Performance indicators that distinguish winning and losing teams in basketball

Abstract

To prepare a team for basketball games, to build up the best tactics, to make good decisions during a game, coaches need to know which elements of matches are the most crucial ones. Especially at close games where there is small difference between the performances of two teams. The main purpose of this study was to identify those critical performance indicators that most distinguish between winning and losing performances within matches. The statistical analysis of basketball games can lead to the identification of many significant performance indicators, not all of which can be analysed in real time. Therefore, a smaller subset of critical performance indicators can be identified by analysing close matches only. Data from 54 matches were gathered from the official score sheets of the European Basketball Championship 2007. Cluster analysis was used to classify the matches into three types such as tight games, balanced games and unbalanced games. There were 28 of these matches that were close matches where the differences between the two teams were 9 points or less. Wilcoxon signed ranks tests were used to compare 18 performance indicators between the winning and losing teams within each type of match. There were 13 significant performance indicators for the full set of matches. This was reduced to 6 critical performance indicators when only the close matches were considered. The analysis of tight matches explored that the winning teams had significantly less 3 point attempts (p<0.05) with higher shooting percentage (p<0.01). The number of successful free throws (p<0.01), the free throw percentage (p<0.001) and the number of defensive rebounds (p<0.01) also contributed to achieve a higher number of scored points and consequently determined success.

Full text

International Journal of Performance Analysis of Sport
2009, 9, 60-66.
Performance indicators that distinguish winning and losing teams in basketball
Gabor Csataljay1, Peter O’Donoghue2, Mike Hughes2 and Henriette Dancs1
1University of West Hungary, Savaria University Centre, Szombathely, 9700, Karolyi
G. square 4, Hungary
2Cardiff School of Sport, University of Wales Institute Cardiff, Cyncoed Campus,
Cardiff, Wales, CF23 6XD, UK
Abstract
To prepare a team for basketball games, to build up the best tactics, to
make good decisions during a game, coaches need to know which
elements of matches are the most crucial ones. Especially at close games
where there is small difference between the performances of two teams.
The main purpose of this study was to identify those critical performance
indicators that most distinguish between winning and losing performances
within matches. The statistical analysis of basketball games can lead to the
identification of many significant performance indicators, not all of which
can be analysed in real time. Therefore, a smaller subset of critical
performance indicators can be identified by analysing close matches only.
Data from 54 matches were gathered from the official score sheets of the
European Basketball Championship 2007. Cluster analysis was used to
classify the matches into three types such as tight games, balanced games
and unbalanced games. There were 28 of these matches that were close
matches where the differences between the two teams were 9 points or less.
Wilcoxon signed ranks tests were used to compare 18 performance
indicators between the winning and losing teams within each type of
match. There were 13 significant performance indicators for the full set
of matches. This was reduced to 6 critical performance indicators when
only the close matches were considered. The analysis of tight matches
explored that the winning teams had significantly less 3 point attempts
(p<0.05) with higher shooting percentage (p<0.01). The number of
successful free throws (p<0.01), the free throw percentage (p<0.001) and
the number of defensive rebounds (p<0.01) also contributed to achieve a
higher number of scored points and consequently determined success.
Keywords: basketball, game analysis, close games, performance indicators
60

