Journal of Exercise Science & Fitness 19 (2021) 143e149

Contents lists available at ScienceDirect

Journal of Exercise Science & Fitness

journal homepage: www.elsevier.com/locate/jesf

Short and long-term differences in anthropometric characteristics and

physical performance between male rugby players that became
professional or remained amateur
Michael J. Hamlin a, b, *, Richard W. Deuchrass c, Catherine E. Elliot a,
Nuttaset Manimmanakorn d
a
Department of Tourism, Sport and Society, Lincoln University, Christchurch, New Zealand
b
Faculty of Health and Environmental Science, Sports Performance Research Institute New Zealand (SPRINZ), Auckland University of Technology, Auckland,
New Zealand
c
Recreation Centre, Lincoln University, Christchurch, New Zealand
d
Department of Rehabilitation Medicine, Faculty of Medicine, Khon Kaen University, Khon Kaen Thailand

a r t i c l e i n f o a b s t r a c t

Article history: Objective: The objective of this study was to investigate which anthropometric and physical performance
Received 20 October 2020 variables characterised players that advanced to professional teams (professionals) and how these var-
Received in revised form iables changed over time, compared to those that did not secure professional contracts (i.e. remained
22 January 2021
amateurs).
Accepted 23 January 2021
Methods: Differences in anthropometry, strength, speed, power and intermittent running ability in 83
Available online 31 January 2021
male rugby players collected between 2015 and 2019 were determined using repeated measures
analysis.
Keywords:
Body composition
Results: When arriving for the first year of the program, forwards that went on to become professional
Developmental players players were older (0.4 ± 0.3 yr, mean ± 95% CI, p ¼ 0.004), heavier (4.6 ± 2.5 kg, p < 0.001) and stronger
Rugby union (range 6.2e16.4%) than forwards that remained amateur. Professional forwards were also slower at
Speed sprinting (range 2.7e2.9%, p < 0.001) and had lower Yo-Yo IRT L1 (10.8%, p ¼ 0.03). When first arrived
Strength on the program, professional backs were taller (3.5 ± 1.8 cm, p < 0.001), heavier (4.6 ± 2.4 kg, p < 0.001)
Yo-Yo intermittent recovery Test and faster over 20 m (1.9 ± 1.7%, p ¼ 0.03) and 30 m (1.7 ± 1.6%, p ¼ 0.04) compared to amateurs.
Compared to amateurs, professionals had a smaller increase in body mass (4.2 ± 2.0%, p < 0.001) and
greater improvement in sprinting (3.7, 2.8, 2.8% over 10, 20 and 30-m, respectively) and Yo-Yo IRT L1
(14.7 ± 11.0%, p ¼ 0.05) over 3 years training.
Conclusion: Characteristics that are likely to assist players in becoming professionals include being older,
heavier, taller and stronger.
© 2021 The Society of Chinese Scholars on Exercise Physiology and Fitness. Published by Elsevier
(Singapore) Pte Ltd. This is an open access article under the CC BY license (http://creativecommons.org/
licenses/by/4.0/).

Introduction and dynamic exertions.1,2 Since rugby union turned professional in

1995, research has sought to understand, among other things, what
Rugby union is a field-based team sport that requires players to movement and performance characteristics are required of the
endure a large number of high-impact collisions along with modern rugby union player to reach professional status.3 Results
numerous maximal sprints, high speed running and other static from such research has informed strength and conditioning staff on
position-specific characteristics required of rugby players including
speed, strength, power, aerobic ability and anthropometry. For
example, forwards are mainly involved in situations that require
* Corresponding author. Department of Tourism, Sport and Society, P O Box
greater body mass, strength and power like tackling, scrummaging
85085, Lincoln University, Lincoln, 7647, Canterbury, New Zealand.
E-mail addresses: [email protected] (M.J. Hamlin), Richard.Deuchrass@ and mauling.4,5 On the other hand, backs are involved in high-
lincoln.ac.nz (R.W. Deuchrass), [email protected] (C.E. Elliot), natman@ speed running, evasion, and movements that require agility and
kku.ac.th (N. Manimmanakorn).

https://doi.org/10.1016/j.jesf.2021.01.002
1728-869X/© 2021 The Society of Chinese Scholars on Exercise Physiology and Fitness. Published by Elsevier (Singapore) Pte Ltd. This is an open access article under the CC
BY license (http://creativecommons.org/licenses/by/4.0/).
M.J. Hamlin, R.W. Deuchrass, C.E. Elliot et al. Journal of Exercise Science & Fitness 19 (2021) 143e149

