Water polo was the first Olympic team sport (in Paris, 1900), but research on water polo is fairly recent (43), and focused on specific aspects (in alphabetical order): biomechanical (9), kinanthropometric (1,2,6,13,27), physical fitness evaluation (13,14,48), physiological (20,36,43,47), and training (16,22,29). For example, some studies have shown that the front crawl technique used to move from one point to another can vary depending on the objectives that correspond to the particular moment of training or situation in the match (9). Thus, improvements in training also influence technical and tactical aspects. This may help in the design of more effective training in water polo (37). Also, scientific knowledge of the technical and tactical actions in collective sports leads to better understanding of the game. This facilitates the acquisition, improvement, and retention of the individual and collective skills needed for such sports (28).
The technique of “notational analysis” (17), for instance, has been applied to a variety of sports: basketball (42), football (21), rugby (33), tennis (19), and volleyball (18). In water polo too, the need for objective feedback on the performance of players in competition has led to recent analyses of this type (5,10,11,23–25,35,43,46). These studies quantify the technical and tactical actions of a game through game-related statistics based mainly on frequencies and effectiveness percentages (22). The results deriving from the analysis of these actions depend, however, on the nature of the situation, which makes reproducibility difficult (23). The skill level of the players naturally affects aspects of the game (29,30). In other sports, notational analysis has focused on determining the differences between winning and losing teams in terms of different game situations, principally (in alphabetical order): beginning match (starter/nonstarter) (15); championship standard (Olympic Games, etc.) (34); dominance of one team over another (42); final game outcome (win, draw, lose) (21); game location (home/away) (19); game type (regular season or play-off) (40); and player's position (41) and sex (male/female) (18). In water polo, the studies have focused on analyzing the differences in game statistics between winning and losing teams in terms of championship standard (23), offensive and defensive coefficients (5), play situations (24,25), player's position (35), relevance of goalkeeper (10), sex (11,46), and types of shot (16,44). Overall, the percentage of shooting success is a variable that distinguishes between winning and losing teams (25), especially for 5-m shots (11,24). In contrast, the penalty shot does not seem to be decisive for the outcome (11,44), although efficacy in this shot can be improved through psychological strategies (26). Counterattack actions are also more prevalent in winning teams (23,24,46). Nevertheless, defensive and offensive actions are also determinants of performance. Examples are goalkeeper saves (11) and blocks (46). In “extra man” situations, the extreme player (or wing) attains a greater proportion of converted shots than the goal-post players and outside players (35). With respect to the differences in game-related statistics between men and women, these seem to be more attributable to technical and tactical actions (7) than to physiological characteristics (20). Nonetheless, the methodological diversity of water polo studies has hindered understanding of the processes involved in the different types of matches (7).
A factor that has yet to be studied in water polo to distinguish between winning and losing teams is the phase of the competition. In basketball, for instance, regular season profiles are known to be best discriminated by successful free throws, whereas play-off profiles are best discriminated by offensive rebounds (40). This suggests that, in water polo too, coaches and players should be aware of these different profiles to increase specificity when it comes to game control. In particular, the games the teams play in the preliminary, classificatory, or final stages could be resolved with technical and tactical actions that differ according to what is required in that specific match. It seems reasonable to hypothesize that at different stages of a championship, the variables that distinguish and predict the winning teams are also different. Thus, the aims of this study were (a) to compare water polo game-related statistics by context (winning and losing teams) and phase (preliminary, classification, and semifinal/bronze medal/gold medal), and (2) to identify characteristics that mark the differences in performances for each phase.
Experimental Approach to the Problem
In water polo, notational analysis has yet to be applied to championship phase game-related statistics. The stimulus for this study was to use this technique to provide scientific support for the game-related statistics that are given to the coaches after each match of international championships. In particular, the aim was to give an answer to the question: What value do these game-related statistics have, and how can they be of use at each stage of a championship? Knowledge of which game statistics differentiate winning from losing teams in each phase of the competition might be an aspect for coaches to consider when planning their training and competitions. One finds in the literature studies in other sports (e.g., basketball) on the influence of the phase of the championship on the game statistics (40), but not in water polo. The present study was therefore an inferential analysis based on the examination of these game-related statistics according to the different phases of a championship.
