
swimming performance. Moreover, there is an absence of studies about
research methodologies and, specifically, assessment tools valid and re-
liable to apply in both training in competition settings. It is crucial that
more research on athletes with DS is available, particularly for coaches
to use the information in the training process across different sports
and from a range of scientific disciplines (e.g., technical, biomechanical,
physiological, psychological variables or nutritional practices).
Trends in the results highlighted the importance of exercise to im-
prove the efficiency in performing everyday activities (e.g., workplace
activities) that typically emphasise physical skills. Exercise is valued due
to its impact on reducing the potential health risks associated with low
fitness and sedentary behaviour, on maintaining an active lifestyle, on
improving functional independence and quality of life in individuals
with DS.
Lastly, athletes with DS struggle to be recognized as athletes on a
single and specific competition class or category. This a very hot topic
related with the inclusion of athletes with DS in Paralympic Games. Now-
adays there are intensive discussions about athletes with DS eligibility
criteria that we believe will have an impact on the sports regulations of
the major sports organizations for individuals with disabilities.

References
1. Van de Vliet P, Rintala P, Frojd K, Verellen J, Van Houtte S, Daly DJ, et al.
Physical fitness profile of elite athletes with intellectual disability. Scand J
Med Sci Sports. 2006;16(6):417-425. doi: 10.1111/j.1600-0838.2006.00539.x.
2. Balic M, Mateos E, Blasco C, Fernhall B. Physical fitness levels of physically
active and sedentary adults with Down syndrome. Adapt Phys Activ Q.
2000;17:310-321.
3. Perán S, Gil JL, Ruiz F, Fernandez-Pastor V. Development of physical re-

265
 sp or t science

sponse after athletics training in adolescents with Down’s syndrome.

Scand J Med Sci Sports. 1997;7(5):283-288.
4. World Health Organization [Internet]. Genomic resource centre – Genes
and Human disease. [cited 2017 May 19]. Available from http://www.who.
int/genomics/public/geneticdiseases/en/index1.html
5. Bull MJ, Committee on Genetics. Clinical report – Health supervision for
children with Down syndrome. Pediatrics. 2011;128(2):393-406. doi:10.1542/
peds.2011-1605.
6. Vicari S. Motor development and neuropsychological patterns with Down syn-
drome. Behav Genet. 2006;36(3):355–364. doi: 10.1007/s10519-006-9057-8.
7. Myśliwiec A, Posłuszny A, Saulicz E, Doroniewicz I, Linek P, Wolny T, Knapik
A, Rottermund J, Żmijewski P, Cieszczyk P. Atlanto-Axial Instability in
People with Down’s Syndrome and its Impact on the Ability to Perform
Sports Activities – A Review. J Hum Kinet. 2015;48(1):17-24. doi: 10.1515/
hukin-2015-0087.
8. Mansoura E, Yaacouba JJ, Bakounya Z, Assia A, Ghanema I. A podoscop-
ic and descriptive study of foot deformities in patients with Down syn-
drome. Orthop Traumatol Surg Res. 2017;103(1):123–127. doi: 10.1016/j.
otsr.2016.10.001.
9. Roizen NJ, Magyar CI, Kuschner ES, Sulkes SB, Druschel C, van Wijngaarden
E, et al. A community cross-sectional survey of medical problems in 440
children with Down syndrome in New York State. J Pediatr. 2014;164(4):871–
875. doi: 10.1016/j.jpeds.2013.11.032.
10. Pitetti K, Baynard T, Agiovlasitis S. Children and adolescents with Down
syndrome: physical fitness and physical activity. J Sport Health Sci.
2013;2(1):47–57. doi: 10.1016/j.jshs.2012.10.004.
11. Pitetti KH, Climstein M, Mays MJ, Barrett PJ. Isocinetic arm and leg strength
of adults with Down syndrome: A comparative study. Arch Phys Med Reha-
bil. 1992;73:847-850.
12. Geijer JR, Stanish HI, Draheim CC, Dengel DR. Bone Mineral Density in Adults
With Down Syndrome, Intellectual Disability, and Nondisabled Adults. Am
J Intellect Dev Disabil. 2014;119(2):107-114. doi: 10.1352/1944-7558-119.2.107.
13. Gonzalez-Agüero A, Ara I, Moreno LA, Vicente-Rodriguez G, Casajus JA. Fat
and lean masses in youths with Down syndrome: gender differences. Res
Dev Disabil. 2011;32(5):1685-1693. doi: 10.1016/j.ridd.2011.02.023.
14. Melville CA, Cooper SA, Morrison J, Allan L, Smiley E, Williamson A. The
prevalence and determinants of obesity in adults with intellectual dis-
abilities. J Appl Res Intellect Disabil. 2008;21(5):425–437. doi: 10.1111/j.1468-
3148.2007.00412.x.
15. Izquierdo-Gomez R, Veiga OL, Villagra A, Diaz-Cueto, Up&Down
Study Group. Correlates of sedentary behaviour in youths with Down
syndrome: the UP&DOWN study. J Sports Sci. 2015;33(14)1-11. doi:
10.1080/02640414.2014.994660.
16. Boer PH, Moss SJ. Effect of continuous aerobic vs. interval training on
selected anthropometrical, physiological and functional parameters of
adults with Down syndrome. J Intellect Disabil Res. 2016;60(4):322-334.
doi: 10.1111/jir.12251.
17. Ordonez FJ, Rosety MA, Camacho A, Rosety I, Diaz AJ, Fornieles G, Garcia

266
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

N, Rosety-Rodriguez M. Aerobic training improved low-grade inflamma-

tion in obese women with intellectual disability. J Intellect Disabil Res.
2014;58(6):583-590. doi: 10.1111/jir.12056
18. Chen C-C, Rigenbach DRS, Snow M. Treadmill walking effects on grip strength
in young men with Down syndrome. Res Dev Disabil. 2014;35(2):288–293.
doi: 10.1016/j.ridd.2013.10.032.
19. Lin H-C, Wuang Y-P. Strength and agility training in adolescents with Down
syndrome: A randomized controlled trial. Res Dev Disabil. 2012;33(6):2236–
2244. doi: 10.1016/j.ridd.2012.06.017.
20. Shields N, Taylor NF. The feasibility of a physical activity program for young
adults with Down syndrome: A phase II randomised controlled trial. J Intel-
lect Dev Disabil. 2015;40(2):115–125. doi: 10.3109/13668250.2015.1014027.
21. Chen C-C, Rigenbach SDR, Albert AR. Assisted Cycling Exercise Improves
Fine Manual Dexterity in Persons with Down’s syndrome. J Appl Res Intel-
lect Disabil. 2014;27(3):264–272. doi:10.1111/jar.12061.
22. Hussein ZA. Strength training versus chest physical therapy on pulmo-
nary functions in children with Down syndrome. Egypt J Med Hum Genet.
2017;18(1):35–39. doi: 10.1016/j.ejmhg.2016.02.008.
23. El Kafy EMA, Helal OF. Effect of rowing on pulmonary functions in children
with Down syndrome. Pediatr Phys Ther. 2014;26(4):437-445. doi: 10.1097/
PEP.0000000000000072.
24. Shields N, Taylor NF. A student-led progressive resistance training program
increases lower limb muscle strength in adolescents with Down syndrome:
A randomised controlled trial. J Physiother. 2010;56(3):187-193.
25. Shields N, Taylor NF, Wee E, Wollersheim D, O’Shea SD, Fernhall B. A com-
munity-based strength training programme increases muscle strength
and physical activity in young people with Down syndrome: A randomised
controlled trial. Res Dev Disabil. 2013;34(12):4385-4394. doi: 10.1016/j.
ridd.2013.09.022.
26. Eid MA, Aly SM, Huneif MA, Ismail DK. Effect of isokinetic training on muscle
strength and postural balance in children with Down’s syndrome. Int J Re-
habil Res. 2017;40(2):127-133. doi: 10.1097/MRR.0000000000000218.
27. González-Agüero A, Vicente-Rodríguez G, Gómez-Cabello A, Ara I, Moreno
LA, Casajús JA. A 21-week bone deposition promoting exercise programme
increases bone mass in young people with Down syndrome. Dev Med Child
Neurol. 2012;54(6): 552-556. doi: 10.1111/j.1469-8749.2012.04262.x.
28. Ferry B, Gavris M, Tifrea C, Bembea M, Courteix D. The bone tissue of chil-
dren and adolescents with Down syndrome is sensitive to mechanical stress
in certain skeletal locations: A 1-year physical training program study. Res
Dev Disabil. 2014;35(9):2077-2084. doi: 10.1016/j.ridd.2014.05.004.
29. Casajús JA, Pueyo D, Vicente-Rodríguez G, González-Agüero A. Improve-
ments in cardiorespiratory fitness through aerobic training in young
people with Down’s syndrome: A longitudinal study. Apunts Med Esport.
2012;47(174):49-54. doi: 10.1016/j.apunts.2011.06.003.
30. Voznesenskiy S, Rivera-Quinatoa JA, Bonilla-Yacelga KA, Cedeno-Zamora
MN. Do equine-assisted physical activities help to develop gross motor
skills in children with the Down syndrome? Short-term results. Procedia
Soc Behav Sci. 2016; 233:307-312. doi: 10.1016/j.sbspro.2016.10.140.

267
 sp or t science

31. Casey AF, Emes C. The Effects of Swim Training on Respiratory Aspects of
Speech Production in Adolescents With Down Syndrome. Adapt Phys Activ
Q. 2011;28(4):326-341.
32. Popa C-E, Galeru O. Study regarding the improvement of postural control in
children who have Down syndrome through swimming. Gymn - Sci J Educ
Sports Health. 2012;13(2):85-98.
33. Carter K, Horvat M. Effect of Taekwondo Training on Lower Body Strength
and Balance in Young Adults with Down syndrome. J Policy Pract Intellect
Disabil. 2016;13(2):165–172. doi: 10.1111/jppi.12164.
34. Gutiérrez-Vilahú L, Massó-Ortigosa N, Costa-Tutusaus L, Guerra-Balic M,
Rey-Abella F. Effects of a Dance Program on Static Balance on a Platform
in Young Adults With Down Syndrome. Adapt Phys Activ Q. 2016;33(3):233-
252. doi: 10.1123/APAQ.2015-0048.
35. Moraru C, Hodorca RM, Vasilescu D. The Role of Gymnastics and Dance in
Rehabilitating Motor Capacities in Children with Down syndrome. Sp Soc
Int J Ph Ed. 2014;14:102-112.
36. Querido A, Marques-Aleixo I, Figueiredo P, Seifert L, Chollet D, Vilas-Boas
JP, et al. Front crawl and backstroke arm coordination in swimmers with
Down syndrome. In: Kjendlie P-L, Stallman RK, Cabri J, editors. Book of
Proceedings of the XI Symposium on Biomechanics and Medicine in Swim-
ming; 2010 Jun 16-19; Oslo, Norway: Norwegian School of Sport Science;
2010. p. 157-159.
37. Querido A, Marques-Aleixo I, Figueiredo P, Soares-Miranda L, Vilas-Boas JP,
Corredeira R, et al. Coordinative characterization of front crawl swimmers
with Down syndrome. Arch Exerc Health Dis. 2010;1(2):58-62.
38. Marques-Aleixo I, Querido A, Figueiredo P, Vilas-Boas JP, Corredeira R, Daly
D, et al. Intracyclic Velocity Variation and Arm Coordination Assessment in
Swimmers with Down Syndrome. Adapt Phys Activ Q. 2013;30(1):70-84.
39. Ulate FV, Campos AM. Effects of Weight Resistance Training on Swimmers
with Down Syndrome. Retos: Nuevas Tendencias en Educación Física, De-
portes y Recreación. 2011;1(19):10-14.
40. Querido A, Corredeira R, Vilas-Boas JP, Daly D, Fernandes RJ. Can swim-
mers with Down syndrome follow a visual pacer in an incremental proto-
col? In: Australian Institute of Sports, editors. Proceedings of the XII Inter-
national Symposium on Biomechanics and Medicine in Swimming, 2014
April 28-May 2; Canberra, Australia: Australian Institute of Sports (AIS);
2014. p. 325-328.
41. Ayán-Perez C, Martínez-Lemos I, Lago-Ballesteros J, Carral JMC, Loira-
Caminã N. Reliability and Validity of Physical Fitness Field-Based Tests
in Down syndrome: A Systematic Review. J Policy Pract Intellect Disabil.
2016;13(2):142-156. doi: 10.1111/jppi.12155.
42. Boer PH, Moss SJ. Test–retest reliability and minimal detectable change
scores of twelve functional fitness tests in adults with Down syndrome. Res
Dev Disabil. 2016;48:176-185. doi: 10.1016/j.ridd.2015.10.022.
43. Ayán-Perez C, Martínez-Lemos RI, Cancela-Carral JM. Reliability and con-
vergent validity of the 6-min run test in young adults with Down syndrome.
Disabil Health J. 2017;10(1):105-113. doi: 10.1016/j.dhjo.2016.07.004.
44. Boer PH, Moss SJ. Validity of the 16-metre PACER and six-minute walk test

268
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

in adults with Down syndrome. Disabil Rehabil. 2016;38(26):2575-2583. doi:

10.3109/09638288.2015.1137982.
45. Aranha VP, Samuel AJ, Saxena S. Reliability and sensitivity to change of the
timed standing balance test in children with Down syndrome. J Neurosci
Rural Pract. 2016;7(1):77-82. doi: 10.4103/0976-3147.165412.
46. Bandong ANJ, Madriaga GO, Gorgon EJR. Reliability and validity of the Four
Square Step Test in children with cerebral palsy and Down syndrome. Res
Dev Disabil. 2015;47:39-47. doi: 10.1016/j.ridd.2015.08.012.
47. Bandini LG, Fleming RK, Scampini R, Gleason J, Must A. Is body mass index
a useful measure of excess body fatness in adolescents and young adults
with Down syndrome? J Intellect Disabil Res. 2013;57(11):1050-1057. doi:
10.1111/j.1365-2788.2012.01605.x.
48. Samur-San-Martin JE, Goncalves EM, Bertapelli F, Mendes RT, Guerra-
Junior G. BMI cutoff point estimation as obesity diagnostic criteria in Down
syndrome adolescents. Nutr Hosp. 2016;33(5):1090-1094. doi: 10.20960/
nh.57.
49. Nascimento EF, Souza MK, Rosa TS, Melo GLR, Soares BRA, Sousa FEV, et
al. Predictive Equation for Fat Percentage Based on Body Mass Index for
Adolescents with Down Syndrome. JEPonline. 2016;19(4):77-84.
50. Freire F, Costa LT, Gorla JI. Indicators of obesity in young people with
Down syndrome. Motricidade. 2014;10(2): 2-10. doi: 10.6063/motrici-
dade.10(2).1196.
51. Esco MR, Nickerson BS, Bicard SC, Russell AR, Bishop PA. Agreement of
BMI-Based Equations and DXA in Determining Body-Fat Percentage in
Adults with Down syndrome. Adapt Phys Activ Q. 2016;33(1):89-96. doi:
10.1123/APAQ.2014-0240.
52. Nickerson BS, Esco MR, Bicard SC, Russell AR, Williford HN, Shaefer G. Va-
lidity of the body adiposity index in adults with Down syndrome. Res Dev
Disabil. 2015;38:92-96. doi: 10.1016/j.ridd.2014.12.010.
53. Loveday SJ, Thompson JMD, Mitchell EA. Bioelectrical impedance for
measuring percentage body fat in young persons with Down syndrome:
validation with dual-energy absorptiometry. Acta Paediatr. 2012;101(5):491-
495. doi:10.1111/j.1651-2227.2012.02821.x.
54. González-Agüero A, Matute-Llorente A, Gómez-Cabello A, Vicente-
Rodríguez G, Casajús JA. Percentage of body fat in adolescents with Down
syndrome: Estimation from skinfolds. Disabil Health J. 2017;10(1):100-104.
doi: 10.1016/j.dhjo.2016.05.013.
55. Usera PC, Foley JT, Yun J. Cross-Validation of Field-Based Assessments of
Body Composition for Individuals with Down syndrome. Adapt Phys Activ
Q. 2005;22(2):198-206.
56. Boer PH. Accuracy of the prediction equation for the determination of
maximum heart rate in adults with Down syndrome. J Intellect Disabil Res.
2017;61(1): 83-88. doi: 10.1111/jir.12345.
57. Chen C-C, Ringenbach SDR, Snow MS, Hunt LM. Validity of a pictorial
Rate of Perceived Exertion Scale for monitoring exercise intensity in
young adults with Down syndrome. Int J Dev Disabil. 2013;59(1):1-10. doi:
10.1179/2047387712Y.0000000005.
58. Agiovlasitis S, Motl RW, Ranadive SM, Fahs CA, Yan H, Echols GH, et al. Pre-

269
 sp or t science

diction of oxygen uptake during over-ground walking in people with and

without Down syndrome. Eur J Appl Physiol. 2011;111(8):1739-1745. doi:
10.1007/s00421-010-1812-0.
59. Seron BB, Greguol M. Assessment protocols of maximum oxygen consump-
tion in young people with Down syndrome – A review. Res Dev Disabil. 2014;
35(3): 676–685. doi: 10.1016/j.ridd.2013.12.008.
60. Agiovlasitis S, Beets MW, Lamberth J, Pitetti, KH, Fernhall B. Accuracy of a
piezoelectric pedometer in persons with and without Down syndrome. Res
Dev Disabil. 2016;59:73-79. doi: 10.1016/j.ridd.2016.08.003.
61. Peiris CL, Cumming TB, Kramer S, Johnson L, Taylor NF, Shields N. Ac-
celerometer use in young people with Down syndrome: A preliminary
cross-validation and reliability study. J Intellect Dev Disabil. 2016;1-12. doi:
10.3109/13668250.2016.1260100.
62. Special Olympcis [Internet]. The Special Olympics – What we do. [cited 2017
May 5]. Available from: http://www.specialolympics.org/Sections/What_
We_Do/What_We_Do.
63. Dykens EM, Cohen DJ. Effects of Special Olympics International on social
acceptance in persons with mental retardation. J Am Acad Child Adolesc
Psychiatry. 1996;35(2):223-229.
64. INAS: The International Federation for Intellectual Impairment Sport [In-
ternet]. Promoting sport worldwide for athletes with an intellectual dis-
ability, autism and Down’s Syndrome. [cited 2017 July 1]. Available from:
http://www.inas.org/.
65. IPC: Paralympic Movement [Internet]. Official website of the Paralympic
Movement. [cited 2017 July 1]. Available from: https://www.paralympic.org/.
66. Smedley G. Making a splash. Learn Disabil Pract. 2007;10(2):24-25.
67. SU-DS: Sports Union for athletes with Down syndrome [Internet]. About Us
– Mission statement. [cited 2017 July 1]. Available from: http://www.su-ds.
org/about-us/.

270
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Aerobic Interval Versus Continuous Training at

Low Volume and Vigorous Intensity
Laura Castro De Garay1, Tony Meireles Dos Santos2, Matthew
Rhea3, Victor Machado Reis1

Introduction
Different aerobic exercises such as walking (1), cycling (2) and running
(3) are often used in physical exercise programs for improving physical
health and performance. Running on the treadmill is an aerobic exer-
cise activity commonly performed by active individuals in fitness centres
with participants often following two primary styles: (a) continuous, un-
interrupted exercise characterized by low-moderate intensity and long
period (4) and (b) interval, with exercise efforts of moderate-high inten-
sity, interspersed with periods of active or passive recovery (5, 6).
The physiological adaptations induced by the aerobic training de-
pend mainly on the relative intensity of the exercise, stressing that vigor-
ous aerobic exercise have positive effects greater than moderate, espe-
cially in improving aerobic power (7) and neuromuscular capacity (8).
But, moderate aerobic training results in an improvement in health sta-
tus (9), adherence (10) and weight loss (11). Often individuals attempt to
perform higher volumes of training than necessary, resulting in a train-
ing intensity below optimal (12). Thus, the exercise intensity pre-deter-
mination is warranted for training prescription, especially when individ-
ualized fitness information is available (13).Although well documented
in the literature that intense exercise have better performance benefits
when compared to moderate intensity (14), either for untrained (9) or
trained individuals (15), there are few contributions (15, 16) that address
the effectiveness of a training program with an interval (IT) and continu-
ous (CT) mode of training at the same total distance, frequency and aver-
age intensity for a period that is sufficient to enable adaptations.
A previous study that compared the effect of IT (with active and pas-
sive recovery) and CT, with the same training volume, found greater ef-

1. Research Center in Sports Sciences, Health Sciences & Human Development

(CIDESD); University of Trás-os-Montes and Alto Douro; Vila Real, Portugal
2. Department of Physical Education; Federal University of Pernambuco; Re-
cife, PE – Brazil
3. Human Movement Program; AT Still University; Mesa, Arizona, USA

271
 sp or t science

fects in IT with a total volume less than 30-min (17). According to the
authors, the duration of training depends on the aerobic fitness level
and the individual’s ability to perform high intensity efforts, followed by
a specific recovery. Another study of ten weeks with cardiac patients on
the treadmill, observed an increase in the total work performed when
IT was performed at high intensities (80 - 90% VO2peak) compared to
CT at moderate intensity (50 - 60% VO2peak) (18). Also, Nybo et al. (3)
compared the effect of a vigorous IT with traditional intervention pro-
grams (CT and strength training) in untrained young individuals. Twelve
weeks of training showed significant increases for IT on VO2max (14 ±
2%) when compared with the other two modes of intervention. Helgerud
et al. (14) examined eight weeks (3 d∙wk-1) of aerobic training and found
improvements in VO2max in two IT groups (long interval, 4-min exercise
interspersed with 4-min rest vs. short-interval, 15 s exercise interspersed
with 15 s rest, both performed at 90 - 95%HRmax) when compared with
two continuous groups (45-min at 70% HRmax; ~24-min at 85% HRmax).
Thus, the aim of this study was to compare physiologic adaptations
to continuous and interval running at low volumes (4.2 km) and vigorous
intensities performed over 14 weeks. We hypothesized that 14 weeks of
interval training at the same volume as continuous training would result
in greater improvements in physiological indicators of aerobic fitness.

Material and methods

Participants
Initially 35 subjects were recruited for participation in the study. Par-
ticipants were randomly assigned to either an interval training (IT; n =
13) or continuous training (CT; n = 13). The subjects from control group
(CG; n = 9) were selected between those who reported unavailability to
complete all 46 training sessions and they did their normal daily physical
activities. Two individuals during maximal exercise test (VO2) showed
clinical impairment with test being interrupted by the doctor and a third
individual, presented a plantar fasciitis in the early familiarization. They
were all removed from the study. The values of pre-training maximal
test showed no differences in maximal VO2 between the three groups.
Thus, thirty-two adult males participated in the study. All subjects were
healthy with no known musculoskeletal or cardiorespiratory disease and
exercised regularly, with at least one year in aerobic activities (moderate
to vigorous intensities), but not currently involved in a training program.
They were instructed to continue normal daily activities and to refrain
from beginning any other exercise training program until the comple-

272
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

tion of the study. Subjects were instructed to refrain from caffeine, heavy
meals or heavy exercises 24 h before each testing session. Before and
after the 14-wk intervention period, the subjects completed a total of
four visits in the following order: pre-testing procedures (1) answer the
questionnaires and explanations to the procedures, (2) anthropometrics
measures, (3) maximal test; post-testing procedures (4) maximal test. All
tests and training program were performed at the same time of the day
in a controlled laboratory condition (temperature = 19-22 o C; relative
humidity = 50-60%).
All subjects completed a Physical Activity Readiness Questionnaire,
and signed an informed consent before participation, which thoroughly
explained the training and testing procedures. The study was performed
according to the Declaration of Helsinki, and the protocol was approved
by Ministry of Health – The Human Ethics Committee of Clementino Fra-
ga Filho Hospital of Rio de Janeiro Federal University, Rio de Janeiro,
Brazil (no 58659/02471312.8.0000.5257). Participants characteristics are
shown in Table 1.

IT (N = 13) CT (N = 11) CG (N = 8)
Age 30.2 ± 0.8 30.5 ± 1.0 29.1 ± 4.4
Height 174 ± 0.1 175 ± 0.1 175 ± 0.1
BM 77.6 ± 13.5 76.6 ± 3.7 76.4 ± 10.9
FM 14.3 ± 6.0 11.9 ± 3.9 12.5 ± 4.9
BMI 25.6 ± 3.64 25.14 ± 1.4 24.8 ± 2.2

Tabel 1. Body mass, fat mass and body mass index for the two training groups
and control.

Data are presented as mean ± SD for the three groups. IT, interval training; CT,
continuous training; CG, control group. Age (years); Height (cm); BM, body
mass (kg); FM, fat mass (%); BMI, body mass index (kg∙m-2).