therefore require a relatively lower body mass, but high levels of Participants
power and speed.2,6
The aim of any training programme is to efficiently and effec- Anthropometric and performance measures of 83 male rugby
tively improve player’s performance in a sports-specific manner players (mean ± SD for age ¼ 18.9 ± 1.3 yr, range ¼ 17.6e26.8 yr)
without incurring illness or injury. By increasing the player’s per- during their time at university between 2015 and 2019 were
formance ability, many professional training programmes (i.e. recorded. Players were involved in a university sport scholarship
Institut National du Football, for footballers or the International program where they received nutritional, psychological, and
Rugby Academy of New Zealand, for rugby union players) hope to medical advice along with individualized strength and condition-
progress the player along the performance pathway from amateur ing training. All participants were young rugby players selected
to professional athlete. To help with program prescription, players from age-group provincial or national representative honours (high
complete regular physical assessments which tend to be used in the level amateur leagues). The study was approved by the Lincoln
short-term to evaluate mesocycles or short conditioning phases. University Human Ethics Committee (Approval Reference No.
However few studies on rugby union specifically have documented 2018e01). All subjects were informed of the benefits and risks of
the long-term anthropometrical and performance changes such the investigation prior to signing an institutionally approved
developmental programmes achieve. Identifying anthropometrical informed consent document to participate in the study. The dataset
and physical performance characteristics that can distinguish be- incorporates measures from February through to October for each
tween players of different abilities in different positions are year, therefore includes the major competition season for rugby
important for the best training regimens, performance outputs, for players in the southern hemisphere.
recruitment, and for player development.7,8 For example, a number
of studies have shown that players with superior speed, strength, Training
aerobic endurance and repeated sprint ability have better chances
of being selected onto a professional team.9e11 Similarly in rugby Individualized training programs were developed by the
league, Till and colleagues, in a number of retrospective studies on strength and conditioning staff at the university for each player,
junior rugby league players, found significant differences in depending on the player’s strengths and weaknesses, playing po-
anthropometric and performance variables in 13e15 year olds that sition, time of year (in-season vs. out of season) and injury status. In
subsequently went on to gain professional contracts, compared to most weeks, players would have at least three training sessions,
others that remained amateur.12e14 one sport-specific skills session, 2 rugby club trainings and one
While research exists on the differences between professional practice game or competition (approximately 9 h) per week.
and amateur rugby union athletes, this research is now 7 years old,
and with the continual evolution of the game, along with changes Testing
in the physical demands required of its athletes,15 these research
findings may be outdated. In addition, this previous research was Performance testing was completed by the same strength and
conducted on players already at the professional level.16 Conse- conditioning trainers and the results were entered into the Metrifit
quently, there is a need for a study to monitor the long-term database after each test. All testing procedures were standardised
anthropometric and performance measures including strength, using each specific test’s documented procedures (see relevant
power, speed and aerobic endurance in developmental rugby references in this section), carried out at the same time of the day,
players that at the time of testing had not reached the professional under similar environmental conditions (e.g. strength measured in
level. The usefulness of these anthropometric and performance the lifting gym, Yo-Yo IRT L1 and speed measured on a suitable-
measures in the development of a rugby player from amateur to sized indoor artificial turf). Familiarization trials were not
professional can then be elucidated and used to guide training of required as all players were familiar with the testing protocols and
such players. had been tested numerous times previously. Similar to previous
Therefore, the purpose of this study was to investigate which research17 individual tests were not systematically different be-
anthropometric and physical performance variables characterised tween trials with a between-test variability (typical error) of <10%.
players that went on to be selected for professional teams and how Data on intermittent running ability (Yo-Yo IRT L1), anthropometry,
these variables changed over 3 consecutive years before selection upper and lower body strength, speed and power from players from
was made. the start of 2015 to the end of 2019 was downloaded from the
database. For each year, players were tested approximately every
2e3 months from March to October.

Methods Anthropometry

This retrospective longitudinal study used a commercially Body mass, reported in kg (to 1 decimal point) was measured on
available software system (Health and Sport Technologies, Ltd., calibrated scales (Seca, model 762, Germany) with the players
trading as Metrifit, Millgrange, Greenore,Co., Louth, Ireland) to shoes and socks removed and in light training clothing. The sum of
collect testing data on players during their time at university. The 8 skinfolds (bicep, triceps, subscapular, abdominal, supraspinale,
data was uploaded into the Metrifit system by the strength and iliac crest, front thigh and medial calf) were taken using Interna-
conditioning team throughout the player’s academic year. The tional Society for the Advancement of Kinanthropometry (ISAK)
players were separated by general playing positions (forwards and guidelines, by an ISAK-qualified level 3 practitioner.18 Skinfold
backs) to establish differences between players in different posi- thickness showed adequate between-test reliability with a typical
tions and those players that went on to gain professional contracts error of 11.0%.
and those players that did not (professional and amateur). The
variables used are consistent with those previously used and Strength and power
included stature, body mass, skinfold thickness, muscular strength
and power, sprint speed (via split times) and intermittent running On the first testing occasion in the participants first year of the
ability. program, 1RM was estimated from a number of resistance training
144
M.J. Hamlin, R.W. Deuchrass, C.E. Elliot et al. Journal of Exercise Science & Fitness 19 (2021) 143e149

