International Journal of Performance Analysis in Sport

ISSN: 2474-8668 (Print) 1474-8185 (Online) Journal homepage: http://www.tandfonline.com/loi/rpan20

Correlates of Team Performance in Volleyball

Sotiris Drikos, Panagiotis Kountouris, Alexandros Laios & Yiannis Laios

To cite this article: Sotiris Drikos, Panagiotis Kountouris, Alexandros Laios & Yiannis Laios (2009)
Correlates of Team Performance in Volleyball, International Journal of Performance Analysis in
Sport, 9:2, 149-156

To link to this article: http://dx.doi.org/10.1080/24748668.2009.11868472

Published online: 03 Apr 2017.

Submit your article to this journal

Article views: 1

View related articles

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=rpan20

Download by: [University of New England] Date: 25 May 2017, At: 23:01
International Journal of Performance Analysis of Sport
2009, 9, 149-156.

Correlates of Team Performance in Volleyball

Sotiris Drikos, Panagiotis Kountouris, Alexandros Laios, Yiannis Laios

University of Athens, Faculty of Physical Education and Sports Science, Department of

Court Games - Volleyball, Greece

Abstract

The overall performance of a Volleyball team depends on many factors,

from which decisive are considered to be the execution of skills that lead
immediately to winning or losing the rally. These are lost serves, aces, kill-
attacks, attack errors and kill-blocks. The analysis of these skills in relation
to team performance, as expressed by the ratio of sets won to the total
number of sets, lead to the formation of two new correlates. These are the
serving efficiency ratio (SER), defined as the ratio of lost serves to aces,
and the attack efficiency ratio (AER), defined as the number of kill attacks
divided by the sum of attack errors and kill-blocks. Analysis of the data
collected from all the matches of the male A1 volleyball professional league
of 2005-2006 in Greece proved that the two efficiency ratios were better
predictors of the teams’ overall performance than the five original
variables. The findings lead to clear-cut definitions of norms both for the
serving and attack efficiency ratio. The leading teams had a SER of around
two and an AER of around three. These criteria are valuable tools
especially for Volleyball coaches in deciding for the appropriate tactics of
their teams.

Keywords: Volleyball, performance, serving efficiency ratio, attack

efficiency ratio.

1. Introduction

The overall performance of a Volleyball team depends on many factors, from which the
skills that lead immediately to winning or losing of the point are considered decisive. In
Volleyball there are five decisive actions that lead immediately to winning or losing the
rally; two of them are inherent to the execution of the serve and they are the serve aces
and lost serves. The other three appear at the execution of attacks and they are the kill-
attacks, attack errors and attacks stopped by kill-blocks.

Consider a game of Volleyball between teams A and B, the score being at a particular set
24-23 in favour of team A. If team A wins the next rally it also wins the set, since the
score will be 25-23. If team B wins the rally the teams will have to play up to 26 points,
since the score will be 24-24. The difference between 25-23 and 24-24 is actually a two-
point difference, since it is the difference between both winning and losing a point. This

149
is the essence of the rally-point system that applies to the game of Volleyball as well as to
a number of other games. In this sense a lost rally carries a negative weight practically
equal to the positive weight of a won rally. Moving from the rally to the set, the team that
wins the set is the team that has a better ratio of actions with a positive or negative
outcome. Finally, the ratio of sets won to the total number of sets played by the team is
considered to be the best reflector of the overall team performance.

The above arguments advocate that the independent reporting of proportions of positive
or negative actions may not convey a complete picture of a team’s or a player’s
efficiency in the specific skill and consequently of the team’s overall performance. Up to
date there have been several studies that attempted to correlate the effectiveness in
specific skill with the overall team performance, aiming to evaluate the order of
significance of these skills.

Eom and Schutz (1992) identified the performance of the attack and block as the most
important skills for winning or losing a match in international level men’s volleyball,
while Palao et al. (2004), using data from the Olympic Games of Men and Women of
Sydney in 2000, undertook the task to correlate performance in technical skills of
Volleyball with team level. The results showed that in male teams there were statistically
significant differences between the levels of teams in the skills of attack and block. The
number of errors in all skills was inversely proportional to the level of the teams.