Anthropometry
The subjects underwent a set of anthropometric assessments, which
followed the norms of the International Society for Advancement of
Kinanthropometry – ISAKA (Stewart, Marfell-Jones, Olds, & de Ridder,
2011). Anthropometric measurements (body mass; fat mass and BMI)
were collected. The seven skinfolds measurement were introduced in
the equation proposed by Jackson and Pollock (1978) and the densities
obtained by the equations were converted in body fat using the formula

273
 sp or t science

proposed by Siri (1961). Skinfold thickness was obtained by Sanny® pro-

fessional skinfold calliper. The Body Mass Index (BMI) was calculated as
the ratio between the weight and the square of the height (BMI = kg∙m-
2), in accordance with the recommendations of the World Health Organi-
zation - WHO (2008). To measure weight and height, a scale with a stadi-
ometer were used (Filizola, S. Paulo, Brazil).
Maximal VO2 test
Electrodes (Micromed) were placed at the manubrium, right and
left iliac crest for measured heart rate (HR) at derivation CM5, with the
individual on the treadmill (Inbrasport Master Super ATL, Porto Alegre,
Brazil) HR values were visualized through Elite software (Micromed Bio-
technology, Brasília, Brazil). The individual was connected to a meta-
bolic cart (VO2000, Aerosport, Medgraphics, St. Paul, Minnesota) and gas
samples were collected and measured every 10 second during the test.
The participants were submitted to a ramp protocol with an initial veloc-
ity of 8.0 km×h-1 (0% of inclination), with a 0.02% inclination increase
every 10 s until a maximum of 18 km×h-1 (with 2% inclination). The test
was terminated when voluntary exhaustion occurred. VO2 max was the
highest VO2 value averaged for a 30-s period. Time to attain VO2max
(tVO2max) was also measured. VO2max was the highest value averaged
during a 30-s period. Maximal HR (HRmax) was the highest value aver-
aged during a 30-s period.
Blood lactate concentration ([La])
Capillary blood samples (25 µl) were obtained from the finger of
each subject during all tests and the [La] were measured using an
(Accutrend®Plus Roche Diagnostics GmbH, Mannheim, Germany). The
measuring range was 0.8–22 mmol∙L-1 and the sample is collected
(Accu-Chek® Softclix) and first applied to a coded yellow test strip (Ac-
cutrend BM-Roche). Blood was added to the strip by letting it drip from
a finger; in accordance with the instrument’s instructions. We never let
the finger touch the strip’s pad to exclude any possible interference due
to the sweat. The [La] were collected pre-post training: on rest (sitting for
five minutes), and 1st, 3rd and 5th-min immediately after the end of the
test (on sitting position).
Heart Rate (HR) and Rating of Perceived Exertion (RPE)
Subjects were fitted with a chest strap monitor (RS800CX Polar, Kem-
pele, Finland) for HR measurement. The HR was collected pre-and post-
training: on rest (5-min sitting; before starting the test); 60-s to 120-s im-
mediately after the end of the test (on sitting position). The assessment
of RPE was made by Category Ratio Scale (CR10) used the Borg’s table

274
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

translated into Portuguese language which consists of a vertical scale

from 0 to 10, where “0” represents no effort and “10” maximal effort. The
subject rated the effort every minute of exercise. All subjects performed
a familiarization session to this scale.

Training intervention
The current study was designed to compare the physiological effects
between the two training regimes (IT and CT) and the control group. The
subjects from IT and CT performed aerobic training running programs
that consisted of three sessions per week, 20-min per session during
a period of 14 weeks. A fourth day was used for missed session during
the same week and to avoid anyone to be excluded by not completing a
minimum of 42 (>90%) of the total 46 training sessions. Training was con-
trolled and supervised for each subject by a physical education profes-
sional trained to control the running intensities and session time. ACSM
(19) states that vigorous (77 - 95%HRmax) and near-maximal to maximal
(≥ 96%HRmax) intensities can be used for developing and maintaining
aerobic fitness. In the present study, the training intensities for both
groups were adapted from the ACSM recommendations (19). Thus, we
established the following criteria to intensities stimulus categories for
IT: 79 - 87%HRmax (vigorous); 88 - 93%HRmax (Vigorous-near maximal);
94 - 99%HRmax (near maximal to maximal). The objective of the three
zones of HR was to give to our interval running program a variety-load
with no rest between the stimuli and a progression of load during 14-wk
(Table 2). To CT we used a ~ 87%HRmax as optimal load for sustained
in a constant-effort during 20-min, as calculated during equated total
amount of work. During every session, HR was controlled with a Polar
monitor, to ensure that the subject would exercise within the pre-defined
HR zone. When a subject was behind the necessary intensity treadmill
velocity was increased by 0.5 km×h-1 and when the subject was above it
was decreased 0.5 km×h-1. Table 2 displays the training load in the two
experimental groups during the 14-week period.

275
 sp or t science

Continuous
Interval training
training
(N = 13)
(N = 11)
Vigorous Vigorous-near Near maximal Vigorous
maximal to maximal
79 - 87%HRmax 88 - 93%HRmax 94 - 99%HRmax ~ 87%HRmax
(138 ± 9 - 164 ± 10 (165 ± 10 - 172 ± 10 (174 ± 10 - 183 ± (164 ± 7 b∙min-1)
b∙min-1) b∙min-1) 11 b∙min-1)
duration (min) duration (min) duration (min) duration (min)
Week 44-min 110-min 36-min 200-min
1-2
18 x 1 min 27 x 1 min 36 x 1 min 10 x 20 min
13 x 2 min 9 x 3 min
13 x 2 min
5 x 4 min
2 x 5 min
Week 28-min 46-min 40-min 120-min
3-4 7 x 1 min 29 x 1 min 20 x 1 min 6 x 20 min
7 x 2 min 2 x 2 min 3 x 2 min
1 x 3 min 3 x 3 min 6 x 3 min
1 x 4 min 1 x 4 min
37-min 36-min 41-min 120-min
23 x 1 min 33 x 1 min 19 x 1 min 6 x 20 min
2 x 3 min 1 x 3 min 5 x 2 min
4 x 3 min

Week 25-min 33-min 56-min 120-min

5-6
7 x 1 min 5 x 1 min 3 x 2 min 6 x 20 min
5 x 2 min 10 x 2 min 6 x 3 min
1 x 3 min 2 x 4 min 8 x 4 min
1 x 5 min
24-min 63-min 27-min 120-min
10 x 1 min 22 x 1 min 27 x 1 min 6 x 20 min
7 x 2 min 9 x 2 min
5 x 3 min
2 x 4 min
Week 26-min 65-min 23-min 120-min
7-8
10 x 1 min 19 x 1 min 23 x 1 min 6 x 20 min
8 x 2 min 8 x 2 min
4 x 3 min
3 x 4 min
1 x 6 min

276
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Week 26-min 48-min 40-min 120-min

9 - 10 14 x 1 min 20 x 1 min 12 x 1 min 6 x 20 min
6 x 2 min 4 x 2 min 4 x 2 min
4 x 3 min 4 x 3 min
2 x 4 min 2 x 4 min
Week 44-min 110-min 36-min 200-min
11 - 12 18 x 1 min 27 x 1 min 36 x 1 min 10 x 20 min
13 x 2 min 9 x 3 min
13 x 2 min
5 x 4 min
2 x 5 min
Week 28-min 46-min 40-min 120-min
13 - 14 7 x 1 min 29 x 1 min 20 x 1 min 6 x 20 min
7 x 2 min 2 x 2 min 3 x 2 min
1 x 3 min 3 x 3 min 6 x 3 min
1 x 4 min 1 x 4 min
37-min 36-min 41-min 120-min
23 x 1 min 33 x 1 min 19 x 1 min 6 x 20 min
2 x 3 min 1 x 3 min 5 x 2 min
4 x 3 min

Week 1 -2 was used to equate training load (20-min/session, 5 d∙wk-1); week

3 to 14 (20-min/session, 3 d∙wk-1). All sessions for both groups (IT and CT)
started with 1-min warm-up at moderate intensity (64-76%HRmax)

Tabel 2. Distribution of work for interval training (IT) and continuous training
(CT) for 14 weeks.

Statistical analysis
Data were expressed as mean and standard deviation. Heart Rate
and Blood Lactate were compared by analysis of variance (ANOVA) with
2 factors (Group x Time) with repeated measures for Time. Absolute and
relative VO2, tVO2, Fat Mass, Body Mass and RPE were compared with
repeated measures ANOVA. Whenever ANOVA was significant, the Tukey-
Kramer post hoc was used. The criterion for significant differences was
set at p £ 0.05 All analyses were performed using the R software version
15.0 (R Foundation for Statistical Computing, Austria, 2012).

Results
All measurements before and after the 14-wk of running training for
the three groups can be seen at Table 3 There was interaction between
Group x Time. There were no differences observed at baseline between
any of the 3 groups for age, height, mass, BMI. Further, there were no

277
 sp or t science

changes for any anthropometric variable after 14-wk of training. Of the

24 participants who completed the training program (IT and CT) all at-
tended at least 90% of the training sessions. There was no difference in
adherence across groups.

IT (N = 13) CT (N = 11) CG (N = 8)
Pre Post Pre Post Pre Post
VO2
mL×kg1×min1 47.8 ± 6.3 51.7±5.9* 50.9 ± 6.0 52.8 ± 5.1 49.7 ± 5.6 49.4± 6.4
L.min 3.67 ± 0.6 4.0±0.6* 3.89 ± 0.3 4.0 ± 0.3 3.8 ± 0.5 3.7 ± 0.4

tVO2max min 9.2 ± 0.8 10.2±0.7* 10.5 ± 1.1 10.7 ± 1.0 9.6 ± 1.8 9.5 ± 1.9

HR
b∙min -1
rest 69 ± 10 59 ± 9* 67 ± 10 57 ± 2* 66 ± 8 64 ± 10
max 185 ± 11 187 ± 11 183 ± 9 185 ± 10 190 ± 6 191 ± 8
60s 122 ± 20 121 ± 10 116 ± 10 117 ± 13 123 ± 23 123 ± 15
120s 110±14** 109±11** 107±9** 108±14** 112±18** 112±14**

[La]
mmol∙L-1
rest 2.55± 0.38 2.73± 0.39 2.40± 0.20 2.29± 0.45 2.36± 0.23 2.68± 0.23
1-min 12.03± 2.89 13.14±2.57† 12.21±2.49 13.23±1.57† 13.41 1.54 13.30±1.39
3-min 11.42± 2.23 12.66±2.50† 11.97±2.9 12.88±1.65† 12.88±1.56 13.44±1.56†
5-min 9.74±3.4*** 13.00± 2.58† 11.5±2.74 12.9±1.49† 12.26±1.74 13.45±1.40†

RPE 9±1 9±1 9±1 9±1 9±1 9±1

Table 3. Changes in physiological parameters pre- to post training.

Data are presented as mean ± SD. The maximal test was carried out running
on a treadmill on ramp protocol for all groups, IT, interval training; CT, contin-
uous training; CG, control group. VO2, oxygen uptake, relative and absolute
VO2max (mL×kg-1×min1 and L×min-1); tVO2max, maximal time reached on
maximal test until voluntary exhaustion; HR, heart rate on rest (5-min before
test on sitting position), max, maximal heart rate on test, 60-s and 120-s after
ending effort (on sitting position); [La], blood lactate concentration on rest
(5-min before test on sitting position), 1-min, 3-min and 5-min after ending
effort (on sitting position); RPE, rating of perceived exertion (maximal value).
* higher compared with pre-training values; ** lower compared with 60-s in

278
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

the same group and measure; *** lower compared with 3-min in the same
group and measure. † higher compared with pre-training values in the same
group and time; p < 0.05 in all the preview.

Relative and absolute VO2max (mL×kg-1×min1 and L×min-1) in-

creased 2.47 and 3.96% only in interval training group, as seen in Table
3. The initial VO2max of all three groups was similar. Time of the maxi-
mal test performed pre- and post-training showed interaction between
the groups. The IT after 14 weeks showed a significant increase (10%; p
<0.001) when compared with CT and CG. resting heart rate showed a de-
crease for IT and CT after 14-wk of training, despite no significant inter-
action between groups. Post-training a lower mean HR was found in all
groups, when compared with pre-, at 120-s of recovery compared with
60-s of recovery (Table 3). There were no differences in blood lactate be-
tween groups. There were no changes in resting [La] after 14-wk of train-
ing, but differences were found between pre-vs. post-training in recovery
[La] measured at minutes one, three and five (Table 3).

Discussion
The major finding of the present study was that aerobic interval train-
ing with a low training duration (60 min×wk-1) at vigorous to maximal
intensities not only resulted in improved on aerobic power (VO2peak)
but also increased maximal time achieved on the VO2 test until volun-
tary exhaustion.

Blood lactate concentration

The present study found higher post-maximal test [La] in both experi-
mental groups at various measurement times (1st,3rd,5th-min), after
14-wk of training. We cannot conclude that these values increased due
to longer maximal effort, because only the IT group showed increases
on tVO2max (11.08%) and less due to training mode because there was
no interaction between training groups for this variable All groups in-
creased metabolic glycolytic potential over 14 weeks of training. Man-
droukas et al. (17) investigated physically active youth, for the same
amount of running (32-min) in three different modes (active recovery IT;
passive recovery IT and continuous training) and concluded that lactate
is removed faster when recovery is active. For the authors, possible ex-
planation is that blood lactate is used as energy substrate in aerobic en-
ergy production by inactive muscles and that the primary factor for the
increase in [La] after exercise is the duration and intensity (high) effort

279
 sp or t science

in which the individual has undertaken. Others (20) pointed out that the
high [La] measured immediately after exercise remain elevated between
3 to 8 min, and up to approximately the 60-min [La] be likely to decrease
progressively. For our study, we hypothesized that training mode (con-
tinuous or interval) would not affect the [La] after maximal efforts. Both
groups showed decrease of [La] between 1st and 5th-min. We conclude
that a 3-min recovery was not enough to induce significant changes in
[La]. In the present study, after maximal efforts, at least five minutes of
passive recovery were required to level-off [La].

Heart Rate
Even though all subjects were fit before the experiment, decreases
on resting heart rate (8.86%) were found, may be because of the exer-
cise mode (running). The literature (21) suggests that this reduction is
due to increased venous return. Typically, individuals with a low resting
HR have increased parasympathetic activity and decreased sympathet-
ic activity. In aerobically trained individuals this response is expected.
The resting bradycardia is one of the adaptations after chronic exercise.
There were no changes after 14-wk in HRmax, HR60s and HR120s (recov-
ery). However, as expected, all groups showed HR decrease immediately
after maximal effort from 60-s to 120-s. Some (22) described training
adaptations in recovery HR, but in the present study these adaptations
were only observed in resting HR, corroborating previous findings with
running (3). In relation to rating of perceived exertion no change was ob-
served despite the interval training group had increased the total time at
maximal effort after 14-wk of training.

Maximal VO2
The most important finding of our results was that only individuals
who performed interval training showed improved values of maximal
VO2 after 46 sessions of running training. The value of VO2 max is an im-
portant predictor of performance and has been used in the prescription
of exercise. The duration and intensity of the stimulus directly affects the
magnitude of the adaptation to aerobic training and these benefits are
verified mainly with interval work durations from three to five minutes
(24). The adaptations observed in this study for interval training group
were higher compared with that in the continuous training group and
may be due to central and peripheral adaptations that may result in
higher performance (14). Some stated that IT when performed at inten-
sities below 80%VO2max offers no improvement over CT at the same

280
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

intensity (23); while others (25) reported that the best gains comes from
IT with a (2:2) exercise/recovery ratio. In agreement with our results, sev-
eral (14, 18 ,28) have verified that IT at 80-90% provides significant im-
provements relative to the CT 50-75 VO2max in healthy subjects. Recent-
ly (29) such benefits were observed at 12-wk of IT with low volume and
high intensity where a group ran (20-min; 2 d.wk-1) at 1 (89%VO2max): 1
(10%VO2max) and other a CT at high volume (30 - 50 min; 2 d.wk-1) and
moderate intensity (58%VO2max).
Some authors have not found differences between training (IT and
CT) for 10 weeks of exercise (26). The limited duration of the continuous
running training herein can also help to explain the minor adaptations,
as others (18) recorded improvements in VO2peak after a higher training
volume than that herein (> 20-min). Training interval when performed in
a maximum intensity or higher and relatively short recovery time (2:1)
apparently shows improvements for the development of aerobic and
anaerobic systems, but if the recovery time is very short, stagnation oc-
curs in early exercise. Some authors (30) pointed out the need to equate
the interval and continuous training by distance performed or other
equivalent variable and that when the intensity is below 80%VO2max)
exercise is maintained without major difficulties and can thus be per-
formed in continuous mode. Although, as noted in the literature, it is
necessary to point out that for high intensity exercise (90-120%VO2max)
is necessary to prescribe training periods or intervals that enable a suffi-
cient volume. Moreover, Gibala (31) recommends that when a minimum
volume is performed a high stimuli intensity should be done to obtain
gains. An important aspect is that our training at “low-volume” showed
benefits without the need to stress individuals on popular high-intensity
training as HIT.

Total time on maximal test (tVO2max)

IT showed a 11.08% increased (9.2 ± 0.8 min vs. 10.2 ± 0.7 min) 14-wk
of running. Authors (32) described IT and TC improvements in exercise
time (4.0%) in tVO2max after 7 weeks of training. Although the authors
have conducted stimuli shorter than that in the present study, the reduc-
tion in training volume (~ 50%) showed benefits on active subjects. Oth-
ers (37) reported that combining low volume of aerobic sessions in con-
tinuous mode (75-85%HRmax) with high-intensity interval training and
sprint training improved by 1-min the time to run 10km, after 6 weeks of
training. Moore et al. (32) found during a 10-wk of running training, im-
provement in the maximum time until volitional exhaustion (16.4 ± 3.2

281
 sp or t science

vs. 17.3 ± 2.8 min, p<0.05), despite not having improvements for maximal
VO2. For the author (31) improvement in performance (time) was due to
peripheral adaptations. In our study, we did not find performance im-
provements in the training group that exercised at constant loads (CT).

Conclusions
Interval training in the current study presented a significant improve-
ment in maximal oxygen uptake and in time on maximal effort. However,
14-wk of continuous running had no impact on aerobic fitness indicators
that were assessed. Hence, we can conclude that interval training at a
minimum of 20-min performed three times per week is sufficient for fit-
ness improvement and more appropriate when compared with continu-
ous training of similar volume.

References
1. Chomistek AK, Cook NR, Flint AJ, Rimm EB. Vigorous-intensity leisure-time
physical activity and risk of major chronic disease in men. Med Sci Sports
Exerc. 2012 Oct;44(10):1898.
2. Støren Ø, Bratland-Sanda S, Haave M, Helgerud J. Improved V [Combining
Dot Above] O2max and Time Trial Performance With More High Aerobic In-
tensity Interval Training and Reduced Training Volume: A Case Study on an
Elite National Cyclist. J Strength Cond Res. 2012 Oct 1;26(10):2705-11.
3. Nybo L, Sundstrup E, Jakobsen MD, Mohr M, Hornstrup T, Simonsen L, Bül-
ow J, Randers MB, Nielsen JJ, Aagaard P, Krustrup P. High-intensity train-
ing versus traditional exercise interventions for promoting health. Med Sci
Sports Exerc. 2010 Oct 1;42(10):1951-8.
4. Kraus WE, Houmard JA, Duscha BD, Knetzger KJ, Wharton MB, McCartney
JS, Bales CW, Henes S, Samsa GP, Otvos JD, Kulkarni KR. Effects of the
amount and intensity of exercise on plasma lipoproteins. N Engl J Med.
2002 Nov 7;347(19):1483-92.
5. Dupont G, Berthoin S. Time Spent at a High Percentage of max for Short
Intermittent Runs: Active Versus Passive Recovery. Can J Sport Sci. 2004
Oct;29(S1):S3-16.
6. Glaister M, Stone MH, Stewart AM, Hughes MG, Moir GL. Aerobic and an-
aerobic correlates of multiple sprint cycling performance. J Strength Cond
Res. 2006 Nov 1;20(4):792-8.
7. Wakefield BR, Glaister M. Influence of work-interval intensity and dura-
tion on time spent at a high percentage of VO2max during intermittent su-
pramaximal exercise. J Strength Cond Res. 2009 Dec 1;23(9):2548-54.
8. McKay BR, Paterson DH, Kowalchuk JM. Effect of short-term high-inten-
sity interval training vs. continuous training on O2 uptake kinetics, mus-
cle deoxygenation, and exercise performance. J Appl Physiol. 2009 Jul
1;107(1):128-38.
9. Bradley PS, Mohr M, Bendiksen M, Randers MB, Flindt M, Barnes C, Hood
P, Gomez A, Andersen JL, Di Mascio M, Bangsbo J. Sub-maximal and maxi-

282
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

mal Yo–Yo intermittent endurance test level 2: heart rate response, re-
producibility and application to elite soccer. Eur J Appl Physiol. 2011 Jun
1;111(6):969-78.
10. Standage M, Sebire SJ, Loney T. Does exercise motivation predict en-
gagement in objectively assessed bouts of moderate-intensity exer-
cise? A self-determination theory perspective. J Sport Exercise Psy. 2008
Aug;30(4):337-52.
11. Boutcher SH. High-intensity intermittent exercise and fat loss. J Obes. 2010
Nov 24;2011.
12. Berg K. Endurance training and performance in runners. Sports Med. 2003
Jan 1;33(1):59-73.
13. American College of Sports Medicine. ACSM’s guidelines for exercise test-
ing and prescription. Lippincott Williams & Wilkins; 2013 Mar 4.
14. Helgerud J, Hoydal K, Wang E, Karlsen T, Berg P, Bjerkaas M, Simonsen T,
Helgesen C, Hjorth N, Bach R, Hoff J. Aerobic High-Intensity Intervals Im-
prove VO2max More Than Moderate Training. Med Sci Sports Exerc. 2007
Apr 1;39(4):665.
15. Swain DP. Moderate or vigorous intensity exercise: which is better for im-
proving aerobic fitness? Prev Cardiol. 2005 Jan 1;8(1):55-8.
16. Glaister M. Multiple sprint work. Sports Med. 2005 Sep 1;35(9):757-77.
17. Mandroukas A, Heller J, Metaxas TI, Sendelides T, Riganas C, Vamvakoudis
E, Christoulas K, Stefanidis P, Karagiannis V, Kyparos A, Mandroukas K.
Cardiorespiratory and metabolic alterations during exercise and passive
recovery after three modes of exercise. J Strength Cond Res. 2011 Jun
1;25(6):1664-72.
18. Rognmo Ø, Hetland E, Helgerud J, Hoff J, Slørdahl SA. High intensity aero-
bic interval exercise is superior to moderate intensity exercise for increas-
ing aerobic capacity in patients with coronary artery disease. Eur J Cardiol
Prev Rehab. 2004 Jun 1;11(3):216-22.
19. Garber CE, Blissmer B, Deschenes MR, Franklin BA, Lamonte MJ, Lee IM,
Nieman DC, Swain DP. American College of Sports Medicine position stand.
Quantity and quality of exercise for developing and maintaining cardiores-
piratory, musculoskeletal, and neuromotor fitness in apparently healthy
adults: guidance for prescribing exercise. Med Sci Sports Exerc. 2011
Jul;43(7):1334-59.
20. Goodwin ML, Harris JE, Hernández A, Gladden LB. Blood lactate measure-
ments and analysis during exercise: a guide for clinicians. J Diabetes Sci
Tech. 2007 Jul;1(4):558-69.
21. Moholdt T, Aamot IL, Granøien I, Gjerde L, Myklebust G, Walderhaug L,
Brattbakk L, Hole T, Graven T, Stølen TO, Amundsen BH. Aerobic interval
training increases peak oxygen uptake more than usual care exercise train-
ing in myocardial infarction patients: a randomized controlled study. Clin
Reab. 2012 Jan;26(1):33-44.
22. Thompson PD, Crouse SF, Goodpaster BR, Kelley D, Moyna NI, Pescatello LI.
The acute versus the chronic response to exercise. Med Sci Sports Exerc.
2001 Jun;33(6 Suppl): S438-45.
23. Overend TJ, Paterson DH, Cunningham DA. The effect of interval and con-
tinuous training on the aerobic parameters. Can J Sport Sci. 1992 Jun

283
 sp or t science

1;17(2):129-34.
24. MacDougall D, Sale D. Continuous vs. interval training: a review for the ath-
lete and the coach. Canadian journal of applied sport sciences. J Can Sci
App Sport. 1981 Jun;6(2):93-7.
25. Gibala MJ, Little JP, MacDonald MJ, Hawley JA. Physiological adaptations
to low volume, high intensity interval training in health and disease. J
Psychol. 2012 Mar 1;590(5):1077-84.
26. Swain DP, Franklin BA. Comparison of cardioprotective benefits of vigor-
ous versus moderate intensity aerobic exercise. Am J Cardiol. 2006 Jan
1;97(1):141-7.
27. Thomas TR, Adeniran SB, Etheridge GL. Effects of different running pro-
grams on VO2 max, percent fat, and plasma lipids. Canadian journal of ap-
plied sport sciences. J Can Sci App Sport. 1984 Jun;9(2):55-62.
28. Currie KD, Dubberley JB, McKELVIE RS, MacDonald MJ. Low-volume, high-
intensity interval training in patients with CAD. Med Sci Sports Exerc. 2013
Aug 1;45(8):1436-42.
29. Tuimil JL, Boullosa DA, Fernández-del-Olmo MÁ, RodrÍguez FA. Effect of
equated continuous and interval running programs on endurance perfor-
mance and jump capacity. J Strength Cond Res. 2011 Aug 1;25(8):2205-11.
30. Gibala MJ. High-intensity interval training: a time-efficient strategy for
health promotion? Curr Sport Med Rep. 2007 Jul;6(4):211-3.
31. Gunnarsson TP, Bangsbo J. The 10-20-30 training concept improves per-
formance and health profile in moderately trained runners. J Appl Physiol.
2012 Jul 1;113(1):16-24.
32. Moore IS, Jones AM, Dixon SJ. Mechanisms for improved running economy
in beginner runners. Med Sci Sports Exerc. 2012 Sep 1;44(9):1756-63.