exercises using a 5e10 repetition maximum protocol and the for- performed 2e3 repetitions and the time to complete the total
mula of Brzycki (1993)19 which has since been shown to have good distance (10, 20 or 30 m) of each sprint was recorded, with the
validity in predicting 1RM (r2 ¼ 0.99).20 Thereafter, a true 1RM was average of the fastest 2 times recorded for analysis. Each of the 2
measured. The strength exercises included deadlift, back squat, efforts was performed after at least a 2-min rest from the previous
bench press, chin-up and prone row. The power clean was used to repetition.
indicate lower body power. All lifting attempts were observed by
qualified and experienced strength and conditioning trainers and Intermittent running ability
only lifts using correct technique were recorded.
During the bench press, player’s used a self-selected hand po- Intermittent running ability was estimated by the 20-m shuttle
sition and their feet remained in contact with the floor while the run test (Yo-Yo IRT L1, BangsboSport, Denmark) which was
gluteus maximus and lower back remained in contact with the completed according to previously-published protocols23 and
bench throughout the lift. During the lift, the bar was lowered to involved players completing 2  20 m shuttle runs in time to an
the chest (with elbows at approximately 90 with no bouncing off audio signal. Players started with both feet on or behind the first
the chest) and returned to the start position where elbows were 20 m line and ran towards the second 20 m line aiming to reach this
fully extended but not locked. The back squat required the player to line in time with the audio signal. Players then turned and ran back
descend in a controlled manner until the top of the thighs were at to the starting line, again in time with the audio signal. At the end of
least parallel with the floor before returning to the standing posi- each 2  20 m shuttle there was a 10 s period where players walked
tion. The deadlift started with the weights resting on the platform or jogged around a cone (placed 5 m past the finishing line) back to
and the participants were instructed to lift the barbell while the starting line again for the next shuttle. The test was concluded
maintaining a neutral, straight back and to extend their knees and if, after one warning, the player failed to complete a shuttle in time,
hip in one movement (to avoid a straight-leg deadlift technique). or the player removed themselves voluntarily. The Yo-Yo IRT L1
The lift was completed when the hip was fully extended (the angle finishing shuttle number was converted to a distance and then used
between the trunk and the thigh was approximately 180 ). For the in the analysis. The CV for Yo-Yo IRT L1 in a similar-aged group of
prone row, each player was instructed to lie on the prone row, high rugby players has been shown to be 3e6.9%.24
bench, face down, keeping their head, chest, and legs flat for the
entire lift. The subjects then gripped the bar with a pronated self- Statistical analyses
selected hand grip. The players then removed the bar from the
bench supports and let it hang freely. When the bar was motionless, Means and standard deviations of the variables along with dif-
the players raised the bar toward their chest until it came in contact ferences between the means of the two groups were estimated
with the high prone row bench. When performing the chin-ups, a using a mixed modeling procedure (Proc Mixed) in the Statistical
reverse underhand grip (palms facing toward face) was used. Analysis System (Version 9.3, SAS Institute, Cary North Carolina,
Players were instructed to start from a stationary position with USA). We analysed the natural logarithm (and transformed back if
arms fully extended and complete a repetition with the chin necessary) of each measure to reduce any effects in non-uniformity
moving over the bar. The power clean required the player to set up of error and to obtain changes in measures and errors as percent-
in a crouched position over the bar on the floor with fully extended ages.25 The fixed effects were test time, year, group (Professional,
arms. From this position, the player was instructed to thrust up- Amateur) and their interaction. The random effects were subject
ward in a triple extension movement, pulling the barbell upward and residual variance. For the first year measurement, we took the
into the catch position on the front of the shoulders with elbows mean of the measurements taken in the first year of the program.
forward.21 The coefficients of variation (CV) for similar strength For the 3-year change measurement, we took the mean change in
testing protocols within professional rugby union players have the variables over 3 years. Results are presented as the standardised
been shown to be approximately 4.5%.22 Testing occurred on a de- mean difference along with the 95% confidence interval. A Pearson
loading week so that the players were coming into the test rela- correlation coefficient was calculated to investigate the association
tively un-fatigued. During this week, squat testing was completed between body mass, skinfold thickness and performance variables.
on Monday with bench press, prone row and chin up on P values are also given for the between group comparisons for
Wednesday and deadlift and power clean on Thursday. Testing those who use traditional hypothesis testing. We used an alpha
commenced after a light warm-up (3 min of light walking at self- level of p  0.05 for significance in this study. This study used a
selected pace) and stretch (3 min self-selected stretches). The convenience sample of 83 players producing up to 4 data points for
1RM protocol included one set of 10 reps at a relatively light load each dependent variable per year over a total of 3 years. The mean
that served as a specific warm-up, followed by a gradual increase in typical error (as a co-efficient of variation) over all tests was body
load until 1RM was achieved. The rate of progress in load was mass ¼ 2.7%, sum of 8 skinfolds ¼ 11.0%, 10 m ¼ 2.5%, 20 m ¼ 2.2%,
dependent on the player’s self-perceived capacity and ranged from 30 m ¼ 2.3%, back squat 6.9%, bench press ¼ 4.8%, chin up ¼ 4.0%,
1 to 20 kg. All 1RM’s were achieved within 3e6 attempts. Rest deadlift ¼ 5.8%, prone row ¼ 4.8%, power clean ¼ 7.1%, Yo-Yo IRT
periods between attempts was 3 min, and rest periods between L1 ¼ 10.9%.
exercises was at least 10 min.
Results
Speed
Of the 83 athletes that we collected data on over the study
All sprints were performed in suitable footwear (e.g. gym shoes) period, 24 (10 forwards, 14 backs) went on the gain contracts with
on artificial turf in an indoor stadium. The players were instructed professional rugby teams in New Zealand and overseas.
to sprint maximally for every repetition within the lane formed by
the Smart Speed (Fusion Sport, Queensland, Australia) single-beam First year of the program
electronic timing gates, which was approximately 2-m wide.
Players started each sprint with their foot on a line 0.3 m back from In the first year of the program, forwards that went on to
the light beam of the first timing gate, in a stationary upright po- become professional rugby players later in their career were older
sition, with no rocking back or forth before starting. Each player (0.4 ± 0.3 years, mean ± 95% confidence interval, p ¼ 0.004) and
145
M.J. Hamlin, R.W. Deuchrass, C.E. Elliot et al. Journal of Exercise Science & Fitness 19 (2021) 143e149