Table 1 lists the dependent variables (game-related statistics) of the study. These game-related statistics are already of general use among water polo coaches and technicians and are those that have been used in earlier studies (10,11,25). The analysis specifically of goalkeeper-related game statistics is because the goalkeeper is the most important defensive player and has clearly different characteristics from those of the other players (36). The independent variable was context (winning or losing teams), and the analysis was carried out for each phase (preliminary, classificatory, and semifinal/bronze medal/gold medal). The preliminary phase is that which starts the competition, and in which the teams face each other in a group league format. The next phase is the classificatory phase in which the teams are paired off in a knockout format, with the winning team passing to the next round (usually, the last 16 or the quarter finals), whereas the loser has matches for the 5th to 16th place classification. The semifinal/bronze medal/gold medal phase includes the 2 semifinals of each championship, and the matches for the bronze and the gold medals. Other studies have used similar phases (44).
We analyzed the results and game-related statistics of 230 men's matches played in 5 International Championship (12th FINA World Championships 2007, Melbourne, Australia; 13th FINA World Championships 2009, Rome, Italy; 28th European Water Polo Championships 2008, Málaga, Spain; 29th European Water Polo LEN Championships 2010, Zagreb, Croatia; and 14th FINA World Championships 2011, Shanghai, China). All the championships were contested in open-air Olympic pools in summer months (July and/or August in the northern hemisphere).
All the results were retrieved from the box scores on the Official Website of OMEGA Timing (http://www.omegatiming.com/). These official box scores provide information on the game statistics analyzed both for each player individually and for the team collectively. The data were retrieved by one of the authors (M.M.), and entered manually into an Excel file. They were then subjected to a random check by another author (Y.E.) to detect possible errors. No informed consent was necessary because the information used is in the public domain on the Web site. Nevertheless, the study was approved by the Bioethics and Biosafety Committee of the University of Extremadura (Spain) and respected the principles of the Declaration of Helsinki.
Basic statistical descriptors (mean and standard deviation) were calculated by context (winning and losing teams) and phase (preliminary, classification, and semifinal/bronze medal/gold medal). To test the hypothesis that at the different stages of a championship the variables that distinguish between the winning and the losing teams are also different, 2 types of analysis were made: a chi-squared analysis and a discriminant analysis. Thus, chi-squared statistics were used to reveal the differences between the context (winning and losing teams) in each of the 3 phases. This is the recommended technique when the descriptors are discrete frequency response variables (31,32). The effect sizes of the differences were calculated (8). This was followed by a discriminant analysis, using the sample-splitting method according to context (winning and losing teams) and phase (preliminary, classification, and semifinal/bronze medal/gold medal). The criterion used to determine whether or not a variable was discriminatory was Wilks' lambda (λ), which measures the deviations within each group with respect to the total deviations. The sample-splitting method included initially the variable that best minimized the value of λ, provided that the value of F was greater than a certain critical value (F = 3.84, “include”). From that point on, the method combines the variables pairwise. The new variable is selected if λ is greater than the value of the input F. However, before introducing a variable one tries to eliminate some of those already selected, as long as the increase in the minimized λ is below a critical threshold (F = 2.71, “remove”). We thus calculated λ, the canonical correlation index (deviations of the between-group discriminant scores relative to the total deviations), and the percentage of correctly classified matches for each phase (preliminary, classificatory, and semifinal/bronze medal/gold medal). This methodological approach has been used in studies of other aquatic disciplines such as swimming (39). A p value of <0.05 was considered to be statistically significant. The statistical power was 0.99, 0.99, and 0.50 for the preliminary, classification, and semifinal/bronze medal/gold medal phases, respectively (alpha value of 0.05; effect size value of 0.03) (12). The statistical analysis was performed with the software package SPSS version 15.0 (SPSS, Inc., Chicago, IL, USA) and G*Power 3 (Institut für Experimentelle Psychologie, Universität Düsseldorf, Germany).