284
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Comparison of energy expenditure during

self-selected walking in field tracks and in
treadmill at various grades
João Paulo Brito1, Ana Conceição1,2, Félix Romero1, Adenilson
Targino de Araújo Junior3, Victor Machado Reis2,3

Introduction
Despite the growing popularity of performing regular walks and the
medical recommendations for their achievement, little is known about
the level of effort induced by the practice of field tracks walks (1). Most
previous studies (2, 3) examined the practice of walking in laboratory
conditions. Energy Expenditure (E) has a critical role in locomotion (4),
as body adjusts/controls the E according to the energy needed for the
task (5).
Using data from the kinematics, Hoffman and colleagues (2) con-
firmed that the conditions of instability of field tracks increase the E
when compared with walks under controlled conditions. Larger steps
caused by rough ground may involve a higher E because of the cost of
transition between feet (6). However, for Menz and colleagues (7) walk-
ing in areas with small obstacles (up to 25 mm) does not seem to be a
significant factor in changing patterns of motion. In a comparative study,
Hall and colleagues (8) say that even when walking on land is held on flat
surfaces, we need more energy for the field tracks than in the motorized
treadmill. According to these authors, it is not possible to estimate the E
of field tracks motion based on laboratory values.
Pearce and colleagues (9) stated that the E of motion in a rigid surface
is higher than that in a motorized treadmill for a pre-determined speed.
On the other hand, walking on a slope can change significantly the E (10,
11). Thus, when prescribing field tracks walks as a physical exercise it is
necessary to know the E involved while walking on sloping terrain and
its reliability in a motorized treadmill. The physiological responses are

1. Sports Sciences Research Laboratory, Sports Sciences School of Rio Maior,

Polytechnic Institute of Santarém, Portugal
2. Research Center in Sport Sciences, Health Sciences & Human Development
(CIDESD), Portugal
3. Department of Sport Sciences, Exercise & Health, University of Trás-os-
Montes and Alto Douro, Portugal

285
 sp or t science

sensitive to the increase in slope (10) and increasing the difficulty of the
terrain may cause a significant increase in E (12).
When walking in field tracks, it may occur various changes in the rela-
tive contribution of the E and in the optimization of the motion stability
(13). If the E constraint contribution was critical, it would be expectable
to present adaptation responses to reduce energy costs. Nevertheless,
the adaptation processes during locomotion tend to optimize the rela-
tion between efficiency of movement and E (14). The use of a self-select-
ed speed seems to result, for most subjects, in E lower than when it is
imposed (10).
DeVoe and colleagues (12) reported that despite the various con-
strains imposed by the ground irregularities, the self-selected speed is
the most economical for most subjects allowing the execution of move-
ment for long periods of time. When walking takes place in varies soils,
there may be variations in the speed selected by the subjects to offset
the soil difficulties (15). Recent research reports scarce conclusive results
in what concerns field versus laboratory walking; Therefore, the present
study aims to compare the energy expenditure. Oxygen consumption,
heart rate and speed necessary to walk in field tracks and on motorized
treadmill at different slopes.

Material and Methods

Participants
Twenty-one male subjects (age 23.43 ± 2.43 years; body weight 77.46
± 13.43 kg, height 1.75 ± 0.08 m; fat mass estimated 15.09 ± 5.42%; con-
sumption of oxygen at rest 3.17 ± 0.37 ml.kg-1.min-1), physically active
(exercising at least for three times per week over the past 6-month pe-
riod) and who had no history of orthopaedic injury, volunteered for this
study. All procedures followed the Helsinki Declaration of 1975 and were
approved by the ethics committee of the institution were the study took
part.

Experimental protocol
On their first visit to the laboratory, subjects’ body mass and height
were measured with a scale and stadiometer (SECA, Germany, Hamburg)
and fat mass was measured estimated using Bio Impedance Analysis
method (Omron BF300, Matsusaka Co. LTD., Japan). The subjects did not
exercise for 12-h prior to the exercise sessions and did not ingested caf-
feine within 3-h before exercise. These measurements were performed
under a minimum 8-h fasting.

286
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

After this evaluation, the subjects had a light meal and 3-h later they
started walking in field tracks. Wenty-four hours later the subjects com-
pleted three periods of walking in the countryside tracks with an interval
of 30 min (a starting VO2 with a difference less than 2.1 ml.kg-1.min-1 to
that observed before the start of the first period of effort was required).
The three segments of the course in field tracks had a mean slope and
distance of 0%, 6% and 14% and 821, 618 and 598 metres, respectively.
On each track, maximum slope variation was 1%.
We used forest segments of footpaths and roads with hard soil
ground predominantly of gravel, without any obstacles. The route seg-
ments were evaluated via planimetric and altimetric survey (Sokkia
130R, Casagiove, Italy) and only the routes that respected a variability of
slope equal to or inferior to 1% over the entire distance of the segment
were selected. The tests took place during the morning with tempera-
tures between 20 and 22ºC and humidity between 50 and 60%.
Before and during the sessions in field tracks, the wind speed was
measured with a portable anemometer instrument brand Xplorer, mod-
el SkywatchXplorer 3 (JDC Electronic, Yverddon-les-Bains, Switzerland).
The tests took place whenever the wind speed was inferior to 15 km.h-1,
speed up to which the aerodynamic resistance is considered negligible,
according to Di Prampero (16). When the wind speed exceeded 15 km.h-1,
the test was stopped and repeated 24-h later. The walking speed on the
field tracks was self-selected as suggestion by literature (17, 18).
Prior to the test, subjects were asked to select a comfortable walking
speed that would allow them to perform up to a 15-km track. Through-
out the route, subjects were followed by an investigator riding a bicycle
with a digital velocimeter (MSC-2DXC, AT EYE Co., LTD, Osaka, Japan)
and were asked to return to the walking speed they self-selected during
the first two minutes of the trial whenever they deviated more than 0.2
km.h-1 from the self-selected speed. Forty-eight hours later, all field tests
were repeated on a motorized treadmill RunRace (Tecnogym, Gambot-
tela, Italy), calibrated according to the manufacturer´s guidelines. The
subjects exercised in the treadmill with the same distance, slope and
walking speeds that were performed during the three segments of the
field tracks
During all exercise sessions, VO2 was measured using direct oxime-
try with a gas analyser Cosmed K4b2 (Cosmed, Rome, Italy) and heart
rate (HR) was measured using a POLAR chest belt coupled to the K4b2.
The E was calculated by the software K4b2 based on the quantities of
the expired VCO2 and VO2 (19). Data were processed using the software

287
 sp or t science

Cosmed PFT Suite version 7.4b (Cosmed, Rome, Italy). During each seg-
ment of the route, the values of the HR and VO2 of the subjects were
monitored by telemetry. The VO2 values were recorded every 20-s (20)
and then averaged per minute to subsequent calculations. To ensure the
existence of a plateau in VO2, the maximum difference allowed between
three 20-s consecutive average VO2 values was less than 2.1 ml.kg-1.min-1.
Heart rate was measured concomitantly to VO2, E was calculated by the
manufacturer’s software. Tests were carried out to the calibration of the
analyser, indicated by the manufacturer, before each use: reference gas
calibration with O2 = 16% and CO2 =5%, turbine calibration with a 3 litres
syringe) delay calibration and room air calibration.

Statistics
The data were analysed using SPSS software version 17.0 (Science,
Chicago, USA). A repeated measures ANOVA was used to confront the
average values of the final min of each variable in the different condi-
tions of exercise. To verify the ‘sphericity’, it was used the ‘sphericity test
of Mauchly.” Whenever the assumption of “sphericity” was not the ob-
served, we used the correction factor of Greenhouse-Epsilon Geisse, as
it is the most conservative and appropriate for a small size sample (21).
For all statistical procedures, the minimum level of significance admit-
ted was P ≤ 0.05.

Results
After analysing the curves of the minute values of VO2, it was con-
firmed that the stabilization of VO2 occurred, basic condition for the use
of these numbers as an indicator of E. Figure 1 shows the typical behav-
iour of VO2 in the six journeys made (values of a representative subject).

288
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Figure 1. Means (full lines and dotted lines) and standard deviations
(error bars) of VO2 while walking in motorized treadmill (TW) and in na-
ture (FW) with 0% gradient (upper panel), 6% gradient (middle panel)
and 14% gradient (lower panel).

The values of the average per minutes of the VO2, E and HR of walking
on a motorized treadmill (TW) were higher than those of the walking in
field track (FW) on all slopes. In all the average comparisons of the stud-
ied variables, there were no differences in the following comparisons: FW
slope with 6% vs TW with slope of 6% to the variable FC; FW with slope of
0% vs FW with slope of 6%, for variable speed (see Table 1). A change of
6% in the slope did not cause differences in the walking speed. However,
when the comparisons slope 0 vs 14% slope and slope 6 vs 14% slope

289
 sp or t science

were made, there is a lower rate of displacement, for a significance of, P

≤ 0.05 and 0.01, respectively.

Exercise VO2 HR E S
condition
(ml.kg-1. (b.min-1) (kcal. (m.s-1)
min-1) min-1)
FW slope vs TW slope 0% dif. 6.44** 12.59** 2.29** -
0%
IC 95% 4.88 to 3.44 a 1.56 to -
8.01 21.75 3.01
FW slope vs TW slope 6% dif. 5.66** 3.17 1.10* -
6%
IC 95% 1.56 to -5.31 a 0.14 to -
9.75 11.65 2.06
FW slope vs TW slope dif. 12.25** 15.01** 2.95** -
14% 14%
IC 95% 9.28 to 6.43 a 1.92 to -
15.23 23.6 3.99
FW slope vs FW slope 6% dif. 11.24** 25.58** 4.01** 0.05
0%
IC 95% 7.54 to 19.35 a 3.14 to -0.02 to
14.94 31.8 4.89 0.12
FW slope vs FW slope dif. 16.86** 37.89** 6.31** 0.15*
0% 14%
IC 95% 14.16 to 31.05 a 5.39 to 0.05 to
19.57 44.73 7.23 0.24
FW Slope vs FW slope dif. 5.63** 12.31** 2.30** 0.10**
6% 14%
IC 95% 2.78 to 8a 1.63 to 0.05 to
8.47 16.63 2.97 0.15
TW slope vs TW slope 6% dif. 10.45** 16.15** 2.83** -
0%
IC 95% 8.07 to 10.35 a 2.03 to -
12.82 21.95 3.62
TW slope vs TW slope dif. 22.67** 40.31** 6.98** -
0% 14%
IC 95% 19.39 to 31.46 a 5.68 to -
25.96 49.16 8.28
TW slope vs TW slope dif. 12.22** 24.16** 4.15** -
6% 14%
IC 95% 10.06 to 20.33 a 3.43 to -
14.38 27.98 4.88

290
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

VO2 = oxygen uptake; HR = heart rate; E = energy expenditure; S = speed; *,

**Statistically significant differences in average for P≤0.05 and 0.001, respec-
tively.
Table 1. Average difference (dif. ) and confidence interval (95% CI) of differences
of the comparisons of different exercise conditions and slope - walking on a
motorized treadmill (TW) vs in field tracks (FW) in the slopes 0% (n = 21), 6%
(n = 21) and 14% (n = 19) for variable speed (S), oxygen consumption (VO2),
heart rate (HR) and the energy (E).

Discussion
Results show a progressive increase in VO2, E and HR with slope, ei-
ther in FW or in the TW conditions of exercise. The E and the VO2 were
higher in TW compared with FW condition. Average speed was also high-
er in TW (1.64 ± 0.16 ms-1 at slope 0%, 1.59 ± 0.14 ms-1, at slope 6%; 1.49
± 0.13 ms-1 and at slope 14%) when compared with self-selected speed.
These corroborate the results of Bertram and Ruina (22), Falola and col-
leagues (23) and that by DeVoe and colleagues (12), which state that the
self-selected speed by the subjects is not always the most economical.
Although Bobbert (11) and Margaria (24) claim that the optimal walking
speed in slopes above 10% is 1.0 ms-1, in the present study, at slopes of
6% and 14%, we observed that the self-selected velocities were higher
than those by Bobbert (11) and by Margaria (24).
The values of E, VO2 and HR are lower in the field tracks walking for
self-selected velocities and are similar to those of Church and colleagues
(25), for the slope of 0% in land, 5.7 ± 1.3 kcal.min-1, 12.8 ± 1.8 ml.kg-1.
min-1 and 101.6 ± 12.0 beat.min-1, respectively. These results differ from
those found by Pearce and colleagues (9) where significant differences
were confirmed in VO2 of walking in a motorized treadmill vs on a hard
surface (respectively, 10.58 and 11.04 ml.kg-1.min-1). However, these re-
sults were found with subjects aged from 55 to 66 years while we exam-
ined the response of young adult subjects (aged 23.6 ± 2.43). Despite not
having measured the E, Murray and colleagues (26) found no differences
between the cinematic walking in the motorized treadmill and in rigid
surface, but the electromyography activity and HR were higher in the
motorized treadmill.
Some (27, 28) have previously reported that there is an intra-subject
variability in the E during the walking on the motorized treadmill. On the
contrary, Bertram and Ruina (22) claim that walking on motorized tread-
mill mimics the natural constraints of the walking, presenting itself as an
alternative of replication to the data extrapolating from the laboratory

291
 sp or t science

environment to the natural environment. There is no consensus that the

oxygen consumption values are higher or lower in the motorized tread-
mill vs field tracks (26, 29, 30, 31).
The increase in the E, with the increment of the slope was previously
described by Alexander (10). This study also found that the E increased
progressively with the increases of the slope. This phenomenon was ob-
served in both walking on a motorized treadmill and in field tracks. Nev-
ertheless, in walking on a motorized treadmill there was a greater inter-
subject variability compared with walking in field tracks. Johnson and
colleagues (32) found VO2 values of 10.79, 15.05 and 20.59 ml.kg-1.min-1, in
walking on motorized treadmill with constant speed of 1.11 ms-1, at 0%,
5% and 15%, slope respectively. Our values for the progressive slopes
were 21.91, 32.36 and 44.58 ml.kg-1.min-1, with self-selected velocities of
1.64 ± 0.16, 1.59 ± 0.14 and 1.49 ± 0.13 ms-1, respectively at grade, 6% and
14%.
The higher walking velocities used by the subjects in the current
study (1.64 ± 0.16, 1.59 ± 0.14 and 1.49 ± 0.13 ms-1, respectively), com-
pared with those in the study by Johnson and colleagues (32), may ex-
plain the higher mean values of VO2. Our VO2 mean values are also higher
than those described by Gottschall and Kram (3), which report average
values of 7.4 ± 0.7 ml.kg-1.min-1 walking in the motorized treadmill for
a speed of 1.25 m.s-1. Thus, once more, these results may suggest that
the walking speed seems to be the main factor in the VO2 differences
between the different studies. However, the sample used in that study
comprised young adults of both sexes (average age 27.4 ± 5.1 years), con-
trary to the current study.
The previous study with most similar methods compared with the
current study, was that by Johnson and colleagues (2002). Percentage
increases in VO2 due to increases in slope in both studies revealed dif-
ferent magnitudes. It was found that when the slope increased from 0
to 5 % (32), or 6 % (this study), the increases in VO2 were 39.5 and 47.7
%, respectively. A possible explanation for the largest increase observed
in our study could be the difference in slopes, though small. However,
when comparing slope increases between 0% and the highest slope (14
% in this study and 15 % in the study by Johnson et al. (32), the delta re-
mained higher in our study (103.5 vs 90.8%). Therefore, other differences
such as subjects’ physical fitness may help to confront both studies.

Conclusions
This study extends the previous literature about the E during walking

292
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

in field tracks, by including walking in different slopes and comparing it

with walking at similar pace on a motorized treadmill. The results show
that 1) VO2, E and HR increase gradually with the slope of the route, both
in the field and in the treadmill, 2) E and VO2 are different in the two exer-
cise conditions, being higher during treadmill walking at a similar pace.
Therefore, our results suggest that the estimation of E in field tracks
based on laboratory measurements can overestimate the true E.

References
1. Ainslie P, Campbell I, Frayn K, Humphreys S, Maclaren D, Reilly T. Physiologi-
cal and metabolic responses to a hill walk. J Appl Physiol. 2002; 92:179-187.
2. Hoffman MD, Millet GY, Candau RB, Rouillon JD. Evaluation of a theoretical
model to quantify the sources of metabolic cost in walking. Am J Phys Med
Rehab. 2004; 83(5), May. 353-362.
3. Gottschall JS & Kram R. Energy cost and muscular activity required for pro-
pulsion during walking.2003 J Appl Physiol. 1985; 2003 May;94(5):1766-72.
4. Lovejoy CO. Evolution of human walking. Sci Am. 1988; 259, 118-125.
5. Sparrow WA, Newell KM. Energy expenditure and motor performance rela-
tionships in human learning a motor task. Psychophysiology. 1994; 31, 338-
346.
6. Donelan JM, Shipman DW, Kram R, Kuo AD. Mechanical and metabolic re-
quirements for active lateral stabilization in human walking. J Biomech.
2004; 37(6). 827-835.
7. Menz HB, Lord SR, Fitzpatrick RC. Acceleration patterns of the head and pel-
vis when walking on level and irregular surfaces. Gait Posture. 2003; 18, 35-
46.
8. Hall C, Figueroa A, Fernhall B, Kanaley JA. Energy expenditure of walking
and running: comparison with prediction equations. Med Sci Sport Exer.
2004; 36(12), 2128-2134.
9. Pearce ME, Cunningham DA, Donner AP, Rechnitzer PA, Fullerton GM, How-
ard JH. Energy cost of treadmill and floor walking at self-selected paces. Eur
J Appl Physiol. 1983; 52(1), 115-119.
10. Alexander RM. Energetics and optimization of human walking and running:
the 2000 Raymond Pearl Memorial lecture. Am J Hum Biol. 2002; 14, 641-
648.
11. Bobbert AC. Energy expenditure in level and grade walking. J Appl Physiol.
1960; 15: 1015-1021
12. DeVoe D, Miller L, Gotshall R. Energy expenditure of backpacking for women
on even and uneven terrain. Res Q Exercise Sport. 1998; 69(1), A32-33.
13. Caldwell GE, Van Emmerik REA, Hamill J. Movement proficiency: task de-
mands and constraints. In: Sparrow, W. A. (Eds.). Energetics of human activ-
ity. Champaign, IL: Human Kinetics; 2000.
14. Patla AE, Sparrow WA. Factors that have shaped human locomotor struc-
ture and behavior. In Sparrow, W. A. (Eds.). Energetics of human activity.
Champaign, IL: Human Kinetics; 2000.
15. DeVoe D. Caloric Expenditure During Backpacking. J Hum Movement Stud.

293
 sp or t science

2001; 40, 259-268.

16. Di Prampero PE. The energy cost of human locomotion on land and in water.
Int J Sports Med. 1986; 7, 55-72.
17. Murtagh EM, Boreham CA, Murphy MH. Speed and exercise intensity of rec-
reational walkers. Prevention in Medicine. 2002; 35(4), 397-400.
18. Spelman C, Pate R, Macera C, Ward D. Self-selected exercise intensity of ha-
bitual walkers. Med Sci Sport Exer. 1993; 25(10), 1174-1179.
19. Elia M, Livesey G. Energy expenditure and fuel selection in biological sys-
tems: the theory and practice of calculations based on indirect calorimetry
and tracer methods. World Revue of Nutrition Dietary. 1992; 70, 68-131.
20. Aisbett B, Le Rossignol P. Estimating the total energy demand for supra-
maximal exercise using the V̇ O2-power regression from an incremental ex-
ercise test. J Sci Med Sports. 2003; 6:343-347.
21. Box G EP. Some theorems on quadratic form applied to the study analysis
of variance problems: II Effects inequality of variance and the correlation
between errors in the TRo-way classification. Annals of Mathematical Sta-
tistics. 1954; 25, 484-498.
22. Bertram JEA, Ruina A J. Multiple walking speed-frequency relations are pre-
dicted by constrained optimization. A.J.of Theor. Biology. 2001; 209,445-
453.
23. Falola JM, Delpech N, Brisswalter J. Optimization characteristics of walking
with and without a load on the trunk of the body. Percept Motor Skill. 2000;
91, 261-272.
24. Margaria R. Sulla fisiologia e especialmente sul consumo energetico della
marcia e della corsa a varia velocità ed inclinazione el terreno. Atti Acca-
demia Nazionale dei Lincei. 1938; 7, 299-368.
25. Church TS, Earnest CP, Morss GM. Field tracks-testing of physiological re-
sponses associated with Nordic Walking. Res Q Exercise Sport. 2002; 73(3),
296-300.
26. Murray M, Spurr G, Sepic S, Gardner G, Mollinger L. Treadmill vs floor walk-
ing: kinematics, electromyogram and heart rate. J Appl Physiol. 1985; 59,
87-91.
27. Morgan DW, Martin PE, Krahenbuhl GS, Baldini FD. Variability in running
economy and mechanics among trained male runners. Med Sci Sport Exer.
1991; 23(3), 378-383.
28. Morgan DW, Martin PE, Krahenbuhl GS. Factors affecting running economy.
Sports Med. 1989; 7(5), 310-330.
29. Frishberg BA. An analysis of over ground and treadmill sprinting. Med Sci
Sport Exer. 1983; 15, 478-485.
30. Elliott BC, Blanksby BA. A cinematographic analysis of overground and
treadmill running by males and females. Med Sci Sport Exer. 1976; 8, 84-87.
31. Nelson RC, DilIman CJ, Lagasse P, Bickett P. Biomechanics of overground
versus treadmill running. Med Sci Sport Exer. 1972; 4, 233-240.
32. Johnson A, Benjamin M, Silverman N. Oxygen consumption, heat produc-
tion, and muscular efficiency during uphill and downhill walking. Appl Ergo.
2002; 33, 485-491.

294
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Optimization as adaptability: how movement

variability supports performance in striking
actions used in combat sports
Dominic Orth1, John van der Kamp1, Robert Rein2

The functional role of movement variability: evidence in coordi-

nation and control
In this review, striking tasks derived from combat sports (including
karate, Taekwondo, kung fu, boxing and mixed martial arts [MMA]) are
evaluated to address the role of goal supportive or functional movement
variability. Movement variability can be considered functional in so far
that it ensures action goals are met by the individual, such as maintain-
ing performance as constraints are changed (1).
Constraints specify the specific context of the action and can be fur-
ther differentiated into individual, environmental and task constraints
(1, 2). Constraints are understood as placing boundaries on movement
coordination and control solutions (as opposed to proscribing them,
(3)). Coordination solutions refer to distinctive patterns or classes of
movement such as walking or running. The term control, a level of anal-
ysis embedded within the coordination of action, refers to movements
made that regulate the stability of a given coordinative solution (4). Con-
straints define the space which the movement system can act (i.e., the
perceptual-motor workspace) (5, 6), and as such determine what coordi-
nation solutions are available to maintain performance.
Previously, movement variability has (mistakenly) been conceptual-
ized entirely as noise that needs to be reduced (7). However, more recent
research has shown associations between movement variability and the
maintenance of performance as constraints are changed (8). To exem-
plify, task constraints, might include goals or rules, such as strike a box-
ing bag with the hand as forcefully and efficiently as possible (9). The
environmental constraints, on the other hand, include properties of the
world external to the individual (10), and, might be the properties of the
target and the distance to the target prior to executing a technique (11)

1. Faculty of Behavioral and Movement Sciences, Vrije Universiteit, Amsterdam,

The Netherlands
2. Institute of Training and Computer Science in Sports German Sport Univer-
sity Cologne, Cologne, Germany.