1. Introduction
Performance indicators are used to assess the performance of an individual, a team or
elements of a team (Hughes and Bartlett, 2002). Well-chosen performance indicators
help coaches to identify good and bad performances (Bartlett, 2001; Hughes and Franks
1997, 2004; 2008), either at individual or team level. Performance indicators are often
used to define the differences between winning and losing teams. To build up the best
strategy, to make rational tactical decision and to enhance the team performance,
coaches need to know which elements are the critical ones that most distinguish
between winning and losing performances within matches. The International Basketball
Federation (FIBA) determined 13 variables, which are officially recorded in every
game. Most of the previous researches on performance analysis in basketball are based
on the statistical analysis of the official variables.
Trninic et al. (2002) analysed the differences between the performance of winning and
defeated top quality teams which played in final tournament of the European club
championships from 1992 to 2000. They found that defensive rebounds, field goal
percentage and free throw percentage were the critical factors that determined success.
Other researchers (Mendes and Janeira, 2001; Tsamourtzis et al., 2002) found that
defensive rebounding is the main factor that distinguishes winning and losing teams in
basketball. During the 1997 European Championship a significant difference between
winning and losing teams was determined in the variables successful field goal
attempts, assists and successful free throw attempts (Jukic et al. 2000). Lidor and Arnon
(2000) found that success cannot be described by shooting alone, but a team has to
demonstrate a high level in rebounding and passing as well as in shooting. They also
identified a significant correlation between the number of total rebounds and the number
of points scored by the team, and between the field goal percentage and the number of
assists. In a study of Sampaio et al. (2004) performance indicators discriminated the
teams by gender. Men’s teams were discriminated from women’s teams by their higher
percentage of blocks and successful two point field goals, and lower percentage of
steals. Choi et al. (2006) used Wilcoxon Signed Ranks tests to identify the critical
performance indicators in basketball. 10 basketball matches of the English basketball
league were analysed by game data sets and by quarter data sets. They found that
analysing performances by game data sets gives different valid performance indicators
than analysing by quarter data sets because the performance fluctuates within matches.
According to the opinion of Oliver (2004), four factors may be determinant to win
basketball games, the shooting percentage from the field, the offensive rebounds, the
turnovers and the number of free throw attempts.
According to Sampaio and Janeira (2003) performance indicators are influenced by
game location (home and away games) and game type (regular season or play-off). In
the 1997-98 and 1998-99 Portuguese Professional Basketball League away wins and
regular season profile were best discriminated by successful free throws. Play-off games
were best defined by offensive rebounding, home wins were best discriminated by
committed fouls. To analyse different type of matches Sampaio and Janeira used cluster
analysis to establish three different groups according to the game final score differences.
Tavares and Gomes (2003) identified that the points scored, the percentage of
successful free-throws, the number of fouls and offensive rating were the game
performance indicators that differentiated high performance level junior men teams. In a
study of Renao at al. (2006) identified game related statistics that differentiate winning
61

and losing teams at the U-16 European Championship in 2004. With the use of cluster
analysis the games were classified into three groups such as close games, balanced
games and unbalanced games. It was found that the number of successful 3 point field
goals and assists were significantly different when contrasting winning and losing teams
in close games.
Analysing all the games of any basketball tournament contains also the matches were
there is substantial difference between the performances of two teams. These games
increase the number of significant performance indicators when all the games are
considered.
In close basketball games coaches has big role and responsibility in formation of team
tactics. Results of analysing close games give useful information about the most
important elements that distinguish winning and losing teams. Knowing the crucial
performance indicators of close games allows coaches to prepare more detailed practice
and game plans and to build up the best winning strategy.
The main purpose of the current study is to find those critical performance indicators
that most distinguished between winning and losing teams at different type of matches
of the European Basketball Championship for men in 2007.
2. Methods
In this study the European Basketball Championship 2007 for men was analysed. The
tournament was held in Spain. Sixteen teams competed in four groups at the preliminary
round. Only the top three teams from each group joined to the qualifying round.
These 12 teams that had classified were divided into two groups of six teams. The best
four teams from each group moved to the quarterfinals and played for the 1st - 8th place.
The requiring data were gathered by using the official score sheets on the official
website of the tournament. The official performance indicators in basketball are number
of 3 points attempts, number of successful 3 points shots, percentage of successful 3
points shots, number of 2 points attempts, number of successful 2 points shots,
percentage of successful 2 points shots, number of free throw attempts, number of
successful free throws, percentage of successful free throws, offensive rebounds,
defensive rebounds, total rebounds, assist passes, personal fouls, steals, turnovers,
blocked shots and points scored by the team. All the 54 matches of the European
Basketball Championship 2007 were analysed.
Data processing was made by SPSS 15.0. Cluster analysis was used to classify the
matches into three types such as close games with final score differences between 1 and
9 points, balanced games (10-22 points) and unbalanced games (22-34 points
difference). Wilcoxon signed ranks tests were used to compare 18 performance
indicators between the winning and losing teams within each type of match. The level
of significance was set at p < 0.05.
62