heavier (4.5 ± 2.5 kg, p < 0.001) than forwards that remained meet the threshold for a substantially practical change, except for
amateurs, however, there was no clear difference in the sum of the sum of 8 skinfolds where a change of 5e6% would be required.
eight skinfolds between these two groups (11.3 ± 15.4 mm,
p ¼ 0.14). Forwards that went on to become professional players
Discussion
were slower than amateur forwards (0.05 ± 0.03 s, p < 0.001;
0.08 ± 0.05 s, p < 0.001; 0.12 ± 0.07 s, p < 0.001 for the 10, 20 and
Anthropometrical and physical performance measures that
30 m sprint times, respectively). Differences in muscular power
clearly differentiate between development rugby players that go on
measured by the power clean test between the future professional
to become professionals or remain amateurs would be useful to
and amateur forwards was not different (3.9 ± 9.6 kg, p ¼ 0.43).
sport administrators, coaches, players and physical conditioning
The forwards that went on to become professional players were
staff. The aim of this study was to discover whether selected
stronger than those that remained amateur (15.3 ± 14.2 kg,
anthropometrical and physical performance measures might
p ¼ 0.03; 19.5 ± 9.7 kg, p < 0.001; 8.5 ± 5.9 kg, p ¼ 0.005;
characterize such players and how these measures change over a 3
10.0 ± 5.0 kg, p ¼ < 0.001 for the deadlift, bench press, chin up and
year period. We have found a number of significant differences
prone row, respectively). However, forwards that went on to
between players who became professional compared to players
become professional players were worse than players that
that remained amateur including differences in age, body mass,
remained amateur in the Yo-Yo IRT L1 test (157 ± 149 m, p ¼ 0.03).
sprint times and strength measures.
Overall, there were less differences in measurements between
Similar to previous research on rugby players,26e30 forwards
the backs (professional versus amateur, see Table 2). In the first year
were generally taller and heavier than backs, whether players went
of the program, backs that went on to become professional rugby
on to become professional players or not. Forwards also had higher
players were taller (3.5 ± 1.8 cm, p < 0.001) and heavier
skinfold thickness, lower Yo-Yo IRT L1 performances and slower
(4.6 ± 2.4 kg, p < 0.001) than backs that remained amateurs. In
sprint times, but higher strength and power measures compared to
addition, in the first year of the program, backs that went on to
backs. Such differences have been attributed to the different per-
become professional players were faster (0.06 ± 0.05 s,
formance characteristics between forwards and backs (i.e. forwards
p ¼ 0.03; 0.07 ± 0.07 s, p ¼ 0.05 for the 20 and 30 m sprint times,
require strength and power and need to show physical dominance
respectively) compared to backs that remained amateur.
in securing their own ball or stopping the opposition).31
Players, particularly forwards, that went on to secure profes-
Differences over three years sional rugby union contracts later in their careers tended to be
older and heavier in their first year, than players that did not secure
The mean 3-year change in performance within players that professional contracts. Although the forwards and backs that went
went on the become professionals and those that remained on to become professionals average body mass was lighter than
amateur are given in Table 3. Both groups showed increases in body that reported for Super Rugby Championship players (117 and 96 kg
mass over the 3 years, but the body mass of players that remained for Super Rugby forwards and backs respectively),30 their body
amateur increased more than the players that went on to become mass was significantly greater than the body mass of the amateur
professionals (4.2 ± 2.0%, mean 3-yearly % increase ± 95% confi- athletes. The body mass difference between the two groups high-
dence interval, p < 0.001). Some of this increased body mass in the lights the increased requirement of size within the professional
amateur group particularly, was likely to be body fat since the sum rugby game.32
of skinfolds also increased over the 3 years in this group (8.9 ± 6.2%, Although the players that went on to become professionals had
p ¼ 0.004, Table 2). Sprinting ability in all 3 sprint tests (10, 20 and higher body mass compared to the amateurs, there was no signif-
30 m) improved over the 3 years in the players that went on to icant difference in the skinfold thickness between the groups. In
become professionals (p < 0.001, p ¼ 0.004, p ¼ 0.009), which was players that became professionals, the higher body mass was
not mirrored in the amateur players (p ¼ 0.27, p ¼ 0.53, p ¼ 0.80). associated with longer sprint times (r ¼ 0.61. 0.73, 0.74 for the 10,
Performance in the Yo-Yo IRT L1 test improved in the individuals 20 and 30 m sprints, respectively) and decreased Yo-Yo IRT L1
that went on to become professional players (9.6 ± 10.1% over 3 distance (r ¼ 0.64), but significantly greater performance in most
years, p ¼ 0.05), but decreased in the players that remained strength and power tests (ranged from r ¼ 0.24 to 0.65). Similar to
amateur (5.0 ± 4.6%, p ¼ 0.03). Improvements in performance previous research,33 we found body mass is clearly an important
were found in the deadlift, back squat, bench press, prone row and factor when considering level of performance in rugby.
power clean in the forwards and back squat, bench press, chin up, As might be expected, a concentrated and specific training
prone row and power clean in the backs over the 3 years in the program over 3 years in all players resulted in some significant
separate professional and amateur groups with no significant dif- anthropometrical and physical changes. Because rugby requires
ference between groups. high levels of strength and speed,34 there was a major emphasis on
Overall there was little difference between the two groups in strength and power training over the 3 years (Table 1), with
terms of the smallest worthwhile change for each test (Table 4). In approximately 40% of training spent on strength and power con-
most cases a change in performance of 1e3% would be sufficient to ditioning. Typically, such training results in skeletal muscle