295
 sp or t science

(cf. (10)). As a general pedagogical strategy, by manipulating constraints,

the individual can (learn to) exploit functional movement variability to
maintain or improve performance when facing similar arrangements of
constraints in the future (1, 3).
In striking tasks, there seems to be two levels at which functional
movement variability is important (see Figure 1). On the one hand, in-
dividuals can use movement variability in terms of how they control a
given action or technique. For example, when a kicking action is used,
if a target is far away, the individual can increase the range of motion at
the hips in order to kick it (12). On the other hand, individuals can use
behavioural variability, by changing which coordination solution they re-
alize to strike a target. For example, as the distance to a target is reduced,
individuals might use a hook punch instead of a straight punch because
the ability to generate sufficient force is afforded with the greater trajec-
tory length permitted with a hook punch (9).
Within this context, affordances have been considered an additional
source of constraint (13). Affordances refer to the behavioural opportuni-
ties that become available from the way the individual and their environ-
ment relate to each other during performance (14). The perception and
realization of these affordances are reflected in the coordination solu-
tion used (e.g. crawling, walking, climbing (15)) and the way movement
adjustments are made to satisfy constraints (15). For instance, in the case
where individuals continue to realize an affordance under changing con-
straints, adaptive movements can be made to ensure that an affordance
remains within the limit of achievability (16). For example in Orth et al.
(17) soccer players running to cross a ball, were shown to modify the
speed they ran based on the distance of an opposing player to the same
ball. In this study, soccer players only ran close to their maximum run-
ning speed capabilities when the opponent was placed at a relatively
close distance to the ball.
On the other hand, with larger changes in constraints, in cases where
a different affordance is realized, a different coordination solution may
arise. Warren (18), for instance, showed that the decision to change how
to climb a stair, either in a bipedal fashion or a quadrupedal fashion, was
predicted by the ratio between the stair riser height and the participants
leg lengths. With larger increases in riser height, the same participant
shifted to a different action. Whilst, there are different views on how af-
fordances are perceived (for discussions see (16, 19)), the key implica-
tions are that, the coordination solution and the information used in
order to achieve a certain objective (e.g., to get a ball before an oppo-

296
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

nent or to climb an object) is often based on the collective constraints

of the individual and their environment, such that action capabilities
(e.g., maximum running speed) and body dimensions (limb lengths) are
scaled to properties of the environment when adapting behaviour.

Addressing individual differences in coordination and control in

striking tasks
Consequently, when executing combat techniques, we assume that
behaviour is controlled based on information that specifies whether or
not the situation (e.g., the distance to a target or opponent) affords hit-
ting such that it satisfies constraints (e.g., to impart a big impact). One
of the promising features of striking tasks are in how previous work has
addressed the influence of individual differences in body dimensions on
the organization of striking actions (see (20), and Figure 1). Specifically,
individual differences in body dimensions have been controlled for by
scaling the distance between the individual and target (L) as a function
of limb length (l), such that the scaled distance (D) is:

Equation 1.
Where the subscript E refers to the effector limb segments used in the
scaling computation (typically either the leg, (11, 19) or arm, (9, 21)). By
changing the distance to a target as function of limb length (i.e., ) the
experimenter/practitioner can reduce the amount of between individ-
ual variability in terms of performance outcomes (11), regulatory move-
ments (12) and techniques used (9).
In addition to addressing individual differences in body dimension,
Hristovski, Davids, Araújo, and Passos, (22) proposed that body scaling
be used as a way of manipulating the hypothetical workspace within
which the individuals current behavioural capabilities can be deter-
mined. To exemplify, in the study by Hristovski, Davids, Araújo and But-
ton, (9) individuals were positioned at 10 different distances from a free
hanging boxing bag. At each position participants were asked to execute
as many different techniques as possible (although limited to 10 strikes).
It was shown that at some distances only two techniques (e.g., left and
right hooks) might be used, whereas, at others, six might be used (i.e.,

297
 sp or t science

left and right hooks, uppercuts and jabs). Thus, for linear changes in ,
participants either continued to use some techniques (perhaps by mak-
ing small movements such as leaning in at the hips, (9)) or completely
new actions emerged whilst others disappeared (see also Figure 1).
In this chapter, we explore how by comparing individuals across a
range of positions their degree of functional movement variability bears
a relationship to their skill. For example, an individual’s current capabili-
ties to produce a high impact collision for a given position might include
a variety of techniques (such as a jab or a hook). This would indicate
they have greater number of degenerate solutions for that set of con-
straints (23). Alternatively, for another individual, at the same position,
they may only be able to effectively strike with a jab. However, this indi-
vidual might be able to use the jab across a range of distances by modi-
fying the way the technique is executed. In this context degeneracy, as a
general property of complex systems and hallmark of skill and learning
(24), reflects that the same outcomes can be achieved in different way.
Practically speaking, prior to learning a new skill, the experimenter or
practitioner can uncover an individual’s pre-existing action repertoire
and plan an intervention accordingly. Following the intervention, the
same procedure can be used to determine how the individual’s action
repertoire is influenced by practice (25).
Skill effects can also be examined in a similar way (26). By fixing cer-
tain constraints (e.g., task goal), and manipulating others (e.g., ) a scan-
ning procedure (see (27)) can be used for a variety of experimental and
practical purposes. Figure 1 exemplifies how a scanning procedure can
be carried out by determining an individual’s ability to use punches and
kicks to strike a bag over a range of position.

298
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Figure 1. Upper left: task. Upper right: experimental paradigm for evaluating
behavioural and movement variability. Bottom: example of different levels
at which movement variability, including control (bottom left) and coordi-
nation (bottom right) that can emerge when changing the individual-target
distance.

In this chapter we consider how functional movement variability, both

in terms of movement variability within a specific technique (12, 21) and
in terms of behavioural variability between techniques (20), might help
individuals to maintain performance as constraints change. We focus on
how individuals adapt to changes in the distance to a target (or the inter-
personal-target distance) as this is most frequently researched. In doing
so, we develop the hypothesis that experienced combat sportspeople
can produce a greater range of techniques as constraints are manipu-
lated and can maintain the effectiveness of any given technique over a
larger range of interpersonal-target distances. Toward this end, there is a
need to integrate the current research of striking actions with the hands
and feet. Across the extant literature, numerous striking techniques
have been examined when manipulating the interpersonal-target dis-

299
 sp or t science

tance. However, techniques are typically examined separately, and with

exception of Hristovsky et al. (9), are proscribed to the participant. Con-
sequently, the functional role of variability at the within and between
technique levels are poorly understood.
Thus, the aims of this chapter are: a) to evaluate how functional
movement variability can be identified as a skilled response to varia-
tion in the interpersonal-target distance, and; b) to address implications
for experimental and practice design. We begin, by detailing the current
understanding of what constitutes an effective strike and how these are
executed.

Determining functionality of striking actions

The assessment of effectiveness of striking actions is generally based
on outcome measures including peak impact force (28), execution time
(12), precision (29) and reaction-time (30). Impact force, in particular,
has been a major focus of previous work where, impact force has proven
to be significantly influenced by a range of variables (e.g., (31, 32)) and in
particular skill level (33, 34). In this section, we present a framework for
understanding how athletes of different skill level achieve an effective
striking action. We do this by integrating a generalized muscle activation
model (35) with models of how interceptive actions are regulated.
Generally, forces imparted to the stuck object, involves the transmis-
sion of momentum from the movement system, via the hand or foot, to
the target mass (such as an opponent, force plate or bag). Because of
the mechanical properties of the movement system (e.g., heterogene-
ous segmental dimensions and material properties), the transmission of
forces is not equivalent to the mass of the system and its acceleration at
impact (36, 37). To account for the expected force relative to the actual
force imparted to the target, effective mass is used to reflect the capabil-
ity (or relative efficiency) to transmit momentum (36). Specifically, effec-
tive mass is defined as the mass being used at impact (21, 38). To think of
this intuitively, if a 5 kg weight is dropped to a (solid) floor, the effective
mass will be around 5 kg. However, if we consider a 5 kg arm colliding
with the same floor the effective mass will be less than 5 kg since the arm
system is not rigid.
Effective mass is computed using the displacement (21, 34, 37) or im-
pulse (29) of the struck object during collision relative to the speed of the
effector immediately prior to collision (29, 34, 37). Effective mass is gen-
erally considered to be influenced by behaviour prior to collision (36, 38),
such as the development of effector speed and/or timing of joint stiffen-

300
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

ing (35). Effector speed and effective mass, however, are not entirely cor-
related (29, 38, 39). For example, it has been reported that effector speed
does not influence effective mass as much as the technique utilized (21)
and both are considered important in determining the resultant peak
impact force (29, 39).
In explaining the production of effective mass and hand speed, McGill
et al. (35) put forward an empirically based model of a neuromuscular
activation process involving three phases: the contract-relax-contract
phases, a phenomenon also referred to as double peak muscle activation
(36). McGill et al. (35) found that across a range of techniques, includ-
ing punching and kicking, in some trunk and hip muscles there is firstly
(phase 1); an initial burst of muscle activation associated with move-
ment initiation, followed by (phase 2); a relaxation of muscle activity as-
sociated with production of effector speed, followed by (phase 3); a final
burst of muscle activation associated with collision.
The initial burst of muscle activation (phase 1), occurring around
movement initiation (35), is believed to provide a base of support (or
inertial mass) from which the striking limb segments can rotate around
or pull against (35, 40). Importantly, during the development of effec-
tor speed, previous work has highlighted the role of a proximal-to-distal
sequencing (41), at least in some techniques (42, 43). Here, action is initi-
ated by applying ground reaction forces (GRF) by the feet (40) and can
include additional stepping movements that further increase GRF (29),
or that provide postural stability (44). Lower limb strength and power
measures, such as maximum isometric force and mean propulsive pow-
er in the squat, have also been correlated with peak impact force (45, 46).
In any case, initial lower limb movements help to develop acceleration
of hip and trunk rotations which can then be used by distal segments to
develop end effector speed (35). For instance, there is some evidence
that end effector speed during kicking is developed by an ordered se-
quencing of peak rotational accelerations around the hip and then knee,
(42, 47). Similarly, this has also been shown in punching, with peak rota-
tions first occurring at the shoulder, then elbow, then wrist (48).
The importance of the relaxation phase (phase 2) may be to prevent
interference with the development of effector speed (35, 37). When mus-
cles crossing a joint are stiffened, movement speed can paradoxically be
reduced (39). Thus, maximizing the time over which certain muscles are
relaxed can facilitate greater effector velocity. In terms of skilled striking
then, following the initial muscle burst, there should be a rapid rate of
relaxation (35). It has been argued that the period of relaxation should

301
 sp or t science

be extended for as long as possible such that reactivation (phase 3) oc-

curs as close as possible in time relative to the collision. In both the trunk
musculature and end effector limb, preparation for impact is particularly
important to transmit both the GRFs through the body (41, 45) and the
momentum developed by the end effector limb segment (39).
Roosen and Pain (49) also argued that athletes should attempt to
have peak velocity and contact coincide. Yet, they found that in some
strikes, peak velocity tended to occur prior to impact, see also (42). In
support, VencesBrito et al. (28) showed that beginners reached forearm
peak angular speed earlier than experienced individuals and tended to
show a significantly earlier pronation at the forearms (corroborated in
the timing of EMG activation), suggesting preparation for impact prior
to collision and at least earlier than their more experienced counter-
parts. Similarly, Bolander et al. (21) examined differences between palm
and fist strikes, finding that the largest forces were achieved with palm
strikes. Also, ‘small moments’ were qualitatively observed around the
wrist during punches, which seemed less important in the palm strike,
improving the transfer of force in to the target with this way of striking.
Finally, Kim et al. (42) observed the peak velocity and impact velocity of
a range of kicking techniques in experts, showing that that peak veloc-
ity does not tend to coincide with the collision, further suggesting that
stiffening occurs before impact.
Thus, an important ability seems to be to coordinate the develop-
ment of effector speed with the onset of muscle stiffening, such that
these movements are regulated with respect to the timing of the colli-
sion. The purpose of joint stiffening seems to be to transfer the momen-
tum developed by the body into the object collided with, but, may also
be balanced with the need to protect joints from strains/dislocations. In
any case, pre-collision joint stiffening suggests information is exploited
about when the impending collision will occur and used to optimize
how hard the object is hit. This information is likely to be specific for
the time remaining to contact (50, 51). From an affordance-based con-
trol perspective, where the individual scales behaviour based on their
capabilities (16), we might speculate that if the task is to strike as hard as
possible, individuals will scale when they begin to stiffen based on some
physical limit, such as, speculatively, contractile velocity or a safety mar-
gin related to joint integrity.
In this section, we have identified a range of movements that can
contribute to effector speed, such as the initial use of the feet or rota-
tions at the trunk. These findings emphasize both, that the entire body

302
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

(e.g., trunk, upper and lower limbs) is implicated in the optimization

of impact forces (41, 45, 46, 52, 53) and that the timing of movements
among limb segments relative to the collision event reflect an important
control problem for how individuals are able to optimize impact forces
(35, 36, 39).
Provided this general overview of what constitutes an effective strik-
ing action under ideal constraints (i.e., against stationary targets at self-
preferred distances), we now consider how movement variability may
help to maintain performance as conditions change, focusing on how
individuals adapt to changes to the interpersonal-target distance.

The functional role of movement variability in striking perfor-

mance
Within technique variability as a functional response to changes in
constraints
Movement variability during execution of one particular technique
can have goal supportive effects, even if this goal will often require low
outcome variability such as high end-point accuracy (50, 54). For in-
stance, changes in distance to the struck object prior to execution influ-
ences the mechanical properties (e.g., effector velocities (21), joint kin-
ematics (12)) of the action. Effectiveness, such as impact force (21) and
execution time (11, 19), can deteriorate when compared to a self-pre-
ferred distance (12, 45, 46). Alternatively, there is evidence that skilled
individuals are better able to maintain performance effectiveness of a
given technique over a larger range of interpersonal-target distances
when compared to less experienced counterparts (11, 19). Taken togeth-
er, these studies, summarized in Table 1, suggest that if individuals are
able to functionally vary the mechanical properties of their movements
for different execution distances, this can be adaptive to maintaining
performance when using a particular technique.

Studies1 Outcome Mechanics

Task description Distance
[Participants] measures2 measured2
Bolander et al. Ungloved 1. Short (Hand Normalized peak NA
(21) Moy Tung maximal (force) length) acceleration
Ving Tsung strike (punch and 2. Medium (1/2 Normalized peak
martial artists palm tested) Arm length) impact force
(3 women, 10 at a padded 3. Long (Arm
men). pendulum (eye length)
and chest height
tested).

303
 sp or t science

Estevan Strike (round- 1. Short (0.68 Total response NA

et al. (55) house kick) at ±0.05-m) time (s)
Taekwondo head of human 2. Normal (1.03 Peak impact (N)
athletes (36 mannequin. ±0.07-m) Body weight
men [different 3. Long (1.03 adjusted peak
weight effects ±0.09-m) impact (N/kg)
tested]). [scaled to leg
length, details
not reported]
Estevan Maximal (strong 1. Short: leg Execution time (s) NA
et al. (11) and fast) strike length x 0.67 Peak impact (N)
Taekwondo (round-house (0.68 ±0.04-m) Impact time (s)
athletes (27 kick) at chest 2. Normal: leg
men [different of human length (1.01
skill groups mannequin. ±0.06-m)
tested]). 3. Long: leg
length x 1.33
(1.35 ±0.09-m)
Falco et al. (19) Maximal (force 1. Close: leg Execution time (s) NA
Taekwondo and speed) strike length x 0.67 Peak impact (N)
athletes (31 (round-house 2. Medium: leg
[different kick) at human length
skill groups mannequin 3. Large: leg
tested]). (height = length x 1.33
participants
abdomen level).
Kim et al. (12) Maximal (effort) 1. Short NA Total linear
Taekwondo strike (round- (difference displacement
athletes (12 house kick) at between Long of the hip (%
men). held pad (height and Normal leg length)
= participants [see 2 & 3]) Hip flexion
abdomen level). 2. Normal (deg)*
(preferred Pelvis
distance) rotation
3. Long (self- (deg)*
selected * = only
maximum) angles
(degrees)
reported as
important
given.
Loturco et al. Gloved maximal 1. Self-selected Peak impact (N) NA
(46) (impact) strike 2. Imposed
Elite amateur (jab punch and (arm length
boxers (6 cross punch [no data])
women, 9 men tested) at a flat
[gender effects impact pad
tested]). (perpendicular to
the ground).

304
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Loturco Gloved maximal 1. Self-selected Peak acceleration NA

et al. (45) (impact vs speed 2. Imposed (g)
Professional tested) strike (length of the
karate athletes (giaku-tsuki [aka. arm [no data])
(10 women, 9 cross punch])
men [gender at sternum
effects of human
tested]). mannequin.

1 = Studies delimited to those comparing performance across different dis-

tances. 2 = Not all outcome measures reported are presented. g = accelera-
tion; NA = not applicable.

Table 1. Studies manipulating the target distance, and the outcome

measures reported.

A number of studies have reported deterioration in performance

outcomes in response to modified distances (45, 46). Loturco et al. (45),
showed that at self-preferred distances, in professional karate athletes
carrying out a maximal impact giku-tsuki (variation of the cross punch),
higher impact forces were recorded compared to when they were re-
quired to punch at an imposed distance. Following this, Loturco et al. (46)
went on to show that, at self-preferred distances, elite amateur boxers
had higher impact forces for the jab technique than at an imposed dis-
tance. In support of these findings, Bolander et al. (21) observed impact
force and hand accelerations across short (hand length), medium (about
half the arm length) and long (arm length) distances to a target, find-
ing that the largest forces and hand accelerations were associated with
striking from the long distance. Similar findings have also been reported
in strikes with the legs. Estevan et al. (55), for instance, observed impact
forces and total response time for the roundhouse kick in Taekwondo
athletes across short, normal and long distances. It was found that the
shortest distance led to the highest peak impact forces and shortest to-
tal execution times. Using more or less the same protocol, Estevan et al.
(11) and Falco et al. (19) also showed evidence of performance deteriora-
tion across three distances, an effect that was more pronounced in less
skilled individuals.
By contrast, in other work, performance did not deteriorate as a func-
tion of changing the distance to the target (45, 46, 55). For example, Este-
van et al. (55) reported that body weight might influence results. In this
study, a range of weight categories were tested (58-68, 68-80 and >80 kg)
across the three distances, reporting that participants from the welter

305
 sp or t science

weight group (68-80 kg) did not show any differences in terms of execu-
tion time or impact force. Loturco et al. (45) also did not find a reduction
in cross punch performance when the task instructions were to maxi-
mize speed of the strike. In a later study, Loturco et al. (46) reported that
performance deterioration was also influenced by technique. The cross
punch did not significantly deteriorate in terms of impact force (46), al-
though it did for the jab technique. One of the possible explanations for
this discrepancy in findings is that the later study involved boxers and
the former karate athletes. In competitive karate, the goal is to contact
the opponent, and, striking with excessive force is penalized. In boxing,
on the other hand, delivering damaging punches can indeed knock out
the opponent to win. Thus, these findings allude to the effect of disci-
pline specific experience on how individuals adapt to varying distances.
In any case, several studies have shown that skilled individuals ap-
pear able to avoid (some) significant performance deterioration as
distances are either increased or decreased. Specifically, Estevan et
al. (11) observed impact force, execution time and impact time in the
roundhouse kick in Taekwondo athletes across three distances (short
[0.67 x leg length], medium [leg length], and long [1.33 x leg length]).
In this study two groups were compared based on skill: a group of ath-
letes who had won a medal at official competitive level, versus a group
whose members had not. Based on group-wise comparisons, no differ-
ences were identified in execution time or impact force at the medium
distance between the groups. However, at the short and long distances,
for both outcomes, the less skilled individuals were significantly worse
than the more skilled group. In support, Falco et al. (19) examined the
effect of distance to target and skill on maximal impact force and execu-
tion time when applying the roundhouse kick. Noting that distance was
also scaled as a function of leg length, it was reported that there was no
significant change in impact force across the three distances in the more
skilled group, which consisted of individuals who had at least a medal at
University or National level competition. In contrast, however, the less
skilled group, which was made up of individuals who practiced the sport
but had never won a University or National level competitive medal,
showed significant differences in impact force (19). Falco et al. (19) did
find however, that the skilled group showed slower execution times at
the close distance and medium distances, indicating how, even in skilled
individuals, some performance outcomes can deteriorate whilst others
can be maintained as constraints are changed.
Only one study (12) could be found to contrast the kinematics of tech-

306
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

niques as the target distance was modified. Thus, there is limited insight
into the adaptive adjustment in movements (i.e., execution variability)
occurring during the performance of specific techniques that might help
skilled individuals maintain low performance or outcome variability as
constraints change. Kim et al. (12) examined the effect of three distances
to a target on kinematics of the trunk, hip and kicking leg when apply-
ing the roundhouse kick in experienced Taekwondo athletes. In this
study, kinematic variables at different movement (including push [start
signal to toe-off], release [toe-off to maximum knee flexion to impact]
and striking phases [maximum knee flexion to impact]) were examined.
Increased kick distance seemed to be achieved by increased pivot hip
displacement (i.e., the leg supporting the body weight displaced toward
the target). Pivot hip displacement appeared to be regulated on the ba-
sis of degree of body posture during the strike (i.e., leaning back such
that the trunk becomes more parallel with the ground) and pivot foot
rotation (12). Specifically, when going from short to long distances, the
pelvic posterior tilt and left rotation increased, whilst both the hip flex-
ion and the trunk´s right lateral tilt decreased (i.e., indicating they were
leaning back) (12). This suggests that controlling the reach of the foot to
the target is achieved by coupling the pelvis and kicking leg, and, pelvis
and trunk movements. During the strike phase, these couplings are be-
lieved to be particularly important for adapting to the degree of pelvis
displacement during the push and release phases (12, 42).
In summary, there is clear evidence that changing the distance to the
target can lead to performance deterioration (45, 46). In addition, there
is some evidence that skilled individuals better maintain performance
as the distance is manipulated (11, 19). There is also the possibility that,
some outcome parameters might deteriorate whilst other outcomes can
be maintained, such as execution time being maintained at the expense
of peak impact force (11, 19). Finally, the kinematics of an action or tech-
nique are also influenced by changing the distance to the target (12). Al-
though there is no direct evidence that this sort of movement variability
supports the ability to maintain performance, in terms of impact force or
execution time, it can be said to at least support general task success of
hitting the target (12).
There is a dire need for future research to establish how movement
variability allows individuals to maintain performance of a technique as
conditions are modified. Indeed, this question is similar to Bernstein’s
problem raised in his work on kinematic trajectories of the arm of black-
smiths (56). During hammering iron, Bernstein showed that blacksmiths

307
 sp or t science

while hammering, although they would move using different trajecto-

ries, they maintained very effective precision and impact.