Table 2 presents the basic descriptors of the variables by context (winning/losing teams) in each phase. As the phase advances, the number of variables distinguishing between winners and losers declines (17 variables in the preliminary phase and 2 variables in the classificatory and in the semifinal/bronze medal/gold medal phases). The distinguishing variables were offensive, defensive, and mixed.
Table 3 gives the results of the discriminant analysis for each phase: Wilks' lambda, canonical correlation index, and percentage of teams correctly classified. The predictive models classified correctly 91% of the preliminary phase using 7 variables (shots, extra player shots, counterattacks, turnover fouls, sprints, goalkeeper-blocked shots, total possession time), 90% of the classificatory phase using 6 variables (shots, action shots, sprints, goalkeeper-blocked shots, goalkeeper-blocked action shots, total possession time), and 73% of the semifinal/bronze medal/gold medal phase using only 3 variables (sprints, goalkeeper-blocked penalty shots, and action shots).
To the best of our knowledge, this has been the first study to report the influence of game-related statistics on the outcome of men's water polo matches, followed by a discriminant analysis of those statistics to predict the winning/losing teams in the preliminary, classificatory, and semifinal/bronze medal/gold medal phases. It has examined in depth the application of notational analysis to men's water polo, using a sample (n = 230) far superior to almost the entirety of previous studies (between 11 and 99 matches). The study has also analyzed the data of the recent major international championships held between 2007 and 2011. As the competitive phase advanced, with the matches finally being between top teams, the teams still competing were found to present similar values of the variables studied. Thus, the number of distinguishing variables fell from 17 for the preliminary phase to 2 in the classificatory and medal phases. The predictive power of these variables also fell, with correct classification of 91% of the sample in the preliminary phase, 90% in the classificatory phase, and 73% in the semifinal/bronze medal/gold medal phase.
In the preliminary phase, there were 17 game-related statistics that distinguished between winning and losing teams. The winning teams had higher values for offensive actions (shots, action shots, center shots, extra player shots, 5-m shots, counterattacks, and assists), except for turnover fouls for which lower values represent better control of actions of attack by the winners. Also the winning teams had higher values for defensive actions (steals, blocked shots, goalkeeper-blocked shots, goalkeeper-blocked action shots, goalkeeper-blocked center shots, goalkeeper-blocked extra player shots, and goalkeeper-blocked 5-m shots). Two variables, timeouts and sprints, can be seen as neutral or mixed actions due to their both offensive and defensive nature. The higher value of timeouts obtained by the losing teams indicates less autonomy of the players to resolve game situations. Such autonomy has taken on especial importance with the reduction of possession time from 35 to 30 seconds because coaches now use greater defensive pressure (6). This obliges the players to be better prepared physically, and the offensive players to be faster in their decision making (38) so as to be able to successfully execute their offensive actions. The higher value of sprints obtained by the winning teams indicates their better “conditioning” (reaction time, swimming speed, and anticipation). This is consistent with previous studies that report that both offensive and defensive actions distinguish between the winning and the losing teams (5,11,46). In those studies although there are more offensive than defensive actions, in the present work there was no such difference. This large number of variables that distinguish between the teams in the preliminary phase may be due to the different levels of professionalism and competitive demands of the teams. Indeed, in this phase, countries with a long tradition in the sport compete with others that have only recently introduced it and therefore have less specialization in the collective game (23). Nonetheless, if a team's target is to reach the classification phase, it should consider improving its performance beyond just the variables “action shots” and “goalkeeper-blocked action shots” (25).
