Throwing performance and test-retest reliability in Olympic female water polo players

Shooting performance is critical to successful water polo performance, requiring high levels of speed and accuracy. However, shooting speed and accuracy performance of elite female water polo players has not been described in detail nor has the reliability of accuracy measures been reported. Consequently, the aim of this study was to describe shooting performance in elite female water polo players and compare the test-retest reliability of the currently available accuracy measures. Ten Olympic female water polo players were tested on 3 occasions before Olympic competition for maximal throwing speed and accuracy (total error and hit percentage) toward various target locations, with and without a goalkeeper (GK) present. The current participants (all Olympic athletes) achieved higher speeds (16.8 m·s-1), higher hit percentages (45.3%) of a 20-cm sniper net target, and lower errors (20.7 cm) than any other female water polo players investigated previously. Performance was similar across the different target locations; however, both speed (15.2 vs. 14.8 m·s-1; p ≤ 0.05) and accuracy (50.2 vs. 37.7%; p ≤ 0.05) were reduced in the presence of a GK. Speed and total error was similar across the 3 testing sessions; however, hit percentage was statistically significantly higher in the final session (58.8% vs. 40.4 and 36.4%). Maximal throwing speed showed high levels of test-retest reliability (0.96). Total error (intraclass correlation [ICC] = 0.79) had slightly greater reliability than hit percentage (ICC = 0.73) overall. Hit percentage (ICC = 0.82) showed slightly greater reliability than total error (ICC = 0.72) without a GK present, whereas total error (ICC = 0.79) showed much greater reliability than hit percentage (ICC = 0.34) when a GK was present. This study benchmarks the high degree of throwing speed and accuracy required at the Olympic level in female water polo players. Although resource intensive, total error is a reliable measure of accuracy that allows for the nature of error to be described in detail, providing a great level of insight into performance. Hit percentage, while easier to obtain, should only be used when a GK is not present and in situations where describing the nature of error does not add value to coaching practices.