The role of between technique variability for supporting perfor-

mance
Whilst the above section focused on evidence for functional move-
ment variability as supporting performance when using the same tech-
nique, another feature of striking tasks is the possibility to use different
techniques. Presumably, in combat sports, one of the reasons different
methods of striking have emerged (e.g., hooks, jabs and side-kicks) is
because under different conditions, such as being relatively closer or
further to a target, one technique can be more effective than others. As
an example, if the distance to a target is greater than the ability to reach
it with a punch, the individual must switch to using a leg technique to
meet the objective of striking the target. Other reasons can also be iden-
tified, for instance the nature of an opponent’s guard might mean that a
strike moving primarily through the anteroposterior axis (e.g., a cross or
jab, (31)) is easily blocked and a curved trajectory would be more suc-
cessful (9, 31). Indeed, techniques qualitatively differ, and it is usually
based on the nature of the end effectors trajectory relative to the target
that techniques have been classified. For example, a hook punch and
upper-cut move along different axes (i.e., left-right and craniocaudal re-
spectively), but at the same time both have a curved trajectory (9, 31, 57)
and can conceivably show similar functional angles.
In this section, we review the evidence that shifting between tech-
niques in response to a modified distance between the participant and
the struck object, prior to executing the action, can help maintain or even
improve performance (i.e., is a functional response). In striking tasks
some evidence for this has been uncovered by comparing the relative
effectiveness of techniques across different distances. For example, in
punching techniques, whilst the straight punch might produce the larg-
est impact force, at least in (49, 53, 58), when participants are positioned
at relatively close interpersonal-target distance, this technique does not
emerge as a preferred choice (9). These findings highlight the possibility
that to maintain performance across distances, individuals also have the
option of changing techniques (9, 21, 46, 57).
In Bolander et al. (21), hand speed and impact force were compared
between punch and palm strikes across three distances (short, medium
and long). The results showed that both techniques produced the best
performances at the long distances. Similarly, Loturco et al. (46) com-

308
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

pared the impact forces of jabs and crosses at a self-preferred distance

and at an imposed distance (the length of each individuals arm at full
extension). Notably the jab performance was worse at the imposed dis-
tance, whereas in the cross there were no differences based on distance.
Unfortunately, the self-preferred distances were not reported and thus,
a reason for this technique by distance interaction might be that the self-
selected and imposed distances were very similar for the cross but very
different in absolute terms for the jab. Alternatively, as suggested by the
authors, it is possible that the cross is more effective over a larger range
of distances. According to the authors, the cross is primarily a technique
practiced to achieve high impact, whereas, the jab is a technique primar-
ily used, not for maximal force, but to prepare for other strikes (e.g., used
to ‘open up’ the opponent).
Supporting the idea that individuals have a self-preferred distance
from which they apply different techniques, an innovative work by Petri
et al. (57) showed that at key times (i.e., the hypothetical time needed to
respond to a visual stimulus), experienced karate athletes actively regu-
late the distance from which attacks are initiated based on the technique
being used. Specifically, the jab and reverse punches tended to be initi-
ated from similar inter-personal distances whereas the roundhouse kick
was initiated at a greater inter-personal distance. This study seems to
suggest the performers actively regulated their interpersonal distance to
ensure that a particular technical action remained possible (59).
More challenging to explain are the findings in Hristovski et al. (9).
They suggest that conditions of maximum overlap of possible actions
exist that correspond to the most functional region of performance. That
is, at certain regions multiple actions can be equi-functional reflecting
performance meta-stability. Hristovski et al. (6) conjectured that the
more extensive this region, the greater the individual’s skill, see also
(24). Currently, Hiristovski et al. (9), is the only study identified to ex-
amine conditions where individuals had freedom to select techniques.
To recall, Hristovski et al. (9), had equally skilled individuals (beginner
boxers) placed at 10 equally spaced distances from a heavy bag (begin-
ning with both toes touching the bag, and ending with both toes at an
absolute distance of 1 m). At each position, participants could strike the
bag 10 times and were encouraged to vary the techniques they applied.
Specifically, they were encouraged to use different techniques so long as
they could do so forcefully, with postural stability, and with the percep-
tion of movement efficiency. At small values of Da (~0-0.1) (Da = absolute
distance to the bag/individuals arm length), only left and right hooks

309
 sp or t science

were selected. At small-moderate values of Da (~0.1-0.5), both left and

right hooks and left and right upper cuts emerged. At moderate values of
Da (~0.5-0.9), left and right hooks, left and right upper cuts and left and
right jabs emerged. At large values of Da (1-1.2) only left and right jabs
emerged. Finally, at values of Da greater than 1.2, participants were inac-
tive, unable to reach the target.
According to Hristovski et al. (9) the distance where Da was between
0.5 and 0.9, corresponded to the highest levels of efficiency (for the
striker) and uncertainty (for a hypothetical opponent). It is the region
of meta-stability. A limitation in Hristovski et al. (9) is that ‘…efficiency
was scaled by participants on a 6-point (0-5) continuous scale with 0 re-
flecting absence of a stroke and 5 signifying a maximally efficient stroke.’
This makes it difficult to determine what specifically was influencing the
boxer’s decisions when using strikes at various distances, and, whether
indeed when striking actions overlapped performance was maintained.
There is however, evidence that being able to use more techniques in a
randomized fashion does make it more difficult for opponents to predict
which technique will be used in advance of execution (60). Presumably,
the more techniques an individual can exhibit, the greater opportunity
they have to induce uncertainty in their opponent.
In sum, in this section we have used the interpersonal-target distance
as a parameter to examine the importance of between technique varia-
bility for maintaining performance. In general, the ability to appropriate-
ly shift between techniques with respect to variations in the interperson-
al-target distance seems to be pertinent for skilled combat sport athletes
(57). Future research should, however, uncover clearer evidence that a
greater variety of techniques is associated with improved skill. For exam-
ple, examining skill effects in a task similar to Hristovksi et al. (9) should
reveal that more experienced individuals can demonstrate a larger num-
ber of techniques and maintain them over a larger range of distance (i.e.,
they show a greater region of meta-stability). The main challenge in such
a study is to assess the functionality of strikes in an ‘unconstrained’ set-
up (i.e., where the individual can strike at different angles and positions
– for some innovative approaches see (61, 62)).

Implications that increased movement variability supports per-

formance
A key implication of this review is that individuals able to use a given
striking technique effectively over a broader range of constraint manipu-
lations and able to use more techniques overall, should be more capa-

310
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

ble of adapting to changes in conditions. This would suggest that during

practice, arranging constraints to induce greater movement variability
may be beneficial to the learner (63, 64). It is, however, unclear to what
extent increased practice variability helps individuals to perform effec-
tively in new contexts and also to what extent the purported effect of
practice variability would be modulated by skill. For example, an indi-
vidual who learns to effectively apply a roundhouse kick over a broad
range of conditions is placed in a situation where they cannot use this
kick, may or may not have problems adapting another technique in its
place. The same empirical questions could be raised for an individual
who practices a range of different techniques (see for example the dis-
cussion in Ranganathan and Newell (65) on execution redundancy verse
task variation).
Another possibility that has not been considered, is that exploratory
behaviour, where the individual may aim to learn about theirs and their
opponent’s capabilities under a particular set of constraints can also be
useful for performance (66). Sève et al. (66) for example showed that in
table tennis, skilled players went through an ‘enquiry phase’, exploring
the effectiveness of different performance actions through adopting a
variety of different types of shots, in order to learn about the opponent’s
weaknesses, strengths and strategies. Following this period of explora-
tion (behaviour that was primarily directed toward information gath-
ering), players settled to a phase of performance, where ‘players seek
optimal playing effectiveness by reproducing the strokes they identify
as perturbing to the opponent...’ In paraphrasing Sève et al. (66), explo-
ration was found to be important to all performers across all competitive
match-ups, even in cases where the individual knew the opponent, as
they could not rely on predicting the events of a current match due to
how features of an opponent’s game differ within and across matches.
A possibility is that increased functional movement variability, between
and within techniques, allows the individual to explore new competitive
match-ups.
Taken together, this review implies that the coach or experimentalist
can, and perhaps should, modify constraints (i.e., in the task and/envi-
ronment) during practice with the aim of increasing functional move-
ment variability (and which may include their capabilities to explore in
support of the overall task goal, (67)). In doing so, the learner may in-
crease their capability to adapt and learn under new constraints, rather
than optimize a technique according to some idealized standard (68).

311
 sp or t science

Conclusions
We have developed the argument that increased movement vari-
ability at the within and between levels of action can help individuals
maintain performance as conditions are modified. Functionality in strik-
ing actions involves a balance of developing speed in the end-effector
with accurate timing of movements, including well timed joint stiffening,
relative to the moment of impact. Different methods of striking, such as
cross punches or side-kicks, afford the individual different ways of hitting
a target, including in terms of different locations and also different tra-
jectories (such a curved or straight, and which may be targeted at differ-
ent or the same location). Furthermore, sensitivity to the information, at
least in terms of interpersonal-target distance, for perceiving the bound-
aries within which techniques are effective also seems to be a capability
of skilled individuals. When dealing with opponents, the ability to use
multiple attack methods in randomized fashion can also help inhibit the
opponent’s capacity to predict actions in advance. These predictions
can be tested using experimental procedures similar to those reported
in Hristovski et al. (9). Other predictions have also been developed, and
include that individuals with greater functional movement variability, ei-
ther at within or between technique levels, can adapt effectively to new
constraints (unfamiliar task and environmental manipulations). We also
argue that, during practice, interventions that increase functional move-
ment variability may be beneficial, however, future research needs to
address these questions experimentally.

References
1. Davids K, Glazier P, Araújo D, Bartlett R. Movement systems as dynamical
systems: The functional role of variability and its implications for sports
medicine. Sports Medicine. 2003;33(4):245-60.
2. Newell KM. Constraints of the development of coordination. In: Wade MG,
Whiting HTA, editors. Motor Development in Children: Aspects of Coordina-
tion and Control. Dordrecht: Martinus Nijhoff Publishers; 1986.
3. Renshaw I, Chow JY, Davids K, Hammond J. A constraints-led perspective
to understanding skill acquisition and game play: A basis for integration of
motor learning theory and physical education praxis? Physical Education
and Sport Pedagogy. 2010;15(2):117-37.
4. Newell KM. Change in movement and skill: Learning, retention, and trans-
fer. In: Latash ML, Turvey MT, editors. Dexterity and its Development. New
Jersey: Psychology Press; 1996. p. 393-429.
5. Sporns O, Edelman GM. Solving Bernstein’s problem: A proposal for the
development of coordinated movement by selection. Child Development.
1993;64(4):960-81.
6. Newell KM, Liu YT, Mayer-Kress G. A dynamical systems interpretation of

312
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

epigenetic landscapes for infant motor development. Infant Behavior and

Development. 2003;26(4):449-72.
7. Riley MA, Turvey MT. Variability and determinism in motor behavior. Jour-
nal of Motor Behavior. 2002;34(2):99-125.
8. Chow JY, Davids K, Hristovski R, Araújo D, Passos P. Nonlinear pedagogy:
Learning design for self-organizing neurobiological systems. New Ideas in
Psychology. 2011;29(2):189-200.
9. Hristovski R, Davids K, Araújo D, Button C. How boxers decide to punch a
target: Emergent behaviour in nonlinear dynamical movement systems.
Journal of Sports Science and Medicine. 2006;CSSI:60-73.
10. Newell KM, Jordan K. Task constraints and movement organization: a com-
mon language. In: Davis WE, Broadhead GD, editors. Ecological task analy-
sis and movement. Champaign, IL: Human Kinetics; 2007. p. 5-23.
11. Estevan I, Álvarez O, Falco C, Molina-García J, Castillo I. Impact force and
time analysis influenced by execution distance in a roundhouse kick to
the head in Taekwondo. The Journal of Strength & Conditioning Research.
2011;25(10):2851-6.
12. Kim JW, Kwon MS, Yenuga SS, Kwon YH. The effects of target distance on
pivot hip, trunk, pelvis, and kicking leg kinematics in Taekwondo round-
house kicks. Sports Biomechanics. 2010;9(2):98-114.
13. Riccio GE, Stoffregen TA. Affordances as constraints on the control of
stance. Human Movement Science. 1988;7(2):265-300.
14. Gibson JJ. The ecological approach to visual perception. Boston: Houghton
Mifflin; 1979.
15. Warren WH. The dynamics of perception and action. Psychological Review.
2006;113(2):358-89.
16. Fajen BR. Affordance-based control of visually guided action. Ecological
Psychology. 2007;19(4):383-410.
17. Orth D, Davids K, Araujo D, Renshaw I, Passos P. Effects of a defender on
run-up velocity and ball speed when crossing a football. European Journal
of Sport Science. 2014;14(sup1):S316-S23.
18. Warren WH. Perceiving affordances: Visual guidance of stair climbing.
Journal of Exerimental Psychology: Human Perception and Performance.
1984;10(5):683-703.
19. Falco C, Alvarez O, Castillo I, Estevan I, Martos J, Mugarra F, et al. Influence
of the distance in a roundhouse kick’s execution time and impact force in
Taekwondo. Journal of Biomechanics. 2009;42(3):242-8.
20. Hristovski R, Davids K, Araújo D. Affordance-controlled bifurcations of ac-
tion patterns in martial arts. Nonlinear Dynamics, Psychology, and Life Sci-
ences. 2006;10(4):409-44.
21. Bolander RP, Neto OP, Bir CA. The effects of height and distance on the
force production and acceleration in martial arts strikes. Journal of Sports
Science and Medicine. 2009;8(3):47-.
22. Hristovski R, Davids K, Araújo D, Passos P. Constraints-induced emer-
gence of functional novelty in complex neurobiological systems: A basis
for creativity in sport. Nonlinear Dynamics, Psychology, and Life Sciences.
2011;15(2):175-206.
23. Mason PH, Winter B, Grignolio A. Hidden in plain view: degeneracy in com-

313
 sp or t science

plex systems. Biosystems. 2015;128:1-8.

24. Davids K, Araújo D, Seifert L, Orth D. Expert performance in sport: An eco-
logical dynamics perspective In: Baker J, Farrow D, editors. Routledge
Handbook of Sport Expertise: Routledge; 2015. p. 130-44.
25. Komar J, Chow JY, Chollet D, Seifert L. Effect of analogy instructions with
an internal focus on learning a complex motor skill. Journal of Applied
Sport Psychology. 2014;26(1):17-32.
26. Seifert L, Leblanc H, Chollet D, Delignières D. Inter-limb coordination in
swimming: effect of speed and skill level. Human Movement Science.
2007;29(1):103-13.
27. Kostrubiec V, Zanone PG, Fuchs A, Kelso JAS. Beyond the blank slate: routes
to learning new coordination patterns depend on the intrinsic dynamics of
the learner: Experimental evidence and theoretical model. Frontiers in Hu-
man Neuroscience. 2012;6:1-14.
28. VencesBrito AM, Ferreira MAR, Cortes N, Fernandes O, Pezarat-Correia P.
Kinematic and electromyographic analyses of a karate punch. Journal of
Electromyography and Kinesiology. 2011;21(6):1023-9.
29. Neto OP, Silva JH, de Miranda Marzullo AC, Bolander RP, Bir CA. The effect
of hand dominance on martial arts strikes. Human Movement Science.
2012;31(4):824-33.
30. Martinez de Quel O, Bennett SJ. Kinematics of self-initiated and reactive
karate punches. Research Quarterly for Exercise and Sport. 2014;85(1):117-
23.
31. Piorkowski BA, Lees A, Barton GJ. Single maximal versus combination
punch kinematics. Sports Biomechanics. 2011;10(1):1-11.
32. Joch W, Fritsche P, Krause I. Biomechanical analyses of punching in boxing.
In: Morecki A, Fidelius K, Kedzior K, Wit A, editors. Biomechanics VII-B. 3 B.
Baltimore: University Park Press; 1981. p. 343-9.
33. Smith MS, Dyson RJ, Hale T, Janaway L. Development of a boxing dyna-
mometer and its punch force discrimination efficacy. Journal of sports sci-
ences. 2000;1(6):445-50.
34. Neto OP, Magini M, Saba MM. The role of effective mass and hand speed
in the performance of kung fu athletes compared with nonpractitioners.
Journal of Applied Biomechanics. 2007;23(2):139-48.
35. McGill SM, Chaimberg JD, Frost DM, Fenwick CM. Evidence of a double peak
in muscle activation to enhance strike speed and force: an example with
elite mixed martial arts fighters. The Journal of Strength & Conditioning
Research. 2010;24(2):348-57.
36. Lenetsky S, Nates RJ, Brughelli M, Harris NK. Is effective mass in combat
sports punching above its weight? Human Movement Science. 2015;40:89-
97.
37. Walilko TJ, Viano DC, Bir CA. Biomechanics of the head for Olympic boxer
punches to the face. British Journal of Sports Medicine. 2005;39(10):710-9.
38. Blum H. Physics and the art of kicking and punching. American Journal of
Physics. 1977;45(1):61-4.
39. Pain MT, Challis JH. Soft tissue motion during impacts: Their potential
contributions to energy dissipation. Journal of Applied Biomechanics.
2002;18(3):231-42.

314
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

40. Cheraghi M, Alinejad HA, Arshi AR, Shirzad E. Kinematics of straight right
punch in boxing. Annals of Applied Sport Science. 2014;2(2):39-50.
41. Lenetsky S, Harris N, Brughelli M. Assessment and contributors of punch-
ing forces in combat sports athletes: Implications for strength and condi-
tioning. Strength & Conditioning Journal. 2013;35(2):1-7.
42. Kim YK, Kim YH, Im SJ. Inter-joint coordination in producing kicking veloc-
ity of Taekwondo kicks. Journal of Sports Science & Medicine. 2011;10(1):31-
8.
43. Preuschl E, Hassmann M, Baca A. A Kinematic Analysis of the Jumping
Front-Leg Axe-Kick in Taekwondo. Journal of Sports Science & Medicine.
2016;15(1):92-101.
44. Cesari P, Bertucco M. Coupling between punch efficacy and body stability
for elite karate. Journal of Science and Medicine in Sport. 2008;11(3):353-6.
45. Loturco I, Artioli GG, Kobal R, Gil S, Franchini E. Predicting punching accel-
eration from selected strength and power variables in elite karate athletes:
A multiple regression analysis. The Journal of Strength & Conditioning Re-
search. 2014;28(7):1826-32.
46. Loturco I, Nakamura FY, Artioli GG, Kobal R, Kitamura K, Abad CCC, et al.
Strength and power qualities are highly associated with punching impact
in elite amateur boxers. The Journal of Strength & Conditioning Research.
2016;30(1):109-16.
47. Tang WT, Chang JS, Nien YH. The kinematics characteristics of preferred
and non-preferred roundhouse kick in elite Taekwondo athletes. Journal
of Biomechanics. 2007;40(S780).
48. Nakano G, Iino Y, Imura A, Kojima T. Transfer of momentum from different
arm segments to a light movable target during a straight punch thrown by
expert boxers. Journal of Sports Sciences. 2014;32(6):517-23.
49. Roosen A, Pain MTG. Impact timing and stretch in relation to foot ve-
locity in a taekwondo kicking combination. Journal of Biomechanics.
2006;39(S1):562.
50. Lee DN, Lishman JR, Thomson JA. Regulation of gait in long jumping. Jour-
nal of Experimental Psychology: Human perception and performance.
1982;8(3):448-59.
51. Caljouw SR, van der Kamp J, Savelsbergh GJ. Timing of goal-directed hit-
ting: Impact requirements change the information–movement coupling.
Experimental Brain Research. 2004;155(2):135-44.
52. Filimonov VI, Koptsev KN, Husyanov ZM, Nazarov SS. Boxing: Means
of increasing strength of the punch. Strength & Conditioning Journal.
1985;7(6):65-6.
53. Turner A, Baker ED, Miller S. Increasing the impact force of the rear hand
punch. Strength & Conditioning Journal. 2011;33(6):2-9.
54. Scott MA, Li FX, Davids K. Expertise and the regulation of gait in the ap-
proach phase of the long jump. Journal of Sports Sciences. 1997;15(6):597-
605.
55. Estevan I, Falco C, Álvarez O, Molina-García J. Effect of Olympic weight cat-
egory on performance in the roundhouse kick to the head in taekwondo.
Journal of Human Kinetics. 2012;31:37-43.
56. Bernstein NA. The co-ordination and regulation of movements. London,

315
 sp or t science

England: Pergamon; 1967.

57. Petri K, Lichtenstein M, Bandow N, Campe S, Wechselberger M, Sprenger D,
et al. Analysis of anticipation by 3D motion capturing–a new method pre-
sented in karate kumite. Journal of Sports Sciences. 2017;35(2):130-5.
58. Bartsch A, Benzel E, Miele V, Morr D, Prakash V. Impact ‘fingerprints’ and
preliminary implications for an ‘intelligent mouthguard’head impact do-
simeter. Sports Engineering. 2012;15(2):93-109.
59. Dicks M, Davids K, Button C. Individual differences in the visual control of
intercepting a penalty kick in association football. Human Movement Sci-
ence. 2010;29(3):401-11.
60. Milazzo N, Farrow D, Ruffault A, Fournier JF. Do karate fighters use situ-
ational probability information to improve decision-making performance
during on-mat tasks? Journal of Sports Sciences. 2016;34(16):1547-56.
61. Buśko K, Staniak Z, Szark-Eckardt M, Nikolaidis PT, Mazur-Różycka J, Łach
P, et al. Measuring the force of punches and kicks among combat sport
athletes using a modified punching bag with an embedded accelerometer.
Acta of Bioengineering and Biomechanics. 2016;18(1):57-4.
62. Pierce JD, Reinbold KA, Lyngard BC, Goldman RJ, Pastore CM. Direct meas-
urement of punch force during six professional boxing matches. Journal of
Quantitative Analysis in Sports. 2006;2(2):1-18.
63. Schöllhorn WI, Mayer-Kress G, Newell KM, Michelbrink M. Time scales of
adaptive behavior and motor learning in the presence of stochastic pertur-
bations. Human Movement Science. 2009;28(3):319-33.
64. Chow JY, Davids K, Button C, Shuttleworth R, Renshaw I, Araújo D. The role
of nonlinear pedagogy in physical education. Review of Educational Re-
search. 2007;77(3):251-78.
65. Ranganathan R, Newell KM. Changing up the routine: Intervention-in-
duced variability in motor learning. Exercise and Sport Sciences Reviews.
2013;41(1):64-70.
66. Sève C, Saury J, Ria L, Durand M. Structure of expert players’ activity dur-
ing competitive interaction in table tennis. Research Quarterly for Exercise
and Sport. 2003;74(1):71-83.
67. Seifert L, Boulanger J, Orth D, Davids K. Environmental design shapes per-
ceptual-motor exploration, learning, and transfer in climbing. Frontiers in
Psychology. 2015;6:1819.
68. Carroll TJ, Riek S, Carson RG. Neural adaptations to resistance training.
Sports Medicine. 2001;31(12):829-40.

316
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

The use of power output in road cycling to

optimize training and establish performance
Alejandro Javaloyes1, José Manuel Sarabia1 and Manuel Moya-
Ramón1

Introduction
Road cycling is an endurance sport characterized by a high volume
at low and moderate intensities and specific moments of high intensity
in both, training and racing situations (1). The duration of races is vari-
able, from 4 km of a time-trial to more than 250 km of a road race. In
addition, competitions range from one-day (e.g. Spring classics, World
Championships) to three-week competitions (e.g. Tour of France). Inten-
sity is dependent on several factors, such as tactics, type of stage and
the cyclist’s role, making road cycling a sport that is stochastic in nature
(2, 3). In road cycling, there are two types of stages: On the one hand,
individual and team time-trial and on the other hand, mass-start races
with flat, middle mountain and high mountain profile, according to dis-
tance riding uphill (1). Professional riders cycle around 35.000 km during
training and races in a whole season with a range of 70 to 90 days of com-
petition. Furthermore, there is a specialist for any type of race and with
different functions inside a team. There are five rider types, each with
their own anthropometric characteristics and physiological variables
(4): flat terrain riders, uphill riders, all terrain riders, time-trial special-
ist and sprinters. Therefore, regardless their speciality, cyclists must be
able to perform in a wide range of intensity and duration of efforts. Due
to these characteristics, training and racing control is the main objec-
tive to achieve the best performance during the most part of the season.
In recent years, there has been a huge development of power meters.
These devices are fitted on the bike and allow to measure power output
continuously during both, race and training. Power output is an external
measure of the intensity of effort obtained from the force applied to the
pedal, and the angular velocity with which this force is applied.
Performance in this type of sport is established by several factors:
aerobic performance, anaerobic performance, and efficiency (5). There-
fore, these factors regulate the response to training load and competi-

1. Sports Research Centre, Universidad Miguel Hernandez de Elche, Elche, Spain

317
 sp or t science

tion in the cyclist. This book chapter focused on the aerobic performance
and the ability of power output to monitor it. The two main parameters
derived from the aerobic performance are maximal oxygen uptake
(VO2max) and anaerobic thresholds (lactate or ventilatory threshold, LT
or VT, respectively). In addition, methodologies to establish these pa-
rameters based on power output in both, laboratory and field situations
are presented below.

Optimizing training and racing through power output

In the last decade, the use of power output to measure intensity in cy-
cling has increased its popularity. This is because low-cost and high reli-
able power meters have begun to be commercialized (3). This fact allows
most cyclists (professional or amateur) to be able to attach these devices
on their bikes. Power output is a measure of external load, providing an
intensity of effort instantly, in contrast to internal load measurements
like heart rate (HR) (2), that represents a body’s response to the intensity
(internal load). Although HR was the most used variable and considered
the gold-standard parameter to control intensity in endurance sports
during the past decade, there are many disadvantages derived from its
use (6). For example, HR is modified by many external factors that are
not related to training, such as environmental conditions, mood state,
intake of certain substances like caffeine etc., and the ‘physiological lag’
inherent to HR.
However, power output presents difficulties in its analysis due to the
variability presented by the power values. Due to this, having techniques
that we can use to record and analyse is essential to interpret the data
correctly, obtaining conclusions about the performance of cyclists from
a single training or race to a full competition season. Furthermore, the
analysis of power output during long periods is a tool that enables cy-
cling teams to evaluate weaknesses and strengths of their cyclists and
could be used to select the best cyclists for each type of race or competi-
tion.

Mean Power Output

The use of mean power output of a single session provides an aver-
age measurement of the effort. Although this reduction to a single num-
ber is attractive, the main problem of this variable is that road cycling
is mainly a sport with a high variability in intensity, with long periods
of low and moderate intensities and specific moments of high intensity
(Figure 1). Two different training sessions with the same mean power

318
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

could have a different impact on the cyclist. For example, a continuous

training session of one hour at 200 W session could have as a result the
same mean power output as one hour performing bouts of one minute
at 400 W and another minute of passive recovery (i.e. 0 W), both cases
with a mean power of 200 W. However, the impact of each training ses-
sion will be very different.

Normalized Power Output

To solve this problem and taking into account the facilities of using
a single number, Allen and Coggan (7) proposed the use of normalized
power output. Normalized power output is a fourth power-weighted
30-second moving average. The choice of 30-second windows is based
on many physiological processes, like heart rate, which response is
around this period of time (2, 7). During homogeneous training sessions,
mean power output and normalized power output remain at the same
or similar values but in training sessions with high variability in power
output, normalized power output increases in function of time at the
higher intensities (Figure 1). Nevertheless, normalized power output
seems to overestimate the intensity of effort (8) and further research is
needed to establish the usefulness of this metric.