In the classificatory phase, there were only 2 game-related statistics, one offensive and one defensive, which distinguished between the winning and losing teams: action shots and goalkeeper-blocked action shots. These results are consistent with the findings of a similar study of men's matches in Olympic Games that found differences in shooting effectiveness (action shots) between winning and losing teams (11), and show the importance of shooting effectiveness for the outcome in international championships (25). It should be noted that the category “action shots” includes all field shots, and that the type of shot used and the zones from which they are made may vary depending on the level of competition (European League vs. Italian League) (23). The goalkeeper-blocked action shots reflected the importance of the goalkeeper's defensive actions for the final outcome of matches. This finding confirms that the goalkeeper is the most important defensive player, as is known for other team sports (36), without forgetting that his effectiveness also depends on the effectiveness of the defensive actions of the other players (23). This suggests specific physical training for this player centered on the improvement of lower-body power (36,43), segmental speed of the upper limbs, and anticipation (16). This variable (goalkeeper-blocked shots) is the one that distinguished between winning and losing teams in the semifinal/bronze medal/gold medal phase. Again, this highlights the importance of the goalkeeper's defensive action as a determinant of success in the men's game (11). In particular, the winning teams' goalkeepers obtained better values of efficacy in situations of equality, superiority, and defensive fallback (4). This efficacy can be improved by the goalkeeper always positioning himself between the ball and the goal in an area where the shooting angles will be narrowed (16). The fact of winning the ball in the initial sprint of each quarter also distinguishes between the winning and losing teams in the medal phase. These results suggest the need for swimming speed training for the players who contest the initial ball, especially given that in subsequent situations these sprints will be repeated in close succession with hardly any rest period (48). Also, for success in the fight for the initial ball, it would be advisable to train for power in the dive to begin the sprint (47) and for the technique of finally getting the ball (2). In this sense, lower-limb power has been shown to be a physical conditioning factor that distinguishes between elite and subelite players, so that it should be the target of specific training (14).
In general, in none of the 3 phases was there any difference between winning and losing teams in penalty shots, neither with the application of the old rules (5 m) nor the current rules (4 m). This is consistent with previous studies (11,44). The official data made available in every European Championship, World Championship, or Olympic Games do not include variables that quantify technical and tactical actions arising from situations of superiority. Analysis of game-related statistics in these situations would provide highly relevant information because 46% of the actions in a match correspond to such situations (16,46), with there being differences between elite and subelite teams (23). However, the rule changes in recent years mean that some of these situations might have changed.
In the preliminary phase, the 7 variables selected by the discriminant analysis model were shots, extra player shots, counterattacks, turnover fouls, sprints, goalkeeper-blocked shots, and total possession time, with 91% of the teams being correctly classified (winners and losers). These findings indicate the importance of a balance between offensive and defensive actions. Specifically, in offensive actions, the selection of the shots shows how important good shooting effectiveness is (11). It suggests that these teams have a greater ability to move the ball faster (23) to gain an advantageous position for shots at goal, thus increasing their effectiveness. In this sense, the speed of the shot is a determinant factor in scoring (1). This speed depends on the player's technique, coordination, overall strength, and grip strength (13). The work of the center forward in these situations is of vital importance (22) because he is the player who produces the greatest number of exclusions (23), and exclusions lead to power play situations in which extra player shots are made (the second ranking variable in distinguishing between winning and losing teams). Indeed, a high proportion of the total number of goals that a team scores emanate from the “extra man” situation (35). Hence, the attacking team players should develop their speed and passing execution with the aim of gaining advantageous situations of superiority that allow them to score goals. The average possession time per player and action is just over 3 seconds (16), whereas in a situation of superiority it is 30 seconds, and in a situation of equality 20 seconds (23). All this suggests that coaches might program specific training for these situations (35). With respect to counterattacks, these reflect the losing teams' poor defensive capabilities when they lose possession and then are slow to transition to defense, allowing the other team to execute shorter actions (24). Thus, swimming speed training with and without the ball would seem to be essential to develop these decisive skills (25). Likewise, swimming speed using free-style strokes should also be the target of specific training because 92% of displacements from one position in the pool to another use this style (16). The following variables selected by the model (turnover fouls, sprints, and goalkeeper-blocked shots) reflect the winning teams' ability to pressure the attacking team and to gain possession of the ball at the beginning of each period, and the effectiveness of their goalkeepers. Therefore, in addition to the physical training needed to improve swimming speed, lower-body power, and technique, coaches should consider strength training to hit, block, and push other players during game play (43) to improve these specific actions. Finally, the total possession time reflects the importance of controlling the ball during attacking plays, and suggests that winning teams perform more elaborate offensive actions (23).