Laios and Kountouris (2005) examined the changes in the effectiveness of technical skills
of Volleyball after the application of the new regulations, through the comparison of the
matches in two consecutive Olympiads of Sydney in 2000 and Athens in 2004. Five skills
were recorded and analyzed (serve, reception, attack, block and defence). The results
showed an increase in the effectiveness of jump serves, as well as a reduction of errors in
the serve. There were no differences in the effectiveness of attack, except from a slight
reduction in the proportion of lost attacks. The most important change was observed in
the effectiveness of the block. In conclusion, practically all the teams displayed a
reduction of errors in all the technical skills under consideration.

In a most recent research, Lobieti et al. (2006) examined the relationships between
performance parameters and final ranking in both male and female Italian A1 volleyball
professional leagues. It was found that attack has to be considered the most important
skill for men’s volleyball, followed by block and serve aces.

It is evident that the approach in all the above studies was to examine separately the
association between the proportion of positive and negative actions with overall team
performance. The aim of the present study was to determine whether latent derivative
parameters, computed from relative proportions of decisive actions that lead immediately
to winning or losing the rally, can be better predictors than the original proportions of
overall team’s performance in Volleyball expressed as the ratio of sets won to the total
number of sets played by the team.

150
2. Method

The primary data were collected from the Greek Men’s Volleyball Championship of
2005-2006 with the participation of twelve teams in a two round system, so that each
team played with each other, both home and away. This gives a total of 132 matches, in
which 21570 points in 491 sets were played.

By convention in the Greek Men’s Volleyball Championship every team has a

specialized person responsible for the data entry of all the events during a match. All
teams use the same computer software coding program that has been specifically
designed for volleyball: Data Volley (Data Project, Salerno, Italy, Release 2.1.9).

The primary dependent variable was the ratio of the number of sets won to the total
number of sets played by the team (Set Ratio=SR). Also examined was the final ranking
of the team.

The five independent variables used in the analysis are from actions that immediately win
or lose a point. Thus for each team there was the proportion of serves lost and serve aces
to the total number of serves. From these two variables a new variable was formed, which
was the ratio of serves lost to serve aces. From the total number of attacks for each team
three proportions were taken into consideration: Attacks won, attacks lost, attacks
stopped by kill-block. An attack can be immediately lost in two ways: either the ball hits
the net or goes out of bounds, or the hitter is stopped by a kill-block. Therefore, from
these three proportions a new ratio variable was formed with the attacks won divided by
the sum of lost attacks and kill-blocks. In this manner the two new variables can be
characterized as the serve efficiency ratio (SER) and attack efficiency ratio (AER).

The validity and reliability of the data entry with regards to the five original independent
variables were checked as follows: since practically all matches are video-recorded, a
particular match day was randomly selected and all six matches of this match day were
recoded into the program by an independent observer one week after the matches.
Consequently this was an inter-observer analysis. The accordance of the recordings of the
two observers for each of the variables used in the analysis was well above 95% that is
usually set as a criterion of acceptance.

Statistical analyses included Pearson’s and Spearman’s correlation coefficients and

multiple stepwise linear regression models. Pearson’s correlations were employed when
the independent variables (proportions or ratios) were compared with SR or with total
points. Spearman’s rank correlations were used when comparing the teams’ final ranking
with the ranking of the teams according to the proportion or ratio under investigation.
Linear regression has been carried out to detect the relationship among the parameters
and their decisive role in the success of a volleyball team.

151
3. Results

Table 1 shows the final ranking of the teams according to their total points and their set
ratio (SR). The correlation between total points and SR is practically a perfect one
(r=0.996, p<0.01). There are, however, a couple of noteworthy discrepancies. Firstly,
while teams 4 and 5 have the same SR (0.59), team 4 surpassed team 5 by two points.
Secondly, while team 9 has a higher SR than team 8 (0.40 vs. 0.38), the team 9 gathered
two points fewer (21 vs. 23). An SR value more than 0.5 signifies that the team has won
more sets than it has lost. As it can be seen from the table, this was accomplished by the
first five teams according to the final rankings.

