The purpose of this study was to compare the differences of identification of performance indicators which were distinguishing the winning and losing performances in Tennis Gland Slam competitions. Especially, the consistencies of identification on Performance Indicators through statistical comparisons using Wilcoxon Signed Ranks tests were considered. For this study, 1,929 matches and 6,888 sets were selected as subjects in 4 Grand Slam tennis competitions (from 2005 to 2011). Totally, 10 performance indicators were selected from the official stats of 4 Grand Slam tennis championships as independent variables for this study that they were % of 1st serves succeed, aces, double faults, unforced errors, winning % of successful 1st serve, winning % of successful 2nd serve, winners(included serves), receiving points won, break points conversions and % of net approaches. As results of the study, there were 3 summarised findings as following below; Firstly, there were significant differences of all 10 performance indicators between winning and losing where the data were arranged as match results (p<.05). Likely, there were also significant differences of 10 performance indicators between winning and losing where the set data were used for comparisons (p<.05). Secondly, there were significant differences on 5 performance indicators distinguishing winning and losing where matches' data were used in the comparisons between championships separately (p<.05). The performance indicators were winning % of successful 1st serve, winning % of successful 2nd serve, winners (included serves), receiving points won, and break points conversions. Thirdly, there were significant differences on only 2 performance indicators distinguishing winning and losing where sets' data were used in the comparisons between championships separately (p<.05). The performance indicators were % of 1st serves and receiving points won. Consequently, this study was shown that there were different results of distinguishing between winning and losing performances in massive size of data. However, there were also similar findings between championships that it would be Key Performance Indicators where the data was critically huge. It intended that those Key Performance Indicators would be used for prediction modelling, coaching feedback and comparisons of winning and losing performances individually. Further researches are required that the development of the systematic structures of identification for KPI, consideration of identifications of Key Performance Indicators for double games and application for a real-time analysis system.
© Copyright 2012 World Congress of Performance Analysis of Sport IX. Veröffentlicht von University of Worcester. Alle Rechte vorbehalten.