3. Results
Analysing all the games of any basketball tournament contains also the matches were
there is substantial differences between the performances of two teams. These games
increase the number of significant performance indicators when all the games are
considered. The analysis of performance indicators in basketball can lead to the
identification of many significant performance indicators, not all of which can be
analysed in real time. Therefore, a smaller subset of critical performance indicators can
be identified by analysing close games only.
By using cluster analysis tight games were identified with final score differences below
9 points. The difference was between 10 and 22 points at balanced games and over 22
points at unbalanced games. There were 28 of the 54 matches that were close matches,
20 of them were balanced games and 6 of them unbalanced games. By analyzing all the
games of the European Basketball Championship 2007 (n=54) there were 13 significant
performance indicators for the full set of matches. Apart from the scored points the most
significant ones were the percentage of successful 3 point shots (p<0.001), the number
of successful free throws (p<0.001) and the defensive rebounds (p<0.001). The 13
significant performance indicators were reduced to 6 critical performance indicators
when only the close matches (n=28) were considered. At closed games the percentage
of successful free throws (p<0.001) seemed to be the most crucial performance indicator
that distinguished between winning and losing teams. Analysis of all the games and
tight matches are summarised in Table 1.
Table 1: Analysis of All the Games and Close Matches
Performance Indicator All matches (n=54) Close matches (n=28)
Winners
(mean+SD) Losers
(mean+SD) Winners
(mean+SD) Losers
(mean+SD)
Successful 2 point shots 19.0+4.2* 16.8+4.8 17.6+3.7 17.5+5.5
2 point attempts 36.5+5.7 36.5+5.4 36.3+6.3 36.8+5.7
%successful 2 point attempts 52.5+8.3** 46.9+9.9 48.6+7.3 49.6+10.4
Successful 3 point shots 8.0+2.6* 6.8+2.3 7.7+2.7 7.0+2.1
3 point attempts 20.6+4.4** 23.3+4.7 20.4+4.2* 22.9+4.5
%successful 3 point attempts 39.1+10.2*** 29.2+7.5 37.8+10.6** 30.3+7.2
Successful free throws made 17.0+5.3*** 12.7+5.5 17.8+5.7** 13.5+5.0
Free throw attempts 22.7+6.3** 19.2+7.8 23.3+6.6 20.6+5.9
%successful Free throws 74.7+10.0** 67.3+14.1 76.0+9-6*** 64.5+14.2
Offensive rebounds 9.7+3.7 10.8+3.5 10.0+4.2 10.6+3.5
Defensive rebounds 26.9+3.9*** 22.1+4.1 26.4+3.7** 22.8+4.7
Total rebounds 36.6+5.4** 32.9+5.2 36.4+5.9 33.5+5.1
Assist passes 12.4+3.5* 10.5+3.9 11.3+3.2 10.6+3.6
Personal fouls 21.0+4.6* 22.9+3.9 22.0+3.7 23.1+4.1
Turnovers 12.6+4.0 13.4+3.8 13.4+4.3 13.0+3.9
Steals 6.4+2.6 6.3+2.6 6.0+2.5 6.6+2.8
Blocked shots 2.7+1.6 2.4+1.3 2.6+1.5 2.6+1.4
Points 79.3+8.7*** 67.7+8.9 76.0+8.4*** 70.8+8.1
Significantly different to losing team: * p < 0.05, ** p < 0.01, *** p < 0.001
At balanced games (n=20) the analysis showed 8 significant performance indicators
(Table 2). The most significant ones were the 2 point shooting percentage (p<0.001) and
the 3 point shooting percentage (p<0.001).
63