Variability of Power Output

Due to the stochastic nature of cycling, a high variability in intensity,
and consequently in power output is produced in road cycling. However,
this variability is dependent on the type of race or training. For example,
a flat 40 km Time-Trial could lead to a low variability in power output,
but a high mountain stage, with long up and down hills, has large varia-
bility in power output. There are many procedures to evaluate variability
in any type of signal, including power output in cycling. Exposure Varia-
tion Analysis (9) is a model that takes different efforts and their duration
into account. This model has been adapted to cycling (10). Another easy
option to evaluate variability in power output is the normalized power
output: mean power output ratio, also called Variability Index (7). Values
close to 1 indicate a low variability, on the other hand, the further away
from 1 the greater the variability of power output (Figure 1).

319
 sp or t science

Figure 1. Power output during two different type of training sessions: Continu-
ous training session at a constant workload (Upper panel) and Intervallic
training session with sets of high intensity followed by periods of recovery
between sets (Lower panel).

Record Power Profile

Another option to visualize power data and its distribution is to or-
ganize it in maximal mean power in function of time. The Record Power
Profile, proposed by Pinot and Grappe (11) is one of the best methods
to establish the relationship Power Output-Time. Record Power Profile
records best power outputs in different time intervals from 1 s to 4 h
in both, absolute power output (W) and relative power output to body
weight (W/kg) (Figure 2, upper panel). Record Power Profile was pro-
posed for monitoring long periods and could be used to track changes
in performance, for example an improvement in the power output in dif-
ferent duration of effort (figure 2, lower panel), to establish a cyclist’s
characteristics and to detect possible weaknesses.

320
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Figure 2. Individual Record Power Profile of a professional cyclist (upper) and

its changes in RPP according to the period of the season (lower).

Methodologies to establish aerobic performance in cycling

Maximal Aerobic Capacity: Maximal Oxygen Uptake and Maximal Aer-
obic Power
Maximal oxygen uptake (VO2max) is defined as the maximal capacity
of an organism to extract and consume O2 (Fernandez-Vaquero, 2008).
VO2max has been considered one of the factors that determinate perfor-
mance in endurance sports. However, after 5 – 8 min at VO2max intensity
there is a drop in systolic volume, an acceleration of fatigue, a reduction
in the oxygen uptake and an increase in the anaerobic metabolism to
produce energy (12). Therefore, during a 40 to 80 km Individual Time-
Trial the intensity is around 75 -85 % of VO2max while in shorter efforts
(e.g. a 4 km Time-Trial) the intensity is closer to VO2max. This fact does
not invalidate the importance of VO2max in performance, considering that
professional cyclists had a high value of this parameter and, for example,
a three-week race is won by cyclists with high VO2max (~85 ml·kg-1·min-1).
Maximal aerobic power (MAP) is the power output obtained at
the VO2max intensity. Likewise to VO2max, MAP is a parameter that indicates

321
 sp or t science

the aerobic potential of athletes (13). There are several methods to es-
tablish MAP, both in laboratory and field conditions. The Graded Exercise
Tests (GXT) is considered the gold-standard methodology for the deter-
mination of MAP. Typical increments in intensity range between 20 and
30 W with durations between 1 and 3 min (14). An example of a typical
protocol with 25 W·min-1 increments is shown in Figure 3. However, the
presence of different protocols in the scientific literature make the choice
of selecting one or another difficult in their selection. These protocols
have some limitations, for example, VO2max and MAP values can fluctuate
according to stage duration or power increments, (15, 16). Furthermore,
the test is performed in laboratory conditions that are not “real” training
or racing conditions.

Figure 3. Power Output during a GXT

Pinot and Grappe (17) proposed a method to assess MAP in field con-
ditions based on an individual’s RPP. MAP is determined by modelling
the aerobic metabolism from RPP with a linear PO-Logt relationship be-
tween 5 min to 4 h (18). The first record power output included in a con-
fident interval was taken as a MAP, and the sustained time of this power
output was the time that this intensity can be sustained (17). This is a
remarkable method to establish variables from the field because it does
not require additional tests. Furthermore, it provides measures of power

322
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

output taken directly from the field (training and racing) with a high level
of accuracy.

Anaerobic threshold
Although MAP is an important parameter to predict performance in
cyclists, when we talk about training, the boundary of sustainable per-
formance seems more important than MAP mainly because this “thresh-
old” has the highest correlation with performance and is more sensitive
to performance changes during the training period (19, 20). This bound-
ary is a dividing line between heavy and very heavy exercise, in other
words, separating sustainable from unsustainable constant exercise.
However, numerous physiological “thresholds” exist in the literature for
exercise prescription and training load quantification, each of them with
their own name and method for determination. In this section, we try
to present the most reliable and easy methods to identify this point in
power for training. Other highly reliable and accurate methods exist (i.e.,
Maximal Lactate Steady State [MLSS]), but nevertheless they require
multiple sessions to assess it (16).
First, we talk about the anaerobic threshold which can be calculat-
ed with a respiratory exchange during GXT, and is considered the gold-
standard method to identify this boundary (16). It is important to know
that a variety of techniques for detecting this point exist, and can receive
different names (i.e., lactate threshold, second ventilatory threshold,
respiratory compensation point, etc.), but all of them refer to the same
theoretical point. Anaerobic threshold occurs at the point in which lac-
tate is rapidly increasing with intensity and represents hyperventilation
relative to the extra CO2 that is being produced. The biggest problem
with these methods was the cost of GXT, which needed expensive de-
vices and a specialist to analyse the ventilatory data.
In the case of having no access to a GXT with respiratory exchange
analyse, other types of test can help to identify this “threshold” so im-
portant to training. With a similar incremental protocol like in GXT or an
incremental cadence protocol against a constant load (21), it is possible
to obtain a heart rate deflection point (HRDP). This method only requires
a heart rate monitor to record the cyclist’s HR during the incremental
test. After that, with a simple visual (or more accurately, with a math-
ematical analysis) it is possible to identify the point in which the slope
of the intensity-HR relationship decreased (Figure 4). However, the prob-
lem arises in cyclists without this HRDP. Many researchers have identi-
fied this point only in a portion of samples, as an example, Lucia, Hoyos

323
 sp or t science

(22) reported the HRDP only in 56% of top-level cyclists and Jones and
Doust (23) in 9 out of 15 well-trained male distance runners.

Figure 4. Detection of anaerobic threshold through the heart rate deflection

point method (HRDP).

With the same mathematical or visual method and recent portable

technology (i.e., Near Infrared Spectroscopy) it is possible to detect a de-
oxyhaemoglobin breakpoint in the evolution of this variable along the
incremental test (24). This point is also related to anaerobic threshold
(25). The popularity of Near Infrared is increasing among athletes be-
cause their size has been reduced, they have become wireless and they
are becoming more accessible. In addition, these devices are compatible
with most of the popular systems used to record training sessions, there-
fore making them easy to integrate for athletes.
Despite the reliability of these methodologies, all of them require
complementary devices to assess anaerobic threshold and its associ-
ated power output. Because of that, methods which do not need more
than a power meter are increasing their popularity. These methods are
related to the ‘critical power’ concept which is defined as the asymptote
of a curvilinear power-time relationship (Figure 5), in other words, it rep-
resents the power that can be theoretically maintained for a very long
period (i.e., more than 30 min). Currently, there is controversy around
this concept because some researchers have related this point with the
anaerobic threshold while others indicate that they are close but are dif-

324
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

ferent boundaries (16). Nevertheless, as a practical application, Critical

Power is being used satisfactorily as a point to differentiate heavy from
very heavy exercise (24, 26). Concretely, the most used tests in the litera-
ture and in the field are the time-to-exhaustion trials (27) and a single 20
min time-trial (7, 28).

Figure 5. An example of the critical power model and their representation

through three time-trials to exhaustion with different constant loads.

Regarding time-to-exhaustion trials for the determination of Critical

Power, the cyclist needs to perform three to five constant-load trials to
volitional exhaustion with a duration between 1 and 20 min (29), and a
rest interval between each 30 min trial (20, 30). Mean power output of
each trial will be used to perform a distribution for a curvilinear time-
to-exhaustion relationship (Figure 5). Although this may sound very
complicate to do, many of the software programs used to control the cy-
cling training integrate the option to calculate the Critical Power indicat-
ing only the mean power output and duration of each trial. In single 20
min time-trials, also known as ‘functional threshold power test’, a term
coined by Allen and Coggan (7), the mean power in this test was directly
related to Critical Power and ventilatory threshold (20). Different recom-
mendations are found between performing the test indoors or outdoors,
and with or without a slope. Even though the previous multiple time-
trial tests seem more reliable, a 20 min time-trial test is less fatiguing and
easier to perform.

325
 sp or t science

Practical applications and Conclusions

Road cycling is a sport with high metabolic demands in both, train-
ing and competition situations. Power meters have become one of the
most used tools to measure the intensity of effort in professional cycling.
Although there is a high amount of data available after finishing any re-
cording, the analysis and interpretation of power output data allows a
measure of the external load and could be used to track changes in fit-
ness. The use of mean power output and normalized power combined
with variability index allows a valid measure of the global impact or the
training load. Furthermore, the RPP is a valid procedure to assess chang-
es in fitness and performance and to establish intensity zones.
Aerobic performance plays a key role to achieve a good readiness-to-
perform. Measuring VO2max and anaerobic threshold and their derived
power output variables allow the measurement of changes in fitness and
performance. Although GXT is considered the gold standard to measure
aerobic performance, field tests provide performance measures in “real”
cycling conditions that could help cyclists and coaches without access to
a laboratory to optimize training and establish performance.

References
1. Padilla S, Mujika I, Orbananos J, Santisteban J, Angulo F, Goiriena JJ. Exer-
cise intensity and load during mass-start stage races in professional road
cycling. Medicine and Science in Sports and Exercise. 2001;33(5):796-802.
2. Jobson SA, Passfield L, Atkinson G, Barton G, Scarf P. The analysis and uti-
lization of cycling training data. Sports medicine. 2009;39(10):833-44.
3. Passfield L, Hopker JG, Jobson S, Friel D, Zabala M. Knowledge is power:
Issues of measuring training and performance in cycling. Journal of Sports
Sciences. 2016:1-9.
4. Mujika I, Padilla S. Physiological and performance characteristics of male
professional road cyclists. Sports Medicine. 2001;31(7):479-87.
5. Joyner MJ, Coyle EF. Endurance exercise performance: the physiology of
champions. The Journal of physiology. 2008;586(1):35-44.
6. Sanders D, Myers T, Akubat I. Training Intensity Distribution in Road Cy-
clists: Objective Versus Subjective Measures. International journal of
sports physiology and performance. 2017:1-20.
7. Allen H, Coggan A. Training and racing with a power meter: VeloPress; 2012.
8. Skiba PF. Evaluation of a Novel Training Metric in Trained Cyclists: 2415:
Board# 94 June 1 2: 00 PM 3: 30 PM. Medicine & Science in Sports & Exer-
cise. 2007;39(5):S448.
9. Mathiassen SE, Winkel J. Quantifying variation in physical load using expo-
sure-vs-time data. Ergonomics. 1991;34(12):1455-68.
10. Abbiss CR, Straker L, Quod MJ, Martin DT, Laursen PB. Examining pacing
profiles in elite female road cyclists using exposure variation analysis. Brit-
ish journal of sports medicine. 2010;44(6):437-42.

326
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

11. Pinot J, Grappe F. The record power profile to assess performance in elite
cyclists. International journal of sports medicine. 2011;32(11):839-44.
12. Gonzalez-Alonso J, Calbet JA. Reductions in systemic and skeletal muscle
blood flow and oxygen delivery limit maximal aerobic capacity in humans.
Circulation. 2003;107(6):824-30.
13. Pinot J, Grappe F. A six-year monitoring case study of a top-10 cycling
Grand Tour finisher. J Sports Sci. 2015;33(9):907-14.
14. Pettitt RW, Clark IE, Ebner SM, Sedgeman DT, Murray SR. Gas exchange
threshold and VO2max testing for athletes: an update. Journal of strength
and conditioning research. 2013;27(2):549-55.
15. Hopkins WG, Schabort EJ, Hawley JA. Reliability of power in physical per-
formance tests. Sports medicine (Auckland, NZ). 2001;31(3):211-34.
16. Pallares JG, Moran-Navarro R, Ortega JF, Fernandez-Elias VE, Mora-Rodri-
guez R. Validity and Reliability of Ventilatory and Blood Lactate Thresholds
in Well-Trained Cyclists. PloS one. 2016;11(9):e0163389.
17. Pinot J, Grappe F. Determination of Maximal Aerobic Power from the Re-
cord Power Profile to improve cycling training. Journal of Science and Cy-
cling. 2014;3(1):26.
18. Grappe F. Methods to assess performance in elite cyclists in the field. As-
petar Sports Medicine Journal. 2016;5:434-8.
19. Lucia A, Hoyos J, Perez M, Santalla A, Earnest CP, Chicharro JL. Which labo-
ratory variable is related with time trial performance time in the Tour de
France? British journal of sports medicine. 2004;38(5):636-40.
20. Nimmerichter A, Williams C, Bachl N, Eston R. Evaluation of a field test to
assess performance in elite cyclists. International journal of sports medi-
cine. 2010;31(3):160-6.
21. Grazzi G, Mazzoni G, Casoni I, Uliari S, Collini G, Heide L, et al. Identification
of a Vo2 deflection point coinciding with the heart rate deflection point
and ventilatory threshold in cycling. Journal of strength and conditioning
research. 2008;22(4):1116-23.
22. Lucia A, Hoyos J, Santalla A, Perez M, Carvajal A, Chicharro JL. Lactic aci-
dosis, potassium, and the heart rate deflection point in professional road
cyclists. British journal of sports medicine. 2002;36(2):113-7.
23. Jones AM, Doust JH. Lack of reliability in Conconi’s heart rate deflection
point. International journal of sports medicine. 1995;16(8):541-4.
24. Keir DA, Fontana FY, Robertson TC, Murias JM, Paterson DH, Kowalchuk
JM, et al. Exercise Intensity Thresholds: Identifying the Boundaries of Sus-
tainable Performance. Med Sci Sports Exerc. 2015;47(9):1932-40.
25. Bellotti C, Calabria E, Capelli C, Pogliaghi S. Determination of maximal lac-
tate steady state in healthy adults: can NIRS help? Med Sci Sports Exerc.
2013;45(6):1208-16.
26. Rossiter HB. Exercise: Kinetic considerations for gas exchange. Compre-
hensive Physiology. 2011;1(1):203-44.
27. Karsten B, Jobson SA, Hopker J, Stevens L, Beedie C. Validity and reliabil-
ity of critical power field testing. European Journal of Applied Physiology.
2015;115(1):197-204.
28. Bentley DJ, McNaughton LR, Thompson D, Vleck VE, Batterham AM. Peak
power output, the lactate threshold, and time trial performance in cy-

327
 sp or t science

clists. Med Sci Sports Exerc. 2001;33(12):2077-81.

29. Morton RH. A 3-parameter critical power model. Ergonomics.
1996;39(4):611-9.
30. Galbraith A, Hopker J, Lelliott S, Diddams L, Passfield L. A single-visit field
test of critical speed. International journal of sports physiology and per-
formance. 2014;9(6):931-5.

328
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Characteristics of open water lifesaving

J. Arturo Abraldes1,2 and Ricardo J. Fernandes3,4

Introduction
Lifesaving sport contains many disciplines that take place in swim-
ming pool and open water. All these disciplines are born from the imita-
tion of the rescues that lifeguards undertake while on duty in inland and/
or open water environments. However, lifesavers reproduce the rescue
sequences and demonstrate the domain of specific rescue materials not
aiming to save the “victim” but to act in the fastest possible way (1,2).
This characteristic of speed makes the sport interesting, challenging
and exciting (3,4), but while in some countries it is popular (e.g. Spain
and Australia), in some others it is unknown or not interesting to be ex-
ercised. This is the case even in other geographic areas close to rivers,
lakes and seas.
Lifesaving in Europe is operated by the International Life Saving Fed-
eration of Europe (5), a member of the International Life Saving Federa-
tion (6). Its main institutional responsibility is the promotion and diffu-
sion of water safety throughout the world. The regular participation in
national and international competitive events promotes higher lifesav-
ing experience and knowledge, particularly of its training process, struc-
ture and organization. This know-how is fundamental and allows the
sport to develop its bases, to be established and spread. In addition, it
assists its promotion at local level and to obtain a greater awareness and
education allowing reducing the number of drowning accidents (7,8).
We consider very important to disseminate and reinforce the existing
knowledge on lifesaving to promote it and increase the number of prac-
titioners. Lifesaving is complex, with a relationship between the aquatic
displacement techniques and the use of specific materials, allowing a
greater control of the shore and aquatic environments. Complementari-
ly, preventive habits instilled with its practice are as (or more) important

1 Faculty of Sport Sciences, University of Murcia, Spain

2 Lifesaving Federation of Galicia, Spain
3 Centre of Research, Education, Innovation and Intervention in Sport, Faculty
of Sport, University of Porto, Porto, Portugal
4
Porto Biomechanics Laboratory, University of Porto, Porto, Portugal

329
 sp or t science

as the rescue techniques, leading to the prevention of accidents and to

a reduction in the number of rescues (4). In this sense, in addition to the
sportive instruction of the practitioner, lifesaving aims enhancing the in-
tegral education of the individual, particularly regarding the respect and
understanding of the dangers that exist in each aquatic environment.
Further to previously published work (e.g. 9,10), the aims of this chap-
ter are: (i) to disclose lifesaving as a regulated sport with a specific com-
petitive calendar; (ii) to identify the specific characteristics of the differ-
ent lifesaving events carried out in natural spaces (at the beach); (iii) to
propose a proper teaching model for the advancement of learning and
the improvement of the sport techniques; and (iv) to summarize the lat-
est findings of scholarly work that has been undertaken about lifesaving.

Development
Competitive lifesaving can take place at the swimming pool or on the
beach, presenting diverse individual and team events in the juvenile (15-
16 years), junior (17-18 years) and absolute (over 15 years) age groups
(11). In addition, contrary to what many people might think that only
professional lifeguards and / or adults are engaged in competition, there
are also competitive events for younger individuals (just above seven
years of age). These events for younger age groups have adapted dis-
tances, materials and technical elements (as described in Table 1).

Table 1. Lifesaving open water competitive events by age group.

Age groups (years old): 7-8 9-10 11-2 13-4 15-6 17-8 >15 >30
Individual events
Surf race · · · · · · · ·
Board race · · · · · · · ·
Surf ski race · · · ·
Beach flags · · · · · · · ·
Beach sprint · · · ·
Run–swim–run · · ·
Ocean man/woman · · ·
Team events
Board rescue · · · · · ·
Rescue tube rescue · · · · · ·
Beach run relay · · · · · · ·
Ocean man/woman relay · · ·
Ocean man mixed · · ·

330
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

For a better understanding of this sport, there is a need to know the

characteristics of the different competitive events (i.e., the disciplines),
namely understanding its general rules (in which the events are de-
scribed) and the specific organization procedures of each championship
(e.g. how many series are there, how many swimmers can access the fi-
nals and how many points the winner can obtain). So, in Tables 2 and 3,
it is described the standard international individual and team events for
youth, junior and absolute age groups, highlighting the details that influ-
ence the final competitive outcomes (1,11-13).

Table 2. Brief description of the lifesaving all age-groups individual events

carried out on natural spaces (adapted from 3,7).

Surf race
After the auditory signal, lifesavers
enter the water, swim around the buoys
clockwise, return to shore and end their
path between the two green flags.

Board race
With three buoys in triangular position
indicating the route, the lifesavers enter
the water after the starting signal, paddling
with the hands their board. They surpass
the buoys clockwise, until arriving at the
finish line in contact with the board.

Surf ski race

Three buoys in triangular position in-
dicate the path that the lifesavers must
follow. They are in-water holding the boat
and, after the exit signal, rise and move on
it, passing the buoys clockwise. They must
cross the finish line on the boat.

331
 sp or t science

Beach flags
Lifesavers are in lying prone position
and some flags are placed 20 m apart (one
less that the number of competitors). Af-
ter the exit signal, each swimmer lifesaver
tries to catch a flag located behind him.
Those that do not get it are eliminated.

Beach sprint
The competition area is marked in 70-90
m lanes separated by coloured ropes. After
the auditory signal, lifesavers must run in
straight-line, barefoot and without help of
any artificial material towards the finish
line.

Run–swim–run
From the start line, competitors run 200
m, surround the flag and enter the water
swimming around the buoys. Then, they
return to the beach, turn on the flag and
run 200 m towards the finish line.

Ocean man/woman
After the exit signal, swimmers’ life-
savers carry out the specified trajectories
for the surf, board race and surf ski race
events, finishing running on the sand until
the finish line. The order of the segments is
draw and decided beforehand.

In team competitions, in addition to individual efficiency, close attention

should be given to the coordination and collaboration between team
members, which will significantly affect the event outcome.

332
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Table 3. Brief description of the lifesaving all age-groups team events carried
out on natural spaces (adapted from 3,7).
Board rescue
At the starting signal, the first competi-
tor swims until the corresponding buoy,
indicating its arrival touching it with one
hand and raising the other. His team-mate
enters the water with the lifeline and ap-
proaches the colleague that will rise to the
board and collaborate on the way back.

Rescue tube rescue

With four participants, at the starting
signal, the first one swim until its corre-
sponding buoy and raises one upper limb.
The second one swim with fins and a res-
cue tube to the buoy attaching it to the col-
league and towing him to shore. The other
two lifesavers can assist in the extraction
until the finish line.

Beach run relay

With four barefoot participants in each
team, the first competitor at the end of his
run gives the baton to the second, who
must be in the changeover zone. He shall
deliver the baton to the third and this one
to the fourth, who will finish the event by
passing the finish line.

Ocean man/woman relay

Each of the four participants will make
a leg (surf, board race and ski surf race, by
the order established in the draw) after be-
ing touched by the incoming team-mate.
The fourth lifesaver ends the relay with a
sprint on the beach until the finish line.

333
 sp or t science

Ocean mixed
A path equivalent to two ocean man/
woman races will be made by a team of
three male and three female lifesavers.
Each participant will take a leg after being
touched by the previous and the last par-
ticipant will end the relay with a sprint on
the beach until the finish line.

Lifesaving classification
Of the various possible classifications for this sport, we use the one
that considers the motor situations as systems whose components inter-
relate and generate a certain degree of uncertainty in the participants
(14). This classification was previously used by others (see 15-17) and is
based on the following assumptions (Figure 1): (i) “C” refers to the exist-
ence of teammates and, consequently, to the need for motor collabora-
tion relationships; (ii) “A” refers to the existence of adversaries and the
relationship of motor opposition; and (iii) “I” refers to the fact that the
environment where the sport occurs offers uncertainty, i.e., if the sport
context provides relevant information that implies adaptation by the
practitioner or if it is closed and normalized.
The three above mentioned criteria, combined binary and underlined
(just in case that one, two or three elements are absent), lead to the es-
tablishment of eight distinct categories arranged in tree shape, in which
lifesaving is situated (18). Thus, depending on the presence/absence of
one, two or three of these elements, it will be defined the category in
which the sport in question is placed (when one of the three elements is
missing, the corresponding letter is highlighted; Figure 1). After observ-
ing the general social and psychomotor characteristics of lifesaving, and
understanding the motor actions of each, it will be possible to dispose of
a rigorous base to propose motor games, exercises and/or tasks that fit
each group of events and what they require.

334
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Figure 1. Tree diagram represent- Figure 2. Representation of the beach

ing the eight categories and sprint event in lifesaving (adapted
corresponding classification from 17)
criteria of sport events (adapt-
ed from 18).

According to the above-mentioned criteria, only the beach sprint

event would have the psychomotor profile (Figure 2), in which the re-
lationship with the other lifesavers is not relevant. This is a typical case
of a psychological non-motor interaction, in which the opponent does
not interfere with (or influence) the motor actions of the other partici-
pants. Here, there are situations of co-motricity, existing an action in the
presence of others but without a body or instrumental interaction (18),
allowing the lifesaver to see the actions of his opponent that can serve
as an emotional stimulus. In these events there are not colleagues and
opponents within a stable physical environment and, because they can-
not leave their respective lanes where exercise is performed, each com-
petitor does not have the collaboration of teammate(s), nor opposition,
hindering or interactions of opponents.
No doubt that the beach flags event is more complex from a strategic
point of view, since members of the same team can compete in the same
race. Therefore, “CAI” gives a new classification of lifesaving (Figure 3),
making essential the collaboration with peers and the clear interaction
with opponents (occurring in a stable and non-changeable environ-
ment). The existence of interactions with colleagues requires a well-
established preparation of the event, since there will be moments that
one, two or three lifesavers of the same club are running simultaneously
and should cooperate to obtain benefits to the team. This is, without a
doubt, an event with multiple strategic possibilities (that can be estab-
lished before and during the race). According to the lifesaving rules, the
competitor cannot grasp or push the opponents, but it is possible to use
the body positioning before the flag collection, which makes evident the
“opposition” nature of the event.