Table 1
Weekly training schedule for rugby players.

Mon Tue Wed Thur Fri Sat Sun

Morning Speed/strength (90 min) Speed/strength/power

(60 min)
Afternoon Strength Aerobic/small-sided games (20 min) Skills Rugby match
(60 min) (60 min) (90 min)
Evening Rugby club training Rugby club training (90 min)
(90 min)

The strength, speed and power training sessions could be scheduled in the morning or afternoon depending on player’s university commitments.

146
M.J. Hamlin, R.W. Deuchrass, C.E. Elliot et al. Journal of Exercise Science & Fitness 19 (2021) 143e149

Table 2
Average anthropometrical and physical performance characteristics of the players in the first year of the scholarship program that went on to become professional or amateur
players.

Forwards Backs

Professional (n ¼ 10) Amateur (n ¼ 36) Professional (n ¼ 14) Amateur (n ¼ 23)

a b
Age (yr) 19.2 ± 0.8 18.8 ± 1.6 18.8 ± 0.7 18.6 ± 0.6
Stature (cm) 187.5 ± 8.8 186.7 ± 5.9 183.9 ± 7.3b 180.4 ± 4.7yc
Body mass (kg) 104.3 ± 5.4 99.7 ± 10.2a 86.9 ± 7.9b 82.3 ± 6.7yc
Skinfold thickness (mm) 105.5 ± 26.4 94.1 ± 25.6 62.9 ± 12.0b 65.6 ± 16.7c
10 m sprint (s) 1.89 ± 0.12 1.83 ± 0.07a 1.72 ± 0.04b 1.75 ± 0.05c
20 m sprint (s) 3.23 ± 0.15 3.14 ± 0.11a 2.95 ± 0.09b 3.00 ± 0.11yc
30 m sprint (s) 4.50 ± 0.23 4.37 ± 0.14a 4.08 ± 0.08b 4.16 ± 0.12yc
Yo-Yo IRT L1 (m) 1593 ± 413 1751 ± 343a 2065 ± 409b 1979 ± 352c
Deadlift (kg) 180 ± 33 165 ± 13a 158 ± 18b 149 ± 21c
Back Squat (kg) 158 ± 26 149 ± 24 137 ± 13b 135 ± 16c
Bench Press (kg) 129 ± 17 110 ± 14a 100 ± 10b 101 ± 18c
Chin-Up (kg) 137 ± 10 128 ± 10a 121 ± 11b 118 ± 9c
Prone Row (kg) 102 ± 8 92 ± 12a 88 ± 7b 86 ± 7c
Power Clean (kg) 100 ± 9 96 ± 15 84 ± 16b 87 ± 10c

Data are mean ± SD. Skinfold thickness; sum of 8 skinfolds; Yo-Yo IRT L1, Yo-Yo Intermittent Recovery Test-Level 1.
a
Significant difference between professional and amateur forwards (p  0.05); y Significant difference between professional and amateur backs (p  0.05).
b
Significant difference between backs and forwards in professional players (p  0.05).
c
Significant difference between backs and forwards in amateur players (p  0.05).

Table 3
Average 3-yearly change (%) in anthropometrical and physical performance measures in rugby players that went on to become professional or amateur players and differences
between these player groups.

Professional (n ¼ 24) (% change) Amateur (n ¼ 59) (% change) Mean Between Group Difference and 95% confidence interval
a b
Body mass 2.9 ± 1.7 7.0 ± 3.4 4.2 ± 2.0c
Skinfold thickness 9.6 ± 12.0 8.9 ± 6.2b 0.4 ± 12.9
10 m sprint 4.4 ± 2.5a 0.7 ± 1.5 3.7 ± 2.8c
20 m sprint 3.1 ± 2.0a 0.3 ± 1.0 2.8 ± 2.3c
30 m sprint 2.7 ± 2.0a 0.1 ± 0.9 2.8 ± 2.1c
Yo-Yo IRT L1 9.6 ± 10.1a 5.0 ± 4.6b 14.7 ± 11.0c
Deadlift 10.7 ± 10.6a 10.1 ± 14.3 0.6 ± 16.7
Back Squat 7.2 ± 6.4a 9.3 ± 4.4b 2.2 ± 7.0
Bench Press 9.5 ± 6.4a 11.9 ± 5.7b 2.4 ± 7.3
Chin-Up 2.9 ± 3.9 5.6 ± 2.6b 2.6 ± 4.4
Prone Row 11.3 ± 5.4a 13.8 ± 6.6b 2.5 ± 4.8
Power Clean 13.4 ± 9.1a 13.6 ± 6.4b 0.2 ± 9.1

Data are mean percent changes of all players in each group over 3 years ± 95% confidence interval. Skinfold thickness; sum of 8 skinfolds; Yo-Yo IRT L1, Yo-Yo Intermittent
Recovery Test-Level 1.
a
Significant differences within professional group (p  0.05).
b
Significant differences within amateur group (p  0.05).
c
Significant differences between professional and amateur players (p  0.05).