The group of unbalanced games contained only 6 matches. Because of the small number
of matches the statistical analysis of unbalanced games explored only 5 significant
performance indicators that differentiate winners and losers, although substantial
differences can be seen at several performance indicators in Table 2.
Table 2: Analysis of Balanced and Unbalanced Matches
Performance Indicator Balanced matches (n=20) Unbalanced matches (n=6)
Winners
(mean+SD) Losers
(mean+SD) Winners
(mean+SD) Losers
(mean+SD)
Successful 2 point shots 19.8+3.9** 16.3+4.2 23.5+4.2* 14.8+1.8
2 point attempts 36.2+5.1 35.9+5.1 39.0+4.2 37.2+5.3
%successful 2 point attempts 55.5+6.8*** 45.3+9.4 60.2+8.6* 40.3+5.1
Successful 3 point shots 8.5+2.4* 6.9+2.3 8.0+3.0 6.2+3.4
3 point attempts 20.9+4.8 23.9+5.2 20.5+4.9 23.2+4.1
%successful 3 point attempts 40.9+9.6*** 28.7+7.1 38.9+11.5 25.4+10.2
Successful free throws made 16.8+5.0* 13.1+5.7 14.0+4.5 7.8+5.7
Free throw attempts 22.9+5.8* 18.9+8.8 18.7+5.6 10.3+10.2
%successful Free throws 72.9+10.3 72.0+10.5 75.3+11.8 64.9+21.9
Offensive rebounds 9.6+3.2 10.7+3.3 8.7+3.1 11.7+4.5
Defensive rebounds 26.9+4.4** 21.5+3.4 29.0+3.0* 20.7+3.1
Total rebounds 36.5+5.2 32.2+5.3 37.7+3.3 32.3+5.5
Assist passes 13.2+3.4* 10.8+3.7 14.8+3.6 8.8+6.5
Personal fouls 20.9+5.0 22.9+3.8 16.8+5.6 21.8+3.3
Turnovers 12.1+3.7 13.1+3.1 10.5+2.0* 16.2+4.5
Steals 6.5+2.3 6.0+2.5 8.3+3.4 6.2+1.3
Blocked shots 2.5+1.3 2.3+1.0 4.0+2.8 2.0+1.3
Points 82.4+7.3*** 66.8+7.0 85.0+8.2* 56.0+9.2
Significantly different to losing team: * p < 0.05, ** p < 0.01, *** p < 0.001
4. Discussion
The main aim of this study was to identify those critical performance indicators that
most distinguish between winning and losing teams, according to the game final score
differences. Analyzing all the 54 matches of the European Basketball Championship
2007 led to the identification of 13 significant performance indicators. The three most
significant ones were the shooting percentage of 3 point shots (p<0.001), the number of
successful free throws (p<0.001), and the number of defensive rebounds (p<0.001).
Analysis of all the games of the tournament contained also the easy winnings where
there were huge differences between the performances of the two teams and winner
teams often achieved better results in most of the notated performance indicators. These
13 significant performance indicators were reduced to 6 critical performance indicators
when only the close matches (n=28) were considered.
To prepare a team for basketball games, to build up the best tactics, to make good
decisions during a game, coaches need to know which elements of matches are the most
crucial ones. Especially at close games where there is small differences between the
performance of two teams. During close matches where the difference between the final
results of the two teams were 9 points or less the winning teams had significantly less 3
point attempts (p<0.05) with higher shooting percentage (p<0.01). It means that winner
teams in defence covered the most dangerous area close to the basket and forced the
opposite players to shoot from outside. The significantly higher number of defensive
rebounds (p<0.01) also mean that they kept attention to guard the area around the basket
64