In the classificatory phase, the variables selected by the discriminant analysis model were shots, action shots, sprints, goalkeeper-blocked shots, goalkeeper-blocked action shots, and total possession time, with 90% of the teams being correctly classified (winners and losers). The smaller number of predictive variables is indicative of greater equality between the teams in technical and tactical actions. This is consistent with other work which also found that the number of variables selected in the model decreases as the skill levels of the teams become more evenly matched (46). Hence, the winning teams probably generate more elaborate static attacks with the main objective being to get the ball moving quickly from one extreme to the other. The idea is to cause a momentary numerical imbalance in a specific space, so that a shot can be made with a greater chance of success (5). The speed with which the ball moves from one end to the other of the attack zone will allow these momentary situations of imbalance to be achieved. In the semifinal/bronze medal/gold medal phase, only 3 variables were selected by the model: sprints, goalkeeper-blocked penalty shots, and action shots. The fact of winning the first ball in each quarter has shown itself to be a decisive game statistic for the final result (3). This is possibly because of the great difference made to the situation at the beginning of the game if the first possession leads to a goal. It therefore seems appropriate for coaches to train their specialists who contest the initial ball of each quarter in speed swimming. Also, although the conversion of penalties has not been found to be a decisive game statistic in water polo (11,44), that the goalkeeper stops a penalty was indeed a distinguishing factor in the final phase of the championships. This is perhaps because of the different psychological predisposition of the team missing the penalty relative to the team that stopped it. Thus, given the distance from which the penalty is taken, it is necessary to use closed/open-skill situations (45) to work on both the physical skills that allow this situation to be satisfactorily resolved and the goalkeeper's capacity for anticipation (16). Likewise, there is a need for the goalkeeper to receive training that involves small amounts of high-intensity and medium duration exercises in which anaerobic metabolism predominates, the aim being, for example, to simulate man-down situations (36). Finally, the last variable selected by the model—action shots—indicates that more important than scoring goals in positions of superiority (extra player shots or counterattack) is to score them in situations of numerical equilibrium. This may be because these latter situations are common in the course of the evenly balanced games corresponding to the final phase of championships. For this reason, in addition to developing aerobic endurance—the average distance swum in a game is greater than 1300 m (16)—it is necessary to increase shooting speed in situations of equality because this has to be greater than in situations of superiority or counterattack (1). It has to be borne in mind that in counterattacks, the swimming speed of the players produces situations of imbalance in the transition (defense–attack) (23). Indeed, swimming speed is decisive in this game-related statistic (counterattack).
This study has some limitations. First, the distribution of the total number of matches analyzed into the different phases was naturally uneven (preliminary phase, n = 134; classificatory phase, n = 76; and semifinal/bronze medal/gold medal phase, n = 20). Nevertheless, there stand out as endowing the results with reliability and relevance to national players and coaches the large total size of the sample (n = 230) and the level of the competitions (the top international level). Second, in the preliminary phase, there usually occur some matches in which neither team any longer has the possibility of passing to the next round, a situation that could well influence the corresponding game-related statistics. Third, the discriminant analysis used post hoc prediction. In interpreting the results, one must bear in mind that this type of prediction usually gives higher values for the classification than a priori predictions.
This study has shown that water polo game-related statistics distinguish between the winning and the losing teams in each phase of an international championship. The distinguishing variables were both offensive and defensive: action shots, sprints, goalkeeper-blocked shots, and goalkeeper-blocked action shots. However, the number of these variables decreased as the phase becomes more demanding and the skill of the teams becomes more evenly matched. The predictive models based on the game-related statistics marked the differences in performance in all phases of the high-level championships studied (with 73% or more being correctly classified). Again, both defensive and offensive variables were selected by the models, suggesting the need for a good balance between the 2 groups of variables. Two variables distinguished the winning teams in the classificatory and final phases: sprints and action shots.