335
 sp or t science

Figure 3. Representation of the Figure 4. Representation of the other

beach flags event (adapted individual events (adapted 17).
from 17).

When other open water individual events (surf race, boar race, surf ski
race, run-swim-run and ocean man/woman) are analysed, it is observed
no essential communicative interaction, i.e., there is no collaboration
(although the presence of adversaries that can influence the course of
the event; cf. Figure 4). It is also evident the high uncertainty of the sea,
plagued by currents and waves (depending on the place on the event
occurs) determining the lifesavers’ actions. The improbability of the sea
requires a large environment deciphering because, although it is not a
completely unknown environment as it is partially signalised (e.g. as a
river descent), the lifesaver needs to interpret the signals and adapts his
motor actions. In fact, despite being partially signalized, there are no
lanes delimiting the right course, evidencing its socio-motor character-
istics. All these imply that participants make their decisions depending
on the environment and the opponents, having automatized their cyclic
propelling movements (with or without materials) but, simultaneously,
needing to adapt to the opponents’ actions.
Allteameventspresentthetreeexponentialbranch(Figure1)thatendswith-
in the socio-motor situations, even if they occur in a space divided by lanes and
without opponents’ relevant interactions (e.g. beach run relay). In this event,
“CAI” shows a cooperative type interaction with other participants, namely
during the moment of the touch when giving the baton (Figure 5), i.e., it re-
quires evident cooperation in-between team components. Despite being
conducted in open water and on the sand, this is a closed event since the
paths are specifically delimited and the competitive space is well estab-
lished. Without a doubt, it is a rather disputed competition, with more em-
phasis on the physical conditioning than on tactics.
The board rescue, rescue tube rescue, ocean man/woman and ocean
mixed events are much more complex, with interactions with peers and

336
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

opponents, within an unstable environment such as the sea (Figure 6).

These “ingredients” evidences their high complexity, with actions of all
kinds and well-coordinated cooperation, where team training is decisive.
They are highly complex events in terms of motricity, i.e., competitions
in which lifesavers should decipher the opponent actions that might
have strategic significance, need to interpret the sea fluctuating infor-
mation, as well as to adapt themselves to the other participants’ actions.

Figure 5. Representation of the Figure 6. Representation of the other

beach run relay event (adapt- team events (adapted from 17).
ed from 17).

The different lifesaving events present five different branches in the

exponential tree diagram, indicating a great motor and situational com-
plexity. In that diagram it is possible to observe that four “CAI” groups
of events end in socio-motor situations. Thus, lifesaving is considered
a sport of great socio-motor character, contrary to what common sense
could tell. The fundamental motor actions technical aspect in all events
(which would not agree with the greater tactical aspect of the socio-mo-
tor practices in general) should not be considered an error, but as the
consequence of the variety of its fundamental characteristics, a very rare
feature in most sports. Therefore, lifesavers must master those aspects
that relate to psychomotor situations, with characteristics such as (18):
(i) a tendency to motor stereotype; (ii) a pre-programed behaviour and
anticipation; (iii) predominantly proprioceptive regulation; (iv) omis-
sion of motor decoding; (v) intense training; and (vi) competitive results.
However, the large number of branches and socio-motor events in which
it could be found performance relevant aspects should not be unconsid-
ered, such as: (i) the high importance of the prior strategy and strategic
decision of the player and (ii) the training of key moments at the time of
collaborate with colleagues effectively.

337
 sp or t science

Analysis of sports initiation and high-performance participation

Sport initiation begins when the individual has a basic and general
enrolment with the overall set of activities, being included on a learn-
ing process aiming to acquiring a practical (and conceptual) contents
domain (19-21). This teaching process is of special importance to the
learning success and contrasts with the typical focus on the result that
coaches usually privilege.
A detailed analysis of sport initiation specifically focusing on lifesav-
ing is already available (cf. 1,3,9), in which the topics sport initiation,
training context, reasons for selecting this sport and the context in which
the lifesaver is inserted are well described. The analysis of these items
shows similar data for the two lifesaving modalities (swimming pool and
open water events), particularly regarding the fact that lifesavers engage
in this sport at the age of 12 (with a decrease tendency in the last years
due to the growing number of participants in lower age-groups; 15).
Participants usually begin their competitive participation in swim-
ming pool conditions rather than in open water events, with conven-
tional swimming skills as the basis for the posterior development of
specific rescue techniques. In the most cases, lifesavers keep practicing
two sports (lifesaving and swimming) during the training season, with
their training processes improving in quality with the increasing number
of coaches with higher education in sports science, as well as with bet-
ter quality specific material resources. The fact that they train, at least,
four to five workouts per week, combining water and dry land exercises,
is leading to significant performances improvement with numerous re-
cords in the last decade.
The main motivation for lifesaving practice is intrinsic, mainly sup-
ported in the enjoyment for the activity (22). This leads to a strong en-
gagement on the sport since the lifesavers that have initiated practice
in yearly ages are still involved and competing in absolute or masters
categories. This is also reinforced by extrinsic factors, particularly by the
pleasure to achieve success in an ample number of events and competi-
tions, and by the environment around the lifesaver. In fact, family and
coaches are the lifesavers main supporters (23) as their parents are also
engaged in sport practice (in general) and coaches are also involved with
lifesaving practice.

Analysis of technical elements

In the following points it will be described, in the order that they
are used in competition, the technical elements included in open wa-

338
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

ter lifesaving events (1,2,7,24): (i) running, which relates to the different
displacements of the lifesaver along the beach soft or hard sands (drier
and wetter, respectively) to reach the water or as a part of the event (e.g.
beach flags, beach run relay or run-swim-run); (ii) water entry, which
reflects to the moment of water entering in contact with the material
later used during the event or without it (in those events that material
is not needed); (iii) swimming approach, i.e., the propelling techniques
for moving in-water between the points previously defined, depending
directly on the event characteristics, specific rules and lifesavers physi-
cal conditioning; (iv) swimming approach with and aid, regarding to the
lifesaver in-water movement mounted on a specific material (according
to the event rules), never loading or pulling it; (v) placing material, refer-
ring to the placement of different material by the lifesaver to itself during
the event (not beforehand);
Other open water lifesaving events technical elements are: (vi) plac-
ing material to the “victim”, related to the placement of different ma-
terial by the lifesaver to a “drowning” partner (according to the event
rules) that will later be used to tow and continue or end the event (the
“victim” can grab the rescue aid and assist in the towing); (vii) towing,
referring to the transportation of a colleague through the water, always
with the airways above the surface so the “victim” can breathe (the di-
rect towing is done with body to body contact and without material, with
the indirect technique requiring the use of material); (viii) moving on the
material, by transporting the “victim” on lifesaving specific material,
who can help with upper and/or lower limbs actions to make the course
as fast as possible; (ix) visual signalling, i.e., visually perceiving a specific
teammate gesture authorizing the beginning of the following lifesaver
action; (x) coordination with the partner, referring to the collaboration of
two or more lifesavers of the same team with the intention of correctly
executing different event actions aiming for a better final result; and (xi)
water extraction, related to one or more lifesavers action when extract-
ing a “victim” from the water to a firm and safe place (the beach sand).
Some technical elements are used in most of the events (evidenc-
ing its importance), while others only in one (with lower transference to
the professional lifeguards’ activities; Table 5). To begin practicing this
sport, a great variety of technical aspects should be taught. In fact, even
if not performed perfectly, it is very important to have a substantive va-
riety of actions, leading to a future better control of the overall lifesaving
techniques.

339
 sp or t science

Table 5. Motor actions present in open water lifesaving events (adapted from 7).
Events

Beach run relay

Run-swim-run

Board rescue

Ocean mixed
Beach sprint

Rescue tube
Surf ski race

Ocean relay
Beach flags
Board race
Surf race

rescue
Ocean
Running · · · · · · · · · · ·
Water entry · · · · · · ·
Water entry with material · · · · · ·
Swimming approach · · · · · · ·
Swim approach material · · · · · ·
Placing material on itself ·
Placing material on victim ·
Towing (indirect) ·
Moving on the material · · ·
Visual signalling · · ·
Coordination with partner · · · ·
Water extraction ·

In addition, the use of materials in the different lifesaving events evi-

dences a high complexity comparatively, for example, to conventional
swimming (Table 6). This forces the lifesaver to dominate specific mate-
rial, overcoming him aiming for greater performance effectiveness.

Table 6. Materials used in open water lifesaving events (adapted from 7).

Events
Board rescue

Ocean mixed
Beach sprint

Rescue tube
Surf ski race

Ocean relay
Beach flags
Board race

Run-swim-

Beach run
Surf race

rescue
Ocean

relay
run

Fins ·
Rescue tube ·
Rescue board · · · · ·

Ski · · · ·

Baton · ·

Signalling ropes · · ·
Signalling flags · · · · · · · · ·
Signalling buoys · · · · · · · · ·

340
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Some specific materials differ accordingly to the distinct competi-

tive levels, while others have the same characteristics regardless the age
groups in competition. For example, the rule regarding fins area allows a
maximum of 30 x 65 cm in width and length for all categories, but these
dimensions imply a greater effort for the young lifesavers comparing to
adults, influencing their normal physical development (24). This is the
reason why some authors propose some material adaptations for the
younger age categories, like the use of smaller fins (24,25), respecting
their growing and maturation process.

Bioenergetical analysis
Open water lifesaving events do not have any defined standard times
or records since the sea, sand and wind conditions are extremely vari-
able. Moreover, there is a mix of exercise modes in some events (e.g.
running, swimming and towing) and others include significant exercise
rhythm changes. Thus, it is very hard to well define which are the main
energy systems supporting the different lifesaving events. In addition,
the environmental conditions turn very hard the use of laboratorial in-
strumentation, not allowing the variables that are typically used to asses
aerobic and anaerobic energy contributions, as the oxygen uptake and
blood lactate concentrations to be measured well (cf. 26). There are some
pioneering studies attempting to classify some parts of lifesaving sport
events based on their bioenergetical demands (e.g. 27), but they suffer
serious limitations or are insufficient, and thus they do not describe the
total course of the event.
Notwithstanding these limitations, it is possible to state that, due to
evident duration variability of the lifesaving events (from few seconds
to several minutes), all the energy pathways are relevant for lifesaving.
However, in the short-term future, the bioenergetical characterization of
each event should be done to observe the aerobic and anaerobic (lactic
and alactic) contributions. Following the literature (28-31) we hypothe-
sized that: (i) the ATP-CP contribution during short and explosive courses
lasting between ~5 - 15 s (e.g. beach flags and sprint events) will be the
most determinant; (ii) the anaerobic energy from the glycolysis would be
fundamental in events lasting between 1 - 2 min (e.g. board rescue and
rescue with tube rescue); (iii) the maximal oxygen uptake would signifi-
cantly determine performance on events taking about 4 - 8 min to end
(e.g. surf and board races); and (iv) the anaerobic threshold (the gold
standard indicator of the aerobic capacity development) would be fun-
damental at events of 20 - 40 min duration (as the ocean man/woman).

341
 sp or t science

As the energy for the lifesavers muscular work comes from both aero-
bic and anaerobic sources, it is important to emphasize that their train-
ing process should, mainly during the specific preparation stage of the
preparatory period and the competitive period, focus on the bioenerget-
ical zones that mimic those requested in competition. Considering most
of the durations in the various lifesaving events, it seems obvious that
the importance of the aerobic component, although never negligible (cf.
32), should be secondarily related to the development of the anaerobic
energetic pathway (both in its lactic and ATP-CP components).
Thus, the development of the anaerobic (or glycolytic) power and
anaerobic capacity (also known as lactic tolerance) implies the develop-
ment of (33): (i) the enzymatic activity responsible for the degradation of
glycogen to lactic acid and (ii) the muscular capacity to continue exer-
cising in adverse physiological and cellular environments (due to meta-
bolic acidosis), and the increase in muscle reserves of glycogen. Other
training areas should also be developed, namely those that are adja-
cent to those mentioned above, particularly the aerobic power (aiming
developing oxidative energy processing for muscular work) and speed
(aiming potentiating the phosphagen use and the neuromotor domain).
Training exercises focusing on aerobic capacity development, both for
raising the exercise intensity corresponding to the anaerobic threshold,
as well as to increase the effectiveness of the recovery process, should
also be implemented but mainly at lower ages and sports levels, as well
as at the general preparation stage of the preparatory period.

The lifesavers biotype

Several studies indicate the existence of different physical biotypes
regarding the type of exercise performed (e.g. 34) and, even, the spe-
cific position per sport (35,36). This kind of easy-to-conduct studies is
essential to better know the body characteristics of the participants,
and even to detect sports talents in the initial stages of practice. Obvi-
ously, lifesaving characteristics imply a specific biotype in accordance
with the competitive requirements. The first study in this field, although
using a rather heterogeneous sample and including subjects engaged in
unsystematic training process, indicated that lifesavers have a high per-
centage of body fat and, therefore, a lower percentage of lean mass (37).
Subsequently, it was verified that the meso-endomorphy was the pre-
dominant somatotype category in all lifesaving competitive age groups
(38-40).
A more recent study analysed lifesavers according to their competi-

342
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

tive specialty and verified the existence of balanced mesomorph and

meso-endomorphic somatotypes for men and women swimming pool
experts, respectively (41). It also observed relevant anthropometric dif-
ferences between open water and swimming pool specialists. More pre-
cisely, it was found that: (i) open water females are leaner and have lower
bi-acromial and bi-iliocrystal perimeters; and (ii) open water males dis-
play lower weight, height, body mass and percentage of muscle mass,
which may be more adequate to the techniques used when moving on
the board and ski. However, most studies highlight the lifesavers low
level of specialization, indicating that their main differences are related
to strength related parameters (42). Figure 7 shows the somatotypes for
male and females lifesavers.

Figure 7. Somatochart representing lifesavers somatotypes as it is described in

the specialized literature: male vs female lifesavers in lifesaving events in
general and only for open water events (left and right panels, respectively).

Conclusion
The long-time duration of the training process (no matter the sport
analysed) and the high level of competences it requires, determine that
coaches should well-know how to focus on the essentials of sportive
preparation. Speaking about sports in general, and in lifesaving, this
should be done by responding directly to each lifesaver needs and in the
most incisive, pragmatic and consequent way possible. We hope that
this text can help those involved in the sport promoting the evolution of
lifesavers both in quantitative terms (augmenting the number of people
engaged on its practice), as well as regarding the quality of the future
results.

343
 sp or t science

References
(1) Abraldes JA. Salvamento Acuático Deportivo. Un deporte solidario. La Coru-
ña: Xaniño Editorial. 2004.
(2) Pascual LF. Salvamento Acuático Profesional en piscinas y su relación con el
salvamento acuático deportivo en aguas cerradas. Tesis doctoral no publi-
cada, Universidad de Zaragoza; 2004.
(3) Abraldes JA. Formación deportiva en Salvamento Acuático. Investigación
en el Campeonato de España de Autonomías. La Coruña: Federación de
Salvamento e Socorrismo de Galicia. 2009.
(4) Pelayo P, Cocu PC. Natation: le sauvetage. EPS, Revue Education Physique et
Sport, 1992; 236: 21-23.
(5) International Life Saving Federation of Europe. The official website of the
International Life Saving Federation of Europe. 2017 [cited 2017 oct 3]; Avail-
able from: http://europe.ilsf.org/.
(6) International Life Saving Federation. The official website of the International
Life Saving Federation. 2017 [cited 2017 oct 3]; Available from: http://www.
ilsf.org.
(7) Abraldes JA. Salvamento acuático y deporte: Un estudio de los recursos hu-
manos en las playas de Galicia, intervención en los rescates y su relación
con el ámbito deportivo. Coruña: Federación de Salvamento e Socorrismo
de Galicia. 2008.
(8) Paredes T. El deporte en el Salvamento Acuático. Barcelona: Federación Es-
pañola de Salvamento y Socorrismo. Documentación no publicada; 1991.
(9) Abraldes JÁ, Fernandes RJ. Salvamento desportivo em piscina: caracteriza-
ção, iniciação desportiva, treino e investigação. In: Morouço P, Batalha N,
Fernandes RJ editors. Natação e Atividades Aquáticas: Pedagogia, Treino
e Investigação, Escola Superior de Educação e Ciências Sociais, Instituto
Politécnico de Leiria, Leiria. 2016. p. 220-242.
(10) Avramidis S, Abraldes JA, Hodgetts D. Modern Scientific Approaches of Life-
saving Sport. In: Avramidis S editor. Lifesaving Sport. History, Competition
Rules, Fair Play, Special Issues Athens, Greece: Stamoulis Publishing; 2017.
p.391-403.
(11) International Life Saving Federation. Competition rule book. Rules, Stan-
dards and Procedures for Lifesaving World Championships and ILS-sanc-
tioned Competitions. 2015-2019 Edition. Leuve, Belgium: Author; 2015.
(12) Abraldes JA. Fundamentos del Salvamento Acuático Deportivo. Murcia:
Universidad de Murcia y Open Coruseware; 2011 [cited 2017 Sept 30] Avail-
able from: http://ocw.um.es/cc.-sociales/fundamentos-del-salvamento-
acuatico-deportivo.
(13) Real Federación Española de Salvamento y Socorrismo. Reglamento de
competición. 2015-2016. Madrid: Real Federación Española de Salvamento
y Socorrismo; 2015.
(14) Portí G. Las modalidades de escalada. VI Seminario Internacional de praxi-
ología motriz. Madrid; 2001.
(15) Parlebas P. Elementos de sociología del deporte. Málaga: Unisport; 1988.
(16) Alonso JI, Abraldes JA. Análisis y clasificación del salvamento acuático de-
portivo desde su estructura y su dinámica. In: Palacios P et al. editors. 3er

344
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Congreso de Salvamento y Socorrismo de Galicia. La Coruña: Fundación

IDISSA; 2003. p.303-21.
(17) Alonso JI, Abraldes JA. Análisis de las situaciones motrices lúdicas en el sal-
vamento acuático deportivo. In: Abraldes JA, Rodríguez N editors. El salva-
mento acuático deportivo a estudio. La Coruña: Federación de Salvamento
e Socorrismo de Galicia; 2007. p.183-208.
(18) Parlebas P. Juegos, deporte y sociedad. Léxico de Praxiología Motriz. Bar-
celona: Ed. Paidotribo; 2001.
(19) Blázquez D. La iniciación deportiva y el deporte escolar. Barcelona: INDE.
1995.
(20) Hernández-Moreno J. Baloncesto: Iniciación y entrenamiento. Barcelona:
Paidotribo; 1988.
(21) Sánchez-Bañuelos F. Bases para una didáctica de la Educación Física y el
deporte. Madrid: Gymnos; 1984.
(22) Granero-Gallegos A, Gómez-López M, Rodríguez-Suárez N, Abraldes JA.
Analysis of motivational aspects in lifesaving sport. FEGUI, Revista de Sal-
vamento Acuático y Primeros Auxilios, 2012; 36(4), Supl. 1: 80-81.
(23) Dosil J. Psicología de la actividad física y del deporte. Madrid: McGraw-Hill;
2004.
(24) Abraldes JA, Rodríguez N. Posibles adaptaciones das pruebas realizadas en
aguas abiertas y desarrolladas por las categorías alevín, infantil y cadete. I
Congreso Congrés Català de Salvament Aquàtic. Barcelona. 2002 nov.
(25) Del Castill M, Abraldes JA. Estudio del salvamento acuático deportivo en
la categoría alevín. In: IV Jornadas Técnico-Profesionales de Salvamento
Acuático, Segovia; 2000.
(26) Figueiredo P, Zamparo P, Sousa A, Vilas-Boas JP, Fernandes RJ. An energy
balance of the 200 m front crawl race. European Journal of Applied Physiol-
ogy, 2011; 111: 767-77.
(27) Sousa A, Fernandes RJ, Rodríguez N, Abraldes JÁ. Influence of a 100 m
simulated in-water rescue on cardiopulmonary parameters. Prehospital
Emergency Care. 2017; 21(3): 301-308.
(28) Abraldes JA, Rodríguez N, Editors. El salvamento acuático deportivo a es-
tudio. La Coruña: Federación de Salvamento e Socorrismo de Galicia; 2007.
(29) Calvo J. Aproximación al salvamento acuático en las categorías infantil y
cadete. In: Abraldes JA, Rodríguez N editors. El salvamento acuático de-
portivo a estudio. La Coruña: Federación de Salvamento y Socorrismo de
Galicia; 2007. p. 61-112.
(30) Ferragut C, Vila H, Rodríguez N. Aspectos fisiológicos a tnere en cuenta en
el salvamento acuático deportivo. In: Abraldes JA, Rodríguez N editors. El
salvamento acuático deportivo a estudio. La Coruña: Federación de Salva-
mento e Socorrismo de Galicia. 2007. p.161-82.
(31) Heredia, JM., Aparicio, VA. y López, M. Evaluación y cuantificación de las
cargas de entrenamiento e de los deportistas de salvamento y socorrismo
mediante el uso de tecnología GPS. In: Abraldes JA y Rodríguez N editors.
Estudios en torno a una necesidad: Salvamento y Socorrismo Acuático. La
Coruña: Federación de Salvamento e Socorrismo de Galicia, 2008, p. 109-
117.
(32) Gastin PB. Energy system interaction and relative contribution during max-

345
 sp or t science

imal exercise. Sports Medicine, 2001; 31(10): 725-41.

(33) Olbrecht J. The Science of winning: planning, periodizing and optimizing
swim training. Luton: Swimshop; 2000.
(34) Malina RM, Bouchard C, Bard-Or O. Growth, maturation and physical activ-
ity. Illinois: Human Kinetics; 2004.
(35) Sands WA, Smith LS, Kivi DM, McNeal JR, Dorman JC, Stone MH. Anthro-
pometric and physical abilities profiles: US National Skeleton Team. Sport
Biomechics, 2005; 4(2):197-214.
(36) Canossa S, Abraldes JA, Soares S, Vila MH, Ferragut C, Rodríguez N, Ar-
gudo FM, Fernandes RF, Garganta JM. Contrasting morphology and train-
ing background in waterpolo teams of different competitive levels. Motriz.
Journal of Physical Education, 2014; 3:272-79.
(37) Agrasar C, Abraldes JA, Rodríguez N, Villar JM. Estudio cineantropomé-
trico, de la composición corporal y el somatotipo dos componentes de la
selección gallega de salvamento acuático deportivo. In: Palacios P et al.
editors. 3er Congreso de Salvamento y Socorrismo de Galicia. La Coruña:
Fundación IDISSA; 2003. p.263-69.
(38) Ferragut C, Rodríguez N, Vila H, Alacid F, Abraldes JA. Study of the biotype
and the body composition of the gallega male in lifesaving. In: Abraldes JA,
Rodríguez N editors. Book of abstracts International Lifesaving Congress
2007. La Coruña: Lifesaving Federation of Galicia; 2007. p.77-78.
(39) Rodríguez N, Vila MH, Ferragut C, Abraldes JA. Perfil antropométrico dos
deportistas infantiles de salvamento acuático deportivo. In: Martínez MJ,
García JL, Arufe V editors. III Congreso Nacional de Ciencias del Deporte.
Barcelona: ACUGA y Alto Rendimiento; 2007. p.1-9.
(40) Rodríguez N, Abraldes JA, Ferragut C, Vila H. Salvamento acuático deport-
ivo. Un estudio de la población masculina de Galicia. In: Arufe V, García JL,
Lera A editors. V Congreso Nacional de Ciencias del Deporte y Educación
Física. Pontevedra: Alto Rendimiento; 2009. p.1-10.
(41) Abraldes JA, Rodríguez N, Ferragut F, Vila MH. Características antropomé-
tricas, composición corporal y somatotipo en deportistas de élite de salva-
mento. RETOS. Nuevas tendencias en Educación Física, Deporte y Recre-
ación. 2014; 26(2), 66-70.
(42) Vila H, Ferragut C, Rodríguez N, Abraldes JA. Valoración cineantropomé-
trica dos deportistas de la selección gallega de salvamento deportivo. In:
Arufe V, García JL, Domínguez A editors. II Congreso Internacional de Cien-
cias del Deporte. Pontevedra: Alto Rendimiento; 2008. p.1-10.