Table 4 players at least, indicates that some of the increase in body mass
The smallest worthwhile effects in anthropometrical and physical performance during the training period may be associated with increased fat
measures in players that went on to become professional or remained amateurs.
mass.
Professional (n ¼ 24) Amateur (n ¼ 59) The major differences between groups in their adaptation to
Body mass 2.3% (1.3e3.0) 2.2% (1.6e2.6) training over the 3-year period was substantially improved
Skinfold thickness 6.1% (2.1e8.4) 5.6% (4.1e6.8) sprinting and Yo-Yo IRT L1 ability in the players that went on to
10 m sprint 1.2% (0.6e1.6) 0.6% (0.4e0.8) become professional compared to those that remained amateur.
20 m sprint 1.1% (0.5e1.4) 0.6% (0.4e0.8)
Having higher skinfold thickness which can reasonably indicate fat
30 m sprint 1.0% (0.4e1.4) 0.6% (0.3e0.8)
Yo-Yo IRT L1 5.3% (2.4e7.2) 3.1% (2.0e4.0) mass,35 may result in less power-to-weight ratio and lower sprint
Deadlift 2.7% (1.1e3.1) 1.9% (1.0e2.5) speed,36 therefore the professional players would have an advan-
Back Squat 3.4% (1.7e4.6) 2.5% (1.8e3.1) tage here. The increased mass in the amateur players would also
Bench Press 3.3% (1.6e4.4) 2.2% (1.5e2.7) result in lower Yo-Yo IRT L1 scores, given the number of changes in
Chin-Up 1.9% (0.8e2.6) 1.4% (0.9e1.8)
direction and the accompanying deceleration and acceleration
Prone Row 1.7% (0.1e2.4) 2.0% (1.3e2.4)
Power Clean 3.3% (0.7e4.7) 2.3% (1.4e2.9) required with such tests.37 In a recent study on rugby league
players, Scott et al. (2017) also found that body mass and sum of 7
Data are mean smallest worthwhile effect over the 3-year period (and 95% confi-
dence interval). Skinfold thickness; sum of 8 skinfolds; Yo-Yo IRT L1, Yo-Yo Inter- skinfolds were negatively correlated to intermittent running abil-
mittent Recovery Test-Level 1. ity.38 During the three years of the program, players are exposed to
selection processes for higher team selection and players start to
comprehend whether they will progress as professional athletes or
hypertrophy16 which could account for the substantial increase in not. The gradual realization that a professional career is beyond
body mass of the 2 groups over the 3-year training period. How- some players may help to explain some of the changes in body
ever, the substantial increase in skinfold thickness, in the amateur composition (i.e. increased fat mass) and performance witnessed
147
M.J. Hamlin, R.W. Deuchrass, C.E. Elliot et al. Journal of Exercise Science & Fitness 19 (2021) 143e149