Knowledge of the characteristics of the water polo game-related statistics of the winning teams and their power to predict match outcomes will allow coaches to take these characteristics into account when planning training and match preparation. It should be borne in mind that the game-related statistics are given to the coaches at the end of each match. This study will thus benefit coaches in providing them with scientific support for a tool used in most international championships as post-game feedback. In particular, it would allow them to monitor the effectiveness of how they are managing these variables in international competitions. However, it is necessary to continue work with this type of analysis, including other aspects such as player position (center back, center forward, or wing) and game location (home/away), among others.
The study allows coaches and researchers to see that water polo game-related statistics in international competitions are different in each phase (preliminary phase, classificatory phase, and semifinal/bronze medal/gold medal phase) with there being a balance between offensive and defensive actions. Coaches and players can use these results as a referent against which to assess their performance and plan their team's training as appropriate for each phase of the championship. The variables that are most predictive of the outcome (winning vs. losing teams) are shots (preliminary and classificatory phase) and sprints (semifinal/bronze medal/gold medal phase). Two variables are predictive of the winning teams in the classificatory and final phases: sprints and action shots. Finally, there was another distinguishing variable in the final phase—goalkeeper-blocked penalty shots.
This information can be of help to coaches both in their technical and tactical decision making in the different phases of a competition and in individualizing players' training according to the physiological and conditioning requirements of the actions of the game that we have analyzed. Thus, the results suggest the need to develop swimming speed (initial sprint), shooting precision, throwing strength and speed (shot and action shot), and the goalkeeper's lower-body power and anticipation (goalkeeper-blocked penalty shots). In particular, speed training could be in the form of short-duration heats (16), shooting training in situations with and without opponents (24), and goalkeeper training in the form of small amounts of high-intensity exercises (36). Also, in the water, the goalkeeper could use a load of approximately 8 kg hung on a cord attached to a scuba diver's belt (3 × 3 × 4 with 10 seconds recovery between repetitions, and 3 minutes between sets) (14) so as to increase lower-body power. Nonetheless, it is necessary to consider all the scientific information together to improve the process of training water polo players (37).
The authors thank 2 anonymous reviewers who have helped to improve the quality of this article. The authors would also like to express their gratitude to the water polo coaches J. Madera and V. Denysenko for critically reviewing the article, which helped improve it with a view to its practical application. During the completion of this article, Y. Escalante and J. M. Saavedra were visiting researchers at the Cardiff Metropolitan University, Cardiff (United Kingdom), supported with grants awarded by European Social Fund (Una Manera de Hacer Europa) and the Autonomous Government of Extremadura (Gobierno de Extremadura) (PO10012 and GR10171, respectively). Finally, the authors wish to thank Dr. R. A. Chatwin for checking the English of the text.
1. Alcaraz PE, Abraldes JA, Ferragut C, Rodriguez N, Argudo FM, Vila H. Throwing velocities, anthropometric characteristics, and efficacy indices of women's European water polo subchampions. J Strength Cond Res 25: 3051–3058, 2011.
2. Aleksandrović M, Naumovski A, Radovanović D, Georgiev G, Popovski D. The influence of basic motor abilities and anthropometric measures on the specific motor skills of talented water polo players. Facta Univ Phys Educ Sport 5: 65–74, 2007.
3. Argudo FM, Arias JL, Ruiz E, Alonso JI. Effect of first ball possession on partial and final scores in 2003, 2005, and 2007 Water Polo World Championships. Percept Mot Skills 112: 349–352, 2011.
4. Argudo F, Gabaldón S, García P. Evaluación táctica cuantitativa del portero de waterpolo frente a lanzamientos en el X Campeonato del Mundo Barcelona-2003. Cuadernos Técnicos 2: 11–24, 2006.
5. Argudo FM, Roque JIA, Marín PG, Lara ER. Influence of the efficacy values in counterattack and defensive adjustment on the condition of winner and loser in male and female water polo. Int J Perform Anal Sport 7: 81–91, 2007.
