Resumen indicators for real-time feedback in men’s single tennis matches
The performance analysis of sport intends to investigate performance indicators in order to evaluate performances accurately and objectively (Hughes and Franks, 2004). The set of performance indicators to be used can be reduced to those that exhibit the property of construct validity. This allows the coach to focus on the most critical performance indicators during real-time match situations. Construct validity means that performance indicators discriminate between winning and losing performance. In tennis, the final match outcome does not always reflect the whole match performance and the coach need to identify different areas of the match at different points of the game that need to be addressed. The aim of this study, therefore, was to compare two different data sets; one containing whole matches and the other containing individual sets within matches of the singles event of the 2005 Wimbledon tennis championship. The winning and losing performances were compared using Wilcoxon Signed Ranks tests to compare winning and losing performances within each data set. In this study, 11 of 12 performance indicators significantly distinguished between the winning and losing players (P<0.05) within matches. However, for the data set using individual sets, all 12 performance indicators were significantly different between winning and losing players (P<0.05). Consequently, it is possible that the valid performance indicators have found in matches would not include the significant elements for individual set. Therefore, for the real-time feedback in a match, it is necessary to utilise the valid performance indicators for real-time analysis within the series of individual sets rather than to select the valid performance indicators in matches.
Within the performance analysis of sport, the consideration of discriminations between winning and losing teams has been utilised to identify the key performance indicators in particular sports such as football (Choi et al., 2006a), badminton (Blomqvist et al., 1998; Hong and Tong, 2000) and basketball (Tina, 1998; Evangelos et al., 2005; Tavares and Gomes, 2003). Especially, the selection of valid indicators for the applications of performance analysis techniques has also considered within the analysts or coaches’ perspectives. The performance indicators (Hughes and Bartlett, 2002) in the field of performance analysis of sport are valid elements to explain the performances of successful activities and events in matches. Thus, the outcome data collected by performance indicators (Hughes & Bartlett, 2002) are rationally useful for coaches in order to evaluate the performances and to conduct the further training plan for athletes. In last year, the performance indicators have selected and utilised in the analysis of performances in order to enhance the performances of sports (Hughes and Franks, 2004). The performance indicators selected have traditionally been from the whole matches’ data which were including the data in separated time scale such as the quarter data in basketball (Choi et al., 2006b). In the whole data of matches, the winners and losers might be found in the separated data that it is not perfectly presented the valid performance indicators within the data. In tennis, particularly, the performance indicators from whole matches would be not always reflected to the whole match performances and the coach need to identify different areas of the match at different points of the game that need to be addressed. Therefore, the consideration of valid performance indicators within the separated time scales such as sets in tennis is needed in order to discriminate the winning performances of the match.
The results of tennis matches in 2005 Wimbledon Tennis championship were used as subjects of this study that totally 128 matches were occurred within the tournament of 2005 Wimbledon Tennis Championship, but the data collected in this study was 126 matches that 2 games were missing on the official web site of Wimbledon Tennis championship (IBM Corp., 2005). Microsoft Excel 2003 package was used to copy the data from the web and then the SPSS package 12.0 was used to evaluate the data statistically. The data has copied into clipboard of MS office and then pasted onto the excel package in order to arrange the data efficiently. Whilst the data collected, some of data had to be corrected with data preview that a few data has not been collected in the regular format for this study and the regular format for this study has been decided orders of performance indicators. The performance indicators have compared in different data sets of this study are the below following. Total numbers of participations to Wimbledon tennis championship World ranks of players % of Successful serves Aces Double faults Unforced errors % of successful 1st serves % of successful 2nd serves Winners
- Received points
- % of net approaches
The data set of whole match has used that the indicators of winning and losing performance have collected without data transform. The data of individual set, however, has collected with the data transform that the winning and losing performances have intended to make groups in this study. And then the data determined by Wilcoxon Ranks test have compared with the difference data sets.
Totally, the numbers in subjects were 126 for the whole data set and 446 for the individual set data. The summary of the winning and losing performances were shown as Table 1 that the winning and losing in whole matches and sets are presented. The numbers of won within the whole data and individual data such as set data are not same scale. In other words, the winning performances which are based to identify the key performances in the sport are separately presented in the individual data (Table 1).
Table 1. The summary of the winning and losing performances.
The significant differences between the winning and losing performances have found that the % of successful serves (z=-0.56816, p<0.05) has found not significantly different within the Wilcoxon Signed Ranks test in the whole data set.
The significant differences’ values in the individual data sets, however, have found that all of performance indicators chosen in this study (p<0.05) are significantly different.
The significant differences between the winning and losing performances in the Wilcoxon Signed Ranks test have often used in order to determine the data statistically and to identify the key performance indicators (Choi, 2004; Choi et al., 2006c; Choi et al., 2006b; O’Donoghue, 1998). Especially, to find the key indicators of performances has emphasized in previous researches (Tina, 1998; Hughes & Bartlett, 2002). The previous method to find the key indicators of performances was the collected and determined data based on the whole matches. The difference of the key indicators between different data sets, however, has found that the key performance has not found in the whole data such as the percentage of successful serves was found in the individual data. Perhaps, the data range of subjects (Choi et al., 2006c; Choi et al., 2006b) would influence upon the results in order to identify the key performance indicators which would be used to further training plan by coaches. Additionally, the selection of key performance indicators (Choi et al., 2006b) in the researches have already emphasized the applications of the key performance indicators for real-time analysis. The applications of the key performance indicators in real-time analysis (Choi et al., 2006b) would be utilised to the real-time feedback in a match that the valid performance indicators are able to be used to identify the key performances in real-time.
The valid performance indicators found between winning and losing performances in the whole data set are total numbers of participations to Wimbledon tennis championship, world ranks of players, aces, double faults, unforced errors, % of successful 1st serves, % of successful 2nd serves, winners, received points, breaks and % of net approaches (p<0.05). The valid performance indicators found in individual data are the indicators including the % of successful serves (p<0.05). Thus, all indicators have found as the valid performance indicators in this study. For the further researches, the identifications of efficient usages to the coaching process in real-time (Choi et al., 2004; Choi et al., 2006c) are required.
- Blomqvist, M., Luhtanen, P., & Laakso, L. (1998). “Validation of a Notational Analysis System in Badminton” Journal of Human Movement Studies, 35, 