over time between the groups.9 Although speculative, it is our Conclusion

belief, in our players at least, that players in contention for pro-
fessional team selection tend to train harder and with more intent Forward rugby players that go on to secure professional con-
than players that are not. In addition, players making selection tracts later in their careers tend to be older, heavier and stronger,
teams have increased access to technical and tactical coaching as but not necessarily faster or more aerobically fit than forwards that
well as training more and playing more games over the season remain amateurs. Whereas, backs that go on to become profes-
compared to those not in selection contention. sional players tend to be taller, heavier but faster than backs that
Improvement in strength and power over the training period remain amateur. Players (both forwards and backs) that go on to
was similar in both groups with no clear differences in the change secure professional contracts tend to improve strength and power
scores between groups (Table 3). In most cases, improvement was performance over time while maintaining body fat levels. Addi-
in the range from 3 to 13% over the 3 years. Indeed, when looking at tionally, calculating the smallest worthwhile effects from regular
similar research over longer term studies, the average yearly performance and anthropometrical tests will help coaches verify
improvement in strength and power (taken as an average yearly whether players are making meaningful improvements over time.
change in similar tests; bench press, back squat, chin-up, power
clean) in the players of this study were similar to New Zealand CRediT authorship contribution statement
provincial players (2.8% per year), but substantially lower than
Super Rugby or International players (3.2% and 4.0% per year Michael J. Hamlin: Conceptualization, Methodology, Data
respectively).16 The greater improvements in strength and power of curation, Formal analysis, Writing - original draft. Richard W.
the Super Rugby and International players compared to the players Deuchrass: Investigation, Writing - review & editing, Data curation.
in this study, are probably due to playing more games, training Catherine E. Elliot: Methodology, Writing - review & editing.
more, access to high quality nutrition and effective recovery Nuttaset Manimmanakorn: Methodology, Writing - review &
stratergies.39 editing.
Regular measurement of physical performance and body
composition data over time is an important aspect of monitoring Declaration of competing interest
athletes. Such information is useful in making decisions about the
magnitude of change in these variables and thereby the success or None.
otherwise of the training. To make inferences about practically
important changes we have calculated the average (over all tests) Acknowledgements
smallest worthwhile change for each of the tests used (Table 4).
Such data needs to be used in conjunction with the magnitude of The authors thank the rugby players and their coaches for their
the typical error of each test to identify changes of substantial assistance. No financial or material support of any kind was
practical importance for the players. For example, the typical error received to produce the work described in this article, however the
(as a coefficient of variation) for the Yo-Yo IRT L1 in the rugby publication has been partially fiananced by the Lincoln University
players of this study, indicate intermittent running performance Open Access Fund. The authors have no conflicts of interest relevant
varied by approximately 10.9% from test to test when conducted to this article.
1e2 months apart. This reliability in relation to the smallest
worthwhile change in Table 4 (3e5%) suggests there is too much References
variation in performance to identify changes of the smallest
worthwhile effect and that only larger effect sizes will be noticed 1. Duthie G, Pyne D, Hooper S. Applied physiology and game analysis of rugby
until measures to reduce the test-to-test variation are union. Sports Med. 2003;33:973e991.
2. Takamori S, Hamlin M, Kieser D, et al. Senior club-level rugby union player’s
implemented.
positional movement performance using individualized velocity thresholds
A limitation to this study is the fact that a large number of and accelerometer-derived impacts in matches. J Strength Condit Res. 2020.
variables are involved in rugby union performance and success and https://doi.org/10.1519/JSC.0000000000003523.
3. Duthie GM. A framework for the physical development of elite rugby union
can therefore effect the chances of players securing professional
players. Int J Sports Physiol Perform. 2006:2e13.
contracts. Many of these variables (injury, skill level, attitude to 4. Coughlan GF, Green BS, Pook PT, Toolan E, O’connor SP. Physical game demands
training and recovery, diet, willingness to train, adaptability) have in elite rugby union: a global positioning system analysis and possible impli-
not been measured in this study, and therefore cannot be ruled out cations for rehabilitation. J Orthop Sports Phys Ther. 2011;41:600e605.
5. Cummins C, Orr R, O’Connor H, West C. Global Positioning Systems (GPS) and
as important indicators for players achieving professional con- microtechnology sensors in team sports: a systematic review. Sports Med.
tracts. In addition, a number of the field tests used in this study 2013;43:1025e1042.
have not been validated in a rugby union context and results should 6. King D, Cummins C, Hume PA, Clark TN, Pearce AJ. Physical demands of
amateur senior domestic rugby union players over one round of competition
be viewed with caution until validation studies on these tests have matches in New Zealand assessed using heart rate and movement analysis. Int J
been performed. Because the study has used skinfold thickness and Sports Sci Med. 2018;2:66e71.
not more accurate body composition measures the study cannot 7. Dobbin N, Highton J, Moss SL, Twist C. The discriminant validity of a stan-
dardized testing battery and its ability to differentiate anthropometric and
elucidate how changes in body composition might affect the re- physical characteristics between youth, academy, and senior professional
sults. A further limitation is the relatively small subject sample of rugby league players. Int J Sports Physiol Perform. 2019;14:1110e1116.
83 players, especially when analyzing sub-groups. Due to this small 8. Tredrea M, Dascombe B, Sanctuary CE, Scanlan AT. The role of anthropometric,
performance and psychological attributes in predicting selection into an elite
sample size, the results are not intended to be interpreted as development programme in older adolescent rugby league players. J Sports Sci.
normative for all rugby union players, and differences between 2017;35:1897e1903.
players that became professional compared to those that remained 9. Young WB, Newton RU, Doyle TL, et al. Physiological and anthropometric
characteristics of starters and non-starters and playing positions in elite
amateur should be viewed with caution until further analysis on
Australian Rules football: a case study. J Sci Med Sport. 2005;8:333e345.
greater subject numbers is conducted. Finally, as the subject sample 10. Veale JP, Pearce AJ, Carlson JS. The Yo-Yo intermittent recovery test (level 1) to
is from one university rugby club, the results have the potential to discriminate elite junior Australian football players. J Sci Med Sport. 2010;13:
be biased and influenced by factors associated with this club such 329e331.
11. Le Rossignol P, Gabbett TJ, Comerford D, Stanton WR. Repeated-sprint ability
as coaches, strength and conditioning staff etc. and team selection in Australian football league players. Int J Sports Physiol
Perform. 2014;9:161e165.