Table 1. Final ranking of the teams according to their total points and their sets ratio
(SR).
Team ranking Points in ranking Total sets Sets won SR
2 59 80 64 .80
3 47 77 52 .68
4 42 88 52 .59
5 40 82 48 .59
6 32 88 43 .49
7 31 80 37 .46
8 23 82 31 .38
9 21 85 34 .40
10 20 86 32 .37
11 15 80 23 .29
12 6 76 12 .16

Table 2 shows the Pearson’s and Spearman’s correlations between the independent and
dependent variables. The Pearson’s correlation coefficient between SR and the proportion
of lost serves, although negative as it should be, is not strong (r=-.0385, p=0.217). A
better correlation is observed between SR and the proportion of serve aces (r=0.664,
p=0.019). A significantly improved correlation coefficient is observed between SR and
SER (r=-0.823, p=0.001). This correlation coefficient is slightly better than the
correlation coefficient of points in ranking with SER (r=-0.816, p=0.001). The
Spearman’s correlations between the real rankings and the rankings according to the SER
values are also significantly improved.

Table 2. Pearson’s and Spearman’s correlations between the independent and dependent
variables (* p<0.05, ** p<0.01).
Play Outcome Points in ranking SR Spearman’s ρ
Serve Aces 0.665* 0.664* 0.594*
Lost -0.379 -0.385 0.329
Lost/Aces (SER) -0.816** -0.823** 0.720**
Attack Attacks won 0.889** 0.911** 0.874**
Attack errors -0.703* -0.704* 0.720**
Attacks blocked -0.793** -0.828** 0.741**
AER 0.912** 0.932** 0.853**

152
The importance of SER as a predictor of the team’s performance is illustrated in Figure 1.
The twelve numbered points, representing the twelve teams with their corresponding
rankings, lie very close to a straight line with a negative slope, meaning that a decrease in
SER is associated with an increase in SR and consequently with higher team ranking.
With the exception of team 4, all teams that finished in the first five places and had won
more sets than they had lost had a serve efficiency ratio (SER) less than 2.2, while all the
other teams’ SER was higher than 2.5.

1
2
Sets won/Total sets (SR)

3
5 4
0.5 6
7
8 9
10
11

12

0
1.7 2.2 2.7 3.2 3.7
Serves lost/Aces (SER)

Figure 1. Scatter plot of serve efficiency ratio (SER) versus proportion of won sets by the
teams. Numbers indicate the team’s ranking.

The same is observed for the correlation coefficients between SR and the proportion of
errors in attack (r=-0.704, p<0.05), the proportion of kill attacks (r=0.911, p<0.01) and
the proportion of attacks killed by the opponents’ block (r=-0.828, p<0.01). An improved
correlation coefficient is observed between SR and AER (r=0.932, p<0.001). Again this
correlation coefficient is slightly better than the correlation coefficient of points in
ranking with AER (r=0.912, p<0.001). The importance of AER as a predictor of the
team’s performance is illustrated in Figure 2. The twelve numbered points, representing
the twelve teams with their corresponding rankings, lie very close to a straight line with a
positive slope, meaning that an increase in SER is associated with an increase in SR and
consequently with higher team ranking. With the exception of team 4, all teams that
finished in the first five places had an attack efficiency ratio (AER) higher than 3.

153
 1

1
2
Sets won/Total sets (SR)

3
4 5
0.5 6
7
8 9
10
11

12

0
1.7 2.7 3.7
Attacks won/Attacks lost (AER)

Figure 2. Scatter plot of attack efficiency ratio (AER) versus proportion of won sets by
the teams. Numbers indicate the team’s ranking.

The linear regression model of SR as the dependent variable and AER and SER as the
independent variables produced the following linear equation:

SR=0.001 + 0.272 X AER – 0.087 X SER

The multiple R of the regression model was 0.940 (p<0.001), giving an R2=0.884, which
means that 88.4% of the variance of the dependent variable is explained by the variability
of the two independent variables. Furthermore, the intercept of the equation is practically
equal to 0 (a=0.001).

4. Discussion

The present study employed a novel approach in the quantification of the predictability of
overall team performance in Volleyball through combinations of absolute frequencies of
positive and negative decisive actions. The findings showed that serve efficiency ratio,
defined as the ratio of serves lost to serve aces, and attack efficiency ratio, defined as the
ratio of the number attacks won divided by the sum of lost attacks and kill-blocks, are
better predictors of team performance, expressed as the ratio of sets won to the total
number of sets, than the original proportions of the positive and negative actions.