6. Baramenti E, Platanou T. Description of the anthropometric characteristics of elite female water polo players according to competitive level and playing position. Int J Fitness 6: 11–18, 2010.
7. Canossa S, Garganta J, Argudo F, Fernandes R. Indicators of technical tactics in elite water polo games. Braz J Biomotricity 3: 209–219, 2009.
8. Cohen J. Statistical Power Analysis for the Behavioral Sciences. Hillsdale, NJ: Lawrence Earlbaum Associates, 1988.
9. De Jesus K, Figueiredo P, De Jesus K, Pereira F, Vilas-Boas JP, Machado L, Fernandes RJ. Kinematic analysis of three water polo front crawl styles. J Sports Sci 30: 715–723, 2012.
10. Enomoto I, Suga M, Takahashi M, Komori T, Minami T, Fujimoto H, Saito M, Suzuki S, Takahashi J. A notational match analysis of the 2001 women's water polo world championships. In: Biomechanics and Medicine in Swimming IX. Chatard JC, ed. Saint-Etienne, France: University of Saint Etienne, 2003. pp. 487–492.
11. Escalante Y, Saavedra JM, Mansilla M, Tella V. Discriminatory power of water polo game-related statistics at the 2008 Olympic Games. J Sports Sci 29: 291–298, 2011.
12. Faul F, Erdfelder E, Lang AG, Buchner A. G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods 39: 175–191, 2007.
13. Ferragut C, Vila H, Abraldes JA, Argudo F, Rodríguez N, Alcaraz PE. Relationship among maximal grip, throwing velocity and anthropometric parameters in elite water polo players. J Sports Med Phys Fitness 51: 26–32, 2011.
14. Gobbi M, D'ercole C, D'ercole A, Gobbi F. The components of jumps in expert and intermediate water polo players. J Strength Cond Res 2011. DOI:10.1519/JSC.0b013e318240ebf1
15. Gómez MA, Lorenzo A, Ortega E, Sampaio J, Ibáñez SJ. Game related statistics discriminating between starters and nonstarters players in Women's National Basketball Association League (WNBA). J Sports Sci Med 8, 278–283, 2009.
16. Hughes M, Appleton R, Brooks C, Hall M, Wyatt C. Notational analysis
of elite men's water-polo. In: Book of Abstracts of the 7th World Congress of Performance
Analysis. Dancs H, Hughes M, O'Donoghue P, eds. Szombathely, Hungary: International Society of Performance
Analysis of Sport, 2006. pp. 137–159.
17. Hughes M, Franks I. Notational Analysis
of Sport: Systems for Better Coaching and Performance
in Sport. London, United Kingdom: Routledge, 2004.
18. João PV, Leite N, Mesquita I, Sampaio J. Sex differences in discriminative power of volleyball game-related statistics. Percept Mot Skills 11: 893–900, 2010.
19. Koning RH. Home advantage in professional tennis. J Sports Sci 29: 19–27, 2011.
20. Konstantaki M, Trowbridge EA, Swaine IL. The relationship between blood lactate and heart rate responses to swim bench exercise and women's competitive water polo. J Sports Sci 16: 251–256, 1998.
21. Lago-Peñas C, Lago-Ballesteros J, Dellal A, Gómez M. Game-related statistics that discriminated winning, drawing and losing teams from the Spanish soccer league. J Sports Sci Med 9: 288–293, 2010.
22. Lozovina V, Pavii L, Lozovina M. Analysis of indicators of the load during the game in the activity of the centre in waterpolo. Nae More 51: 135–141, 2004.
23. Lupo C, Tessitore A, Minganti C, Capranica L. Notational analysis
of elite and sub-elite water polo matches. J Strength Cond Res 24: 223–229, 2010.
24. Lupo C, Tessitore A, Minganti C, King B, Capranica L. Notational analysis
of American women's collegiate water polo matches. J Strength Cond Res 25: 753–757, 2011.
25. Madera J, Colado JC, Tella V. Analysis of the type of attack and shoot in water polo. Nat Saltos Waterpolo 3: 7–12, 2007.