148
M.J. Hamlin, R.W. Deuchrass, C.E. Elliot et al. Journal of Exercise Science & Fitness 19 (2021) 143e149

12. Till K, Cobley S, O’Hara J, Chapman C, Cooke C. A longitudinal evaluation of week exercise intervention for improved neck muscle strength in amateur
anthropometric and fitness characteristics in junior rugby league players male rugby union players. J Sport Exer Sci. 2020;4:33e39.
considering playing position and selection level. J Sci Med Sport. 2013;16: 27. Duthie G, Pyne D, Hopkins W, Livingstone S, Hooper S. Anthropometry profiles
438e443. of elite rugby players: quantifying changes in lean mass. Br J Sports Med.
13. Till K, Cobley S, O’Hara J, Morley D, Chapman C, Cooke C. Retrospective analysis 2006;40:202e207.
of anthropometric and fitness characteristics associated with long-term career 28. Quarrie K, Handcock P, Waller AE, Chalmers D, Toomey M, Wilson B. The New
progression in Rugby League. J Sci Med Sport. 2015;18:310e314. Zealand rugby injury and performance project. III. Anthropometric and phys-
14. Till K, Morley D, O’Hara J, et al. A retrospective longitudinal analysis of ical performance characteristics of players. Br J Sports Med. 1995;29:263e270.
anthropometric and physical qualities that associate with adult career attain- 29. Owen C, Till K, Weakley J, Jones B. Testing methods and physical qualities of
ment in junior rugby league players. J Sci Med Sport. 2017;20:1029e1033. male age grade rugby union players: a systematic review. PloS One. 2020;15,
15. Reardon C, Tobin DP, Tierney P, Delahunt E. The worst case scenario: locomotor e0233796.
and collision demands of the longest periods of gameplay in professional rugby 30. Posthumus L, Macgregor C, Winwood P, Darry K, Driller M, Gill N. Physical and
union. PloS One. 2017;12, e0177072. fitness characteristics of elite professional rugby union players. Sports. 2020;8:
16. Smart DJ, Hopkins WG, Gill ND. Differences and changes in the physical 85.
characteristics of professional and amateur rugby union players. J Strength 31. Lindsay A, Draper N, Lewis J, Gieseg SP, Gill N. Positional demands of profes-
Condit Res. 2013;27:3033e3044. sional rugby. Eur J Sport Sci. 2015;15:480e487.
17. Dobbin N, Hunwicks R, Highton J, Twist C. A reliable testing battery for 32. Geeson-Brown T, Jones B, Till K, Chantler S, Deighton K. Body composition
assessing physical qualities of elite academy rugby league players. J Strength differences by age and playing standard in male rugby union and rugby league:
Condit Res. 2018;32:3232e3238. a systematic review and meta-analysis. J Sports Sci. 2020:1e16.
18. Norton K, Whittingham N, Carter L, Kerr D, Gore C, Marfell-Jones M. Mea- 33. Jones B, Weaving D, Tee J, et al. Bigger, stronger, faster, fitter: the differences in
surement techniques in anthropometry. Anthropometrica. 1996;1:25e75. physical qualities of school and academy rugby union players. J Sports Sci.
19. Brzycki M. Strength testing: predicting a one-rep max from reps-to-fatigue. 2018;36:2399e2404.
J Phys Educ Recreat Dance. 1993;64:88e90. 34. Brazier J, Antrobus M, Stebbings GK, et al. Anthropometric and physiological
20. Reynolds JM, Gordon TJ, Robergs RA. Prediction of one repetition maximum characteristics of elite male rugby athletes. J Strength Condit Res. 2020;34:
strength from multiple repetition maximum testing and anthropometry. 1790e1801.
J Strength Condit Res. 2006;20:584e592. 35. Zemski AJ, Keating SE, Broad EM, Slater GJ. Longitudinal changes in body
21. Baker D, Nance S. The relation between strength and power in professional composition assessed using DXA and surface anthropometry show good
rugby league players. J Strength Condit Res. 1999;13:224e229. agreement in elite rugby union athletes. Int J Sport Nutr Exerc Metabol. 2019;29:
22. Argus C, Gill ND, Keogh JW, Hopkins WG, Beaven CM. Changes in strength, 24e31.
power, and steroid hormones during a professional rugby union competition. 36. Huovinen HT, Hulmi JJ, Isolehto J, et al. Body composition and power perfor-
J Strength Condit Res. 2009;23:1583e1592. mance improved after weight reduction in male athletes without hampering
23. Bangsbo J, Iaia FM, Krustrup P. The Yo-Yo Intermittent Recovery Test: a useful hormonal balance. J Strength Condit Res. 2015;29:29e36.
tool for evaluation of physical performance in intermittent sports. Sports Med. 37. Darrall-Jones J, Roe G, Carney S, et al. The effect of body mass on the 30-15
2008;38:37e51. intermittent fitness test in rugby union players. Int J Sports Physiol Perform.
24. Chiwaridzo M, Oorschot S, Dambi JM, et al. A systematic review investigating 2016;11:400e403.
measurement properties of physiological tests in rugby. BMC Sports Sci Med 38. Scott TJ, Duthie GM, Delaney JA, et al. The validity and contributing physio-
Rehabil. 2017;9:24. logical factors to 30-15 intermittent fitness test performance in rugby league.
25. Hopkins W, Marshall SW, Batterham AM, Hanin J. Progressive statistics for J Strength Condit Res. 2017;31:2409e2416.
studies in sports medicine and exercise science. Med Sci Sports Exerc. 2009;41: 39. Gabbett TJ. Physiological and anthropometric characteristics of junior rugby
3e12. league players over a competitive season. J Strength Condit Res. 2005;19:764.
26. Hamlin MJ, Deuchrass R, Elliot C, Raj T, Promkeaw D, Phonthee S. Effect of a 6-

149