154
This approach is vindicated by the application of the rally-point scoring system. As
pointed out by Marelic et al. (2004), “the changes in Volleyball regulations applied by
Federation Internationale Volleyball (F.I.V.B., 2000) have opened a space for new
performance-related investigations, representing a big challenge for research into the
team-characteristics of play which are crucial for success”.

Serve aces seem to be good predictors of the team’s overall ability to win the set. This is
in agreement with findings by Lobieti et al. (2006). Conversely serve errors by
themselves do not seem to bear an appreciable consequence on team performance.
However, it is the ratio of the two (the serve efficiency ratio - SER) that becomes an
important predictor of the team’s performance. According to the present findings a team
aspiring for a high final ranking should be capable of winning an ace for approximately
every two serves that it loses, while for teams in the last rankings this ratio increases to
three.

The criterion of serve efficiency ratio can be easily applied to individual players. Serve
aces are predominantly the result of jump serves, performed with increased risk. The
serve efficiency ratio serves as a norm about whether this risk is worthwhile to be taken.
For example, a server accomplishing 8 aces at the expense of 24 lost serves
(SER=24/8=3) may be less valuable to his team than a server with only 4 aces but also
with only 8 lost serves (SER=8/4=2).

In this respect the serving efficiency ratio, which can be easily calculated in real-time
conditions even without the use of a computer, can be a valuable tool both for the player
and especially for the coach. During practice a player should try to minimize this ratio.
During the match, the coach can use this ratio as a criterion to decide when and which
players should risk their serves.

Overall, the three attack parameters, especially the proportion of kill-attacks are better
predictors of team performance than serve parameters, a finding which has been
consistently reported by several investigators (Marelic et al., 2004; Palao et al., 2004;
Laios & Kountouris, 2005; Lobietti et al., 2006). However, once again the attack
efficiency ratio (AER) seems to be a better predictor of team performance than its
constituent elements. According to the present findings, a team aspiring for a high final
ranking should be capable of winning at least three points from its attack for every point
that it loses through an error or the opponent’s kill-block. On the contrary, for teams in
the last rankings the AER is around two.

Equivalent to the case of SER, the attack efficiency ratio (AER) is a simple measure,
which can provide valuable information to the coach not only for the team’s, but also for
individual hitters’ performance variability both within a game, but also during a
tournament and the whole season.

In consequence to the above, it is not surprising that the set ratio can be so well predicted
from SER and AER. This finding also corroborates the notion that the set ratio reflects

155
better the overall performance of a team than the total points gained according to the
rules of the tournament.

In conclusion, the present study proposed two simple measures of team and player
efficiency that seem to correlate very well with the overall team performance.
Furthermore, both the serving and the attack efficiency ratios seem to provide cut-off
values that qualify elite team performance. Further studies are required in order to
elucidate the status of these criteria in elite tournaments, both for men and women.

5. References

Data Project (2005) Manuale del Data Volley versione professionale.

Webite:http://www.dataproject.com.
Eom, H. J. abd Schutz, R. W. (1992) ‘Statistical analyses of volleyball team
performance’, Research Quarterly for Exercise and Sport, 63: 11-18.
F.I.V.B. (2000) Official Volleyball Rules. Approved by the XXVIIth F.I.V.B. World
Congress in Seville, Spain 2000. F.I.V.B. New Edition.
Laios Y. and Kountouris P. (2005) ‘Evolution in men’s volleyball skills and tactics as
evidenced in Athens 2004 Olympic Games’, International Journal of
Performance Analysis in Sport, 5: 1-8.
Lobietti, R., Michele, R. and Merni, F. (2006) ‘Relationships between performance
parameters and final ranking in professional volleyball’, World Congress of
Performance Analysis of Sport 7, Hungary, August 2006.
Marelic, N., Resetar, T. and Jankovic, V. (2004) ‘Discriminant analysis of the sets won
and the sets lost by one team in A1 Italian volleyball league – a case study’,
Kinesiology, 36: 77-82.
Palao, J. M., Santos, J. A. and Urena, A. Y. (2004) ‘Effect of team level on skill
performance in volleyball’, International Journal of Performance Analysis in
Sport, 4(2): 50-60.

Corresponding author:

Sotiris Drikos
e-mail: [email protected]
Address: 45 Theatrou Str. 18534 Pireus, Greece
Tel: +302104121937 Fax: +302104121937

156