with good box out and positioning. The higher number of successful free throws
(p<0.01) and the free throw percentage (p<0.001) also contributed to achieve a higher
number of scored points and consequently determined success. The importance of
defensive rebounds (Mendes and Janeira, 2001; Trninic et al., 2002; Tsamourtzis et al.,
2002) and free throws (Jukic et al. 2000; Oliver, 2004; Sampaio and Janeira, 2003;
Tavares and Gomes, 2003; Trninic et al., 2002) were highlighted by previous researches
also.
At balanced games (final score difference between 10 and 22 points) the better shooting
performance and defensive rebounding (p<0.01) led teams to victory. The significantly
higher number of defensive rebounds (p<0.01) and assist passes (p<0.05) and the better
shooting percentage could reflect that after good defensive rebounding the winner teams
made easy baskets from fast breaks. Tsamourtzis et al. (2005) identified that fast breaks
and their effectiveness are important factors to achieve the victory. There were
significant differences at 2 and 3 point shooting percentage (p<0.001). Beside the better
offensive performance it could be the result of the difference in quality of defence
between winning and losing teams.
Because of the small number of matches (n=6) the analysis of unbalanced games
explored only 5 significant performance indicators that differentiate winners and losers,
although relatively huge difference can be seen at several performance indicators in
Table 2. The reason of the huge difference between the 2 point shooting percentage
(p<0.05) and turnovers (p<0.05) can be explained with the difference between the
defensive performance of the winning and loosing teams.
Results obtained from balanced and unbalanced games show that winning teams made
better performance in most of the game statistics. At close games winning teams were
discriminated from losing teams by the 3 point performance, the free throws
performance and the defensive rebounding.
5. References
Bartlett, R. (2001). Performance analysis: can bringing together biomechanics and
notational analysis benefit coaches? International Journal of Performance
Analysis in Sport, 1, 122-126.
Choi, H.; O’Donoghue, P. and Hughes, M.D. (2006). A study of team performance
indicators by separated time scale real-time analysis techniques within English
national league basketball. In: Dancs, H.; Hughes, M.D. and O’Donoghue, P.
(eds.) Performance Analysis of Sport VII, Cardiff: CPA Press, UWIC, pp.
138-141.
Hughes, M.D. and Bartlett, R.M. (2002). The use of performance indicators in
performance analysis. Journal of Sport Sciences, 20, 739-754.
Hughes, M.D. and Franks, I.M. (1997) Notational analysis of sports. E and FN Spon,
London.
Hughes, M.D. and Franks, I. M. (2004). Notational analysis of sport: Systems for
better coaching and performance in sport. Second edition. London and New
York: Routledge.
Hughes, M.D. and Franks, I. M. (2008). The essentials of performance analysis – An
introduction. London: Routledge.
65

66
Jukic, I.; Milanovic, D.; Vuleta, D. and Bracic, M. (2000). Evaluation of variables of
shooting for a goal recorded during the 1997 European Basketball
Championship in Barcelona. Kinesiology (Zagreb), 32(2), 51-62.
Lidor, R. and Arnon, M. (2000). Developing indexes of efficiency in basketball: Talk
with the coaches in their own language. Kinesiology (Zagreb), 32(2), 31-41.
Mendes, L. and Janeira, M. (2001). Basketball performance - multivariate study in
Portuguese professional male basketball teams. In Hughes, M.D. and Tavares, F.
(eds.) Notational Analysis of sport IV, Cardiff: UWIC, pp. 103 - 111.
Reano, G. M. A.; Calvo, L. A.; Toro, O. E. (2006). Differences between winning and
losing under-16 male basketball teams. In: Dancs, H.; Hughes, M.D. and
O’Donoghue, P. (eds.) Performance Analysis of Sport VII, Cardiff: CPA
Press, UWIC. Pp. 142-149.
Oliver, D. (2004). Basketball on Paper – Rules and Tools for Performance Analysis.
Washington, D.C.: Brassey’s Inc.
Sampaio, J. and Janeira, M. (2003). Statistical analysis of basketball team performance:
understanding teams’ wins and losses according to a different index of ball
possessions. International Journal of Performance Analysis in Sport, 3(1),
40-49.
Sampaio, J.; Godoy, S.I. and Feu, S. (2004). Discriminative power of basketball game-
related statistics by level of competition and sex. Perceptual and Motor Skills,
99, 1231-1238.
Tavares, F. and Gomes N. (2003). The offensive process in basketball – a study in high
performance junior teams. International Journal of Performance Analysis in
Sport, 3(1), 27-33.
Trninic, S.; Dizdar, D. and Luksic, E. (2002). Differences between winning and
defeated top quality basketball teams in final of European club championship.
Collegium Antropologicum, 26(2), 521-31.
Tsamourtzis E., Salonikidis K., Taxildaris K., Mawromatis G. (2002). Technic and
tactical characteristics of winners and losers in basketball. Leistungssport, 1,
54-58.
Tsamourtzis E., Karypidis A., Athanasiou N. (2005). Analysis of fast breaks in
basketball. International Journal of Performance Analysis in Sport, 5(2), 17-
22.