26. Marlow C, Bull SJ, Heath B, Shambrook CJ. The use of a single case design to investigate the effect of a pre-performance
routine on the water polo penalty shot
. J Sci Med Sport 1: 143–155, 1998.
27. Marrin K, Bampouras TM. Anthropometric and physiological changes in elite female water polo players during a training year. Serb J Sports Sci 2: 75–83, 2008.
28. McGarry T. Applied and theoretical perspectives of performance
analysis in sport: Scientific issues and challenges. Int J Perform Anal Sport 9: 128–140, 2009.
29. Mohr M, Krustrup P, Bangsbo J. Match performance
of highstandard soccer players with special reference to development of fatigue. J Sports Sci 21: 519–528, 2003.
30. Morya E, Bigatão H, Lees A, Ranvaud R. Evolving penalty kick strategies: World cup and club matches, 2000–2002. J Sports Sci 22: 512–513, 2004.
31. Nevill AM, Atkinson G, Hughes M, Cooper SM. Statistical methods for analysing discrete and categorical data recorded in sport performance
and notation analyses. J Sports Sci 20: 829–844, 2002.
32. Nevill A, Balmer N, Williams M. Crowd influence on decisions in association football. Lancet 353: 1416, 1999.
33. Ortega E, Villarejo D, Palao JM. Differences in game statistics between winning and losing rugby teams in the Six Nations Tournament. J Sports Sci Med 8: 523–527, 2009.
34. Papadimitriou K, Taxildaris K, Derri V, Mantis K. Profile of different level basketball centres. J Hum Mov Stud 37: 87–105, 1999.
35. Platanou T. Analysis of the extra man in water polo: A comparison between winning and losing teams and players of different playing position. J Hum Mov Stud 46: 205–211, 2004.
36. Platanou T. Physiological demands of water polo goalkeeping. J Sci Med Sport 12: 244–250, 2009.
37. Platonou T, Nikolopoulos G. Physiological demands of water polo games with different levels of competitiveness. In: Biomechanics and Medicine in Swimming IX. Chatard JC, ed. Saint Etienne, France: University of Saint Etienne, 2003, 493–497.
38. Royal KA, Farrow D, Mujika I, Halson SL, Pyne D, Abernethy B. The effects of fatigue on decision making and shooting skill performance
in water polo players. J Sports Sci 24: 807–815, 2006.
39. Saavedra JM, Escalante Y, Rodríguez FA. A multivariate analysis of performance
in young swimmers. Pediatr Exerc Sci 22: 135–151, 2010.
40. Sampaio J, Janeira M. Statistical analyses of basketball team performance
: Understanding teams wins and losses according to a different index of ball possessions. Int J Perform Anal Sport 3: 40–49, 2003.
41. Sampaio J, Janeira M, Ibáñez S, Lorenzo A. Discriminant analysis
of game-related statistics between basketball guards, forwards and centres in three professional leagues. Eur J Sport Sci 6: 173–178, 2006.
42. Sampaio J, Lago C, Drinkwater EJ. Explanations for United States of America's dominance in basketball at Beijing Olympic Games (2008). J Sports Sci 28: 147–152, 2010.
43. Smith HK. Applied physiology of water polo. Sports Med 26: 317–334, 1998.
44. Smith HK. Penalty shot
importance, success and game context in international water polo. J Sci Med Sport 7: 221–225, 2004.
45. Starkes JL, Ericsson KA. Expert Performance
in Sports: Advances in Research on Sport Expertise. Champaign, IL: Human Kinetics, 2003.
46. Takagi H, Nishijima T, Enomoto I, Stewart AM. Determining factors of game performance
in the 2001 world Water Polo Championships. J Hum Mov Stud 49: 333–352, 2005.
47. Tan F, Polglaze T, Dawson B. Activity profiles and physical demands of elite women's water polo match play. J Sports Sci 27: 1095–1104, 2009.
48. Tan FH, Polglaze T, Dawson B. Reliability of an in-water repeated-sprint test for water polo. Int J Sports Physiol Perform 5: 117–120, 2010.