346
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

The importance of perceptual-cognitive skills

in water polo players
Filipe Casanova1,2, Fernando Tavares1,2, Sofia Canossa1, Maickel
Bach Padilha1,2 and Ricardo J. Fernandes1,3

Introduction
Research on sport expertise is a fruitful domain to explore the va-
lidity of models developed in other fields. This provides a rich source of
empirical evidence on the true potential of human achievement, reason
why it has been defined as the ability to consistently demonstrate su-
perior athletic performance (1, 2, 3). This research area points out that
many elements of the perceptual-cognitive skills, decision-making and
motor skill execution strongly influence sport performance, especially in
tasks in which individuals are required to perform under strict temporal
and spatial constraints (e.g. 4, 5, 6).
During invasion team sports, players need the ability to “read” the
opponents intentions and perform correct techniques (e.g. to overcome
opponents and score goals) aiming to achieve game purposes and win
the match. Although water polo is a team sport, little empirical atten-
tion has been employed by researchers (7, 8, 9), probably because it is
played in the water environment, non-using the habitual bipedal human
references. Thus, studying water polo players decisions may contribute
for a better understanding of their expertise, especially the mechanisms
underpinning the decision-making and perceptual-cognitive skills. Dif-
ferences on superior athletic performance between elite and non-elite
players are typically apparent through observation, but the underlying
perceptual and cognitive processes contributing to an advantage in an-
ticipatory behavior are less evident.
Perceptual-cognitive skills refers to the ability to identify and acquire
environmental information for integration with existing knowledge such
that appropriate responses can be selected and executed (10). There-
fore, the process of selective attention and the need to invoke a detailed

1 Centre of Research, Education, Innovation and Intervention in Sport, Faculty

of Sport, University of Porto, Porto, Portugal
2 Center of Studies of Sports Games, Faculty of Sport, University of Porto, Por-
to, Portugal
3 Porto Biomechanics Laboratory, University of Porto, Porto, Portugal

347
 sp or t science

role for knowledge structures stored in memory are considered essential

to help guide the search for (and effective processing of) task specific in-
formation. Helping to contextualize the purpose of perceptual-cognitive
research, a simple information-processing illustration model involving
the main components in anticipation and decision-making is proposed
(Figure 1; 11).

Figure 1. Brief information-processing based model of sport anticipation and

decision-making skill (adapted from Williams & Ward, 2007).

Knowing the underpinning factors of elite sport performers’ enhance-

ment can help highlighting the key determinants of an effective practice,
its instruction and social support networks required to performance
facilitation and learning in other domains. Elite players pertinently se-
lect and accurately interpret environmental cues (i.e. perceptual com-
ponent) that, combined with their faster selection of an appropriate
response (i.e. decision component), allow performing a smoother and
more efficient movement (i.e. motor component). Furthermore, players
skilled at decision-making depend on elaborate perceptual and cogni-
tive skills to effectively interpret complex information and formulate an
appropriate action plan (12). These findings confirm previous studies in
which it is stated that a distinguishing feature of experts is their adept-
ness on both “knowing what to do” and “doing it”, while less skilled ath-
letes may achieve some success degree with one or the other mentioned
capabilities, unable to “link” both (13, 14).

348
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Situational or strategic sports require players’ fast and accurate de-

cisions in a complex and variable environment (15, 8). These decisions
occur upon information coming from different sources (e.g. ball, team-
mates and opponents) and space (near and far), with decision-making
process occurring under pressure with opponents restricting the time
and space available. In this context, the interaction dynamics between
athlete and sport environment are based on the presupposition of stimu-
lus reception from which the player emits an answer (i.e. action-reac-
tion). Thus, players should focus their attention on the most crucial and
relevant information sources to carry out their performances efficiently
and successfully (16, 17).

Expert perceptual-cognitive skills

With the premise of being essential for effective anticipation and
decision-making processes in sports, some researchers have highlight-
ed superiority of skilled performers over less skilled and novice play-
ers when examined a number of perceptual-cognitive skills separately.
These skills include advance visual cue utilization, pattern recall and rec-
ognition, visual search behavior and the knowledge of situational pro-
babilities. In lay terms, it is often referred to as “game intelligence” (18).

Advance visual cue utilization

Advance visual cue use refers to a player’s ability to predict accurate-
ly, based upon opponent’s posture and bodily orientation, previously to
a key event (19, 20). This perceptual skill is also being considered as the
key factor of fast-ball sports performance because players need to per-
ceive relevant sources under time-constrains faced throughout the game
(21). On this matter, a research was conducted where experienced and
inexperienced players were required to observe near “life-size” filmed
sequences of five different players taking penalty kicks during prepara-
tory stance, approach run and kicking (22). The requirement was to in-
dicate which of the four corners of the goal the ball was to be directed,
prior to temporal occlusion. Results showed that experienced players
exhibited better performance only under the shortest durations (that is,
pre-event or pre-contact occlusion conditions), which is in agreement
with other studies results (e.g. 23, 24).
Only few researchers have attempted to identify the underlying
mechanisms or even the specific perceptual information supporting
the identification process that guides skillful action. This issue is usu-
ally addressed by combining the temporal occlusion approach with spa-

349
 sp or t science

tial occlusion, eye movement registration and verbal report techniques

(e.g. 21, 5). In the event occlusion approach, the presumption is that if
there is a decrement in performance on the trial when a cue is occluded
compared to a full vision control condition, then the importance of the
occluded source of information is highlighted. However, such system-
atic programs of research and attempts to cross-validate findings, and
to extend knowledge by combining different measures, are rare in the
literature.
Although this argument could not be considered, researchers have
recently argued that performers are more likely to extract global mo-
tion-related information from an opponent’s postural orientation than
a specific information cue. The suggestion is that skilled performers use
the relative motion between joints and/or limbs to guide successful per-
formance rather than a specific cue(s) (25), which requires researchers
to convert players action video images into point-light displays. These
displays capture the major joint centers of body motion, which are then
displayed as light points against a black background. The aim of using
this technique is to remove background and contextual information pre-
senting movement in its simplest terms (26). Moreover, contemporary
methods of creating point-light (or stick figure) images using optoelec-
tronic motion capture systems rather than video provides significant ad-
vantages in this regard (cf. 27, 28, 29).
Several researchers have suggested that: (i) both novice and skilled
tennis players, for example, are leaned to change their used information
when moving from normal to point-light conditions, but skilled players
are much less affected than are their counterparts (30); (ii) when exe-
cuting a technical skill, such as controlling a ball, the best skilled play-
ers are able to use several potential sources of sensory information (e.g.
vision and proprioception) in an interchangeable manner to facilitate
effective performance (31); (iii) it is possible that in certain situations
skilled performers may decide not to use these cues during matches due
the possible energetic cost associated with anticipation may result in
performers adopting a ‘wait-and-see’ approach (32). Particularly in wa-
ter polo, the effects of fatigue on decision making and players shooting
skill have been studied by using film-based “temporal occlusion para-
digm”, during goal shooting skill test (8). Results showed that decision-
making accuracy was superior under very high exertion conditions than
lower exertion except for rest.

350
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Pattern recall and recognition

Researchers have made extensive use of the recall paradigm to assess
experts’ cognitive advantage degree over lesser skilled performers. The
recall paradigm comprises both static and dynamic images, portraying
either a structured or unstructured task-specific display where the par-
ticipant is required to recall the location of each player (33). Performance
is then ascertained as the level of agreement between priori-identified
features in the actual display (e.g. player positions) and the participant’s
recall of those features (34).
Another methodological approach termed recognition has been used
to identify players’ ability to recognize whether participants have pre-
viously viewed the action sequences in an earlier viewing phase. Par-
ticipants’ task is to quickly and accurately indicate those clips that they
have or have not seen before. The literature reported that experienced
players recognized previously viewed structured video clips more accu-
rately and, consequently, were able to perceive an evolving play pattern
much earlier in its development than their less experienced counter-
parts (35). Once again, skilled players demonstrated superior recogni-
tion skill when compared with their counterparts (34, 36, 37). If players
can encode specific information to a deeper and more conceptual level,
they can anticipate their opponents’ intentions and plan ahead as to the
most appropriate course of action.
Currently, researchers are attempting to identify the underlying
mechanisms that differentiate skilled from less skilled participants.
Using point-light displays, it was possible to show that skilled players
maintain their superiority over less skilled players in pattern recognition
performance even when players are presented as moving dots of light
against a black background (6). This finding suggests that skilled play-
ers are more attuned than their counterparts to the relative motions be-
tween players and/or the higher-order relational information conveyed
by such motions.

Visual search behavior

Visual search strategy is defined as the ability to pick up advanced
visual cues or to identify playing patterns (19, 38). Eyes search the dis-
play or scene attempting to extract main pertinent information guiding
the performers’ action so that appropriate visual attention allocation
precedes and determines effective motor behavior. An eye movement
registration system has been used to assess visual search behavior by
recording a performer’s eye movements and interspersed fixations (cf.

351
 sp or t science

39). Each fixation length is presumed to represent a cognitive process-

ing degree, whereas the point-of-gaze is assumed to be representative of
the most pertinent cues extracted from the environment, facilitating the
decision-making process (this index is obtained by the number of visual
fixations during a given period).
However, it is noticed that corresponding movements of ≤ 5º are of-
ten considered noise and statistically removed from fixation duration
calculation, which typically ranges from 150 up to 600 ms (39). Research-
ers have recorded short (100 ms) and long fixations (1.500 ms) with cor-
responding movements of ≤ 1º (40). Eye movements between succes-
sive fixations, known as saccades, are believed to suppress information
processing. Main research findings suggest that experts focus their gaze
on more information areas of the display compared to novices, enabling
them to more effectively anticipate action requirements (cf. 40, 42).
One of the earliest studies to examine the importance of visual behav-
ior (43, 44), investigated expert and novice soccer players search patterns
faced with offensive simulations requiring tactical decision-making (e.g.
microstate situations – 3 vs 3, 4 vs 4 – and “set-play” conditions – free-
kicks). Findings lead to the following important considerations, namely:
expert players have faster movement initiation, ball-contact and total
response times, and are more accurate in their decisions, as well as ex-
pert players’ better performance is attributed to an enhanced ability to
recognize structure and redundancy within the display, resulting in more
efficient use of available search time (this was supported by eye-move-
ment data that showed expert visual search patterns to be economical,
with fewer fixations of longer duration on selected areas of the display).
Moreover, experts are more interested in the position of the “swe-
eper” and any potential areas of “free” space, whereas novice players
search information from less sophisticated sources such as other atta-
ckers, the goal and the ball. Some of these results were corroborated by
other authors (cf. 40, 45) and even when athletes’ visual behavior is con-
strained by several factors (such as task nature, performers’ physical and
emotional levels and the environment) experts scan the display more
effectively and efficiently than their counterparts (16, 46, 42).
Maintaining gaze for an extended period of time (the so-called quiet
eye period) might be the key factor in self-paced tasks where the accu-
racy of aiming is important. Specifically, the quiet eye period represents
the elapsed time between the last visual fixation on a target and motor
response initiation (47), being explored in archery (48), soccer, (49) and
hockey (50). Nevertheless, despite some advantages recognizing its use

352
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

on sport performance, some restrictions were pointed out when it aims

dynamic situations (cf. 51), as the requirement to maintain an extended
quiet-eye period prior to response initiation (which is likely to interact
with the need to monitor the positions and movements of teammates
and opponents) and to execute the required action prior to being chal-
lenged by an opponent (52).
So, there is an evidence suggesting that sport performers often use
peripheral and central vision to extract relevant information from the
display. Several researchers have noted that experts are more inclined
to fixate gaze centrally in an attempt to pick up an opponent’s relative
motion profile using peripheral vision (23). Moreover, experts are able to
anticipate an opponent’s intended shot direction by fixating on relative-
ly deterministic and proximal postural cues (such as trunk/hip rotation)
before using more distal cues (e.g. racket; 6).

Knowledge of situational probabilities

This perceptual-cognitive skill has been defined as expert perform-
ers ability to extract meaningful contextual information from the event
outcomes, where scientific evidence suggested that experts are more ac-
curate to create expectations than novices on probable event to occur in
any given scenario (53). Situational probabilities importance and their
relationship with decision-making behavior in squash, tennis, badmin-
ton and racquetball were already examined (54, 55). In which, the results
showed that players evaluated the possible event occurrences and then
they used this information to maximize subsequent behavior efficiency.
Players’ initial anticipatory movements were guided by their expecta-
tions, with subsequent corrective or confirmatory movements being
performed based on current information or contextual cues.
Moreover, Ward and Williams (56) tried to investigate elite and sub-
elite players’ requirements to predict and rank the “best passing options”
available to a player in ball possession. The results founded suggest that
elite players were better than their sub-elite counterparts at identifying
players in the best position to receive the ball, and more accurate in as-
signing an appropriate probability to players in threatening and non-
threatening positions. Skilled players were also better at hedging their
bets, judiciously determining the importance of each potential option
presented, effectively priming the search for new information and ensu-
ring that the most pertinent contextual information was extracted from
each area of the display (56).
In an attempt to clarify event probabilities in sports domain, task

353
 sp or t science

specificity and participant skill level, Williams (19) suggested (i) gene-
ral event probability, that refers to the likelihood that opponents will
typically act in a certain way given the context in question, such as the
typical center-backs options when in ball possession in the middle of
the field of play, the typical drive-in offensive style or the center forward
play offensive predominance; and (ii) specific probabilities relate to op-
ponents’ tendencies knowledge, for example, a particular player may
systematically attempt to finish offensive plays through 5m shooting ac-
tion or a given center forward may always seek the back-hand shot when
receiving the ball (57).

Conclusions
Although there is substantial work in the field of expertise (as previ-
ously reported), it would be of interest in future research: (i) to clarify the
importance of mechanisms underlying perceptual-cognitive expertise in
water polo players; (ii) to highlight the influence imposed by several con-
straints on the expert’s performance in a near or realistic context; and
(iii) to integrate simultaneously in the same research different measures
of the perceptual-cognitive skills, constraints imposed by the task, the
environment and the individual characteristics of the performer.

References
1. Starkes, J. L. (1993). Motor experts: Opening thoughts. In: J. L. Starkes & F.
Allard (Eds.). Cognitive issues in motor expertise. Amsterdam: Elsevier Sci-
ence, 3-16.
2. Ericsson, K. A., & Lehmann, A. C. (1996). Expert and exceptional performance:
Evidence of maximal adaptation to task constraints. Annual Review of Psy-
chology, 47, 273-305.
3. Ericsson, K. A. (2003). The development of elite performance and deliberate
practice: An update from the perspective of the expert-performance ap-
proach. In: J. Starkes & K. A. Ericsson (Eds.). Expert performance in sport:
Recent advances in research on sport expertise. Champaign, IL: Human Ki-
netics, 49-81.
4. Abernethy, B. (1999). The 1997 Coleman Griffith address: A juncture between
psychological theory and practice. Journal of Applied Sport Psychology,
11(1), 126-141.
5. Williams, A. M., & Davids, K. (1998). Visual search strategy, selective attention,
and expertise in soccer. Research Quarterly for Exercise and Sport, 69(2),
111-128.
6. Williams, A. M., Hodges, N. J., & Barton, G. (2006). Identifying patterns of play
in dynamic sport tasks: The essential information underlying skilled perfor-
mance. Perception, 35, 317-332.
7. Efthimios, K., Thomas, K., Maria, M., & Vasiliki, D. (1998). Cognitive, perceptual,
and motor abilities in skilled basketball performance. Perceptual and Motor

354
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

Skills, 86 (3, Pt 1), 899-912.

8. Royal, K. A., Farrow, D., Mujika, I., Halson, S. L., Pyne, D., & Abernethy, B. (2006).
The effects of fatigue on decision-making and shooting skill performance in
water polo players. Journal of Sports Sciences, 24(8), 807-815.
9. Escalante, Y., Saavedra, J. M., Mansilla, M., & Tella, V. (2011) Discriminatory
power of water polo game-related statistics at the 2008 Olympic Games,
Journal of Sports Sciences, 29(3), 291-298.
10. Marteniuk, R. G. (1976). Information processing in motor skills. New York:
Holt, Rinehart & Winston.
11. Williams, A. M., & Ward, P. (2007). Perceptual-cognitive expertise in sport: ex-
ploring new horizons. In G. Tenenbaum & R. C. Eklund (Eds.), Handbook of
sport psychology. New York: Wiley, 203-223.
12. Williams, A. M. (2009). Perceiving the intentions of others: how do skilled per-
formers make anticipation judgments? In R. Markus, J. Joseph, & H. Hauke
(Eds.), Progress in Brain Research, vol 174, Mind and Motion: The Bidirec-
tional Link between Thought and Action. The Netherlands: Elsevier, 73-83.
13. McPherson, S. L., & Thomas, J. R. (1989). Relation of knowledge and perfor-
mance in boys’ tennis: Age and expertise. Journal of Experimental Child
Psychology, 48, 190-211.
14. Allard, F., & Starkes, J. L. (1991). Motor-skill experts in sports, dance, and other
domains. In: K. A. Ericsson & J. Smith (Eds.) Toward a general theory of ex-
pertise: Prospects and limits. Cambridge: Cambridge University Press, 126-
152.
15. Meusen, R. (2002). Fatigue during game play: a review of central nervous sys-
tem aspects during exercise. In: W. Spinks, T. Reilly, & A. Murphy (Eds.). Sci-
ence and Football IV. Routledge, London, 304-307.
16. Williams, A. M., Davids, K., Burwitz, L., & Williams, J. G. (1994). Visual search
strategies of experienced and inexperienced soccer players. Research Quar-
terly for Exercise and Sport, 5(2), 127-135.
17. Roca, A., Ford, P. R., McRobert, A. P., & Williams, A. M. (2013). Perceptual-cog-
nitive skills and their interaction as a function of task constraints in soccer.
Journal of Sport & Exercise Psychology, 35, 144-155.
18. Stratton, G., Reilly, T., Richardson, D., & Williams, A. M. (2004). Youth soccer:
From science to performance. London: Routledge.
19. Williams, A. M. (2000). Perceptual skill in soccer: Implications for talent identi-
fication and development. Journal of Sports Sciences, 18, 737-750.
20. Abernethy, B., & Zawi, K. (2007). Pickup of essential kinematics underpins
expert perception of movement patterns. Journal of Motor Behavior, 39,
353-367.
21. Abernethy, B., & Russell, D. G. (1987). Expert-novice differences in an applied
selective attention task. Journal of Sport Psychology, 9, 326-345.
22. Williams, A. M., & Burwitz, L. (1993). Advance cue utilization in soccer. In: T.
Reilly, J. Clarys & A. Stibbe (Eds.). Science and Football II. London: E & FN
Spon, 239-244.
23. Williams, A. M., & Reilly, T. P. (2000). Talent identification and development in
soccer (Special issue). Journal of Sports Sciences, 18.
24. Starkes, J. L., Helsen, W. F., & Jack, R. (2001). Expert performance in sport and
dance. In: R. N. Singer, H. A. Hausenblas & C. M. Janelle (Eds.). Handbook of

355
 sp or t science

Sport Psychology. New York: Wiley, 174-201.

25. Lavalle, D., Kremer, J., Moran, A. P., & Williams, A. M. (2004). Sport Expertise.
In: D. Lavalle, J. Kremer, A. P. Moran, & A. M. Williams (Eds.). Sport Psychol-
ogy: Contemporary Themes. New York: Palgrave Macmillan, 139-158.
26. Cutting, J. E., & Proffitt, D. R. (1982). The minimum principle and the percep-
tion of absolute, common, and relative motion. Cognitive Psychology, 14,
211-246.
27. Hodges, N. J., Hayes, S., Breslin, G., & Williams, A. M. (2005). An evaluation
of the minimal constraining information during movement observation and
reproduction. Acta Psychologica, 119(3), 264-282.
28. Williams, A. M., & Ericsson, K. A. (2005). Perceptual-cognitive expertise in
sport: Some considerations when applying the expert performance ap-
proach. Human Movement Science, 24 (3), 283-307.
29. Cañal-Bruland, R., Huys, R., Hagemann, N., & Williams, A. M. (2006). The ef-
fects of occlusion, neutralization, and deception of perceptual information
on anticipation in tennis. Journal of Sport and Exercise Psychology, 28, S44.
30. Ward, P., Williams, A. M., & Bennett, S. J. (2002). Visual search and biologi-
cal motion perception in tennis. Research Quarterly for Exercise and Sport,
73(1), 107-112.
31. Williams, A. M., Harris, M., Weigelt, C., & Scott, M. A. (2002). Age related dif-
ferences in vision and proprioception during a lower limb interceptive task:
The effects of skill and practice. Research Quarterly for Exercise and Sport,
73(4), 386-395.
32. James, N., Caudrelier, T., & Murray, S. (2005). The use of anticipation by elite
squash players. Journal of Sports Sciences, 23(11/12), 1249-1250.
33. North, J. S., Williams, A. M., Hodges, N., Ward, P., & Ericsson, K. A. (2009). Per-
ceiving patterns in dynamic action sequences: Investigating the processes
underpinning stimulus recognition and anticipation skill. Applied Cognitive
Psychology, 23, 878-894.
34. Williams, A. M., & Davids, K. (1995). Declarative knowledge in sport: a by-
product of experience or a characteristic of expertise? Journal of Sport and
Exercise Psychology, 17(3), 259-275.
35. Williams, A. M., Davids, K., Burwitz, L., & Williams, J. G. (1993). Cognitive
knowledge and soccer performance. Perceptual and Motor Skills, 76, 579-
593.
36. Smeeton, N. J., Ward, P., & Williams, A. M. (2004). Transfer of perceptual skill
in sport. Journal of Sports Sciences, 19(2), 3-9.
37. Abernethy, B., Baker, J., & Côté, J. (2005). Transfer of pattern recall skills may
contribute to the development of sport expertise. Applied Cognitive Psy-
chology, 19, 1-14.
38. Henderson, J. M. (2003). Human gaze control during real-world scene per-
ception. Trends in Cognitive Sciences, 7(11), 498-504.
39. Irwin, D. E. (1992). Visual memory within and across fixations. In: K. Rayner
(Ed.). Eye movements and visual cognition: Scene perception and reading.
New York: Springer-Verlag, 146-165.
40. Williams, A. M., Davids, K., & Williams, J.G. (1999). Visual Perception and Ac-
tion in Sport. London: E & FN Spon.
41. Vaeyens, R., Lenoir, M., Williams, A. M., & Philippaerts, R. M. (2007). Mecha-

356
 curren t a nd fu t ure t rends for perform a nce op t imiz at ion

nisms underpinning successful decision making in skilled youth soccer play-

ers: An analysis of visual search behaviors. Journal of Motor Behavior, 39(5),
395-408.
42. Casanova, F., Garganta, J., Silva, G., Alves, A., Oliveira, J., & Williams, A. M.
(2013). Effects of prolonged intermittent exercise on perceptual-cognitive
processes. Medicine & Science in Sports & Exercise, 45(8), 1610-1617.
43. Helsen, W. F., & Pauwels, J. M. (1993a). A cognitive approach to visual search
in sport. In: D. Brogan & K. Carr (Eds.). Visual search II. London: Taylor & Fran-
cis, 177-184.
44. Helsen, W. F., & Pauwels, J. M. (1993b). The relationship between expertise
and visual information processing in sport. In: J. L. Starkes & F. Allard (Eds.).
Cognitive issues in motor expertise. Amsterdam: Elsevier, 109-134.
45. Williams, A. M. (2002). Visual search behaviour in sport. (Editorial). Journal of
Sports Sciences, 20(3), 169-170.
46. Vickers, J. N., & Williams, A. M. (2007). Performing under pressure: the effects
of physiological arousal, cognitive anxiety, and gaze control in Biathlon.
Journal of Motor Behavior, 39(5), 381-394.
47. Vickers, J. N. (1996). Visual control while aiming at a far target. Journal of Ex-
perimental Psychology: Human Perception and Performance, 22, 342-354.
48. Behan, M., & Wilson, M. (2008). State anxiety and visual attention: The role
of the quiet eye period in aiming to a far target. Journal of Sports Sciences,
26, 207-215.
49. Piras, A., & Vickers, J. N. (2011). The effect of fixation transitions on quiet eye
duration and performance in the soccer penalty kick: instep versus inside
kicks. Cognitive Processing, 12, 245-255.
50. Panchuk, D., Vickers, J. N., & Hopkins, W. G. (2016). Quiet eye predicts goal-
tender success in deflected ice hockey shots. European Journal of Sport Sci-
ence, doi:10.1080/17461391.2016.1156160
51. Singer, R. N. (2000). Performance and human factors: Considerations about
cognition and attention for self-paced and externally-paced events. Ergo-
nomics, 43, 1661-1680.
52. Martell, S. G., & Vickers, J. N. (2004). Gaze characteristics of elite and near-
elite athletes in ice hockey defensive tactics. Human Movement Science,
22(6), 689-712.
53. Farrow, D., & Reid, M. (2012). The contribution of situational probability in-
formation to anticipatory skill. Journal of Science and Medicine in Sport,
15, 368-373.
54. Alain, C., & Girardin, Y. (1978). The use of uncertainty in racquetball competi-
tion. Canadian Journal of Applied Sport Sciences, 3, 240-243.
55. Alain, C., & Proteau, L. (1980). Decision making in sport. In: C. H. Nadeau, W. R.
Halliwell, K. M. Newell & G. C. Roberts (Eds.). Psychology of motor behavior
and sport. Champaign, IL: Human Kinetics, 465-477.
56. Ward, P., & Williams, A. M. (2003). Perceptual and cognitive skill development
in soccer: the multidimensional nature of expert performance. Journal of
Sport and Exercise Psychology, 25(1), 93-111.
57. Canossa, S., Abraldes, J.A., Soares, S. M., Fernandes, R.J. and Garganta, J.M.
(2016), Vertical jump and shot speed, efficacy and accuracy in water polo,
International Journal of Performance Analysis in Sport, 16, 64-79.

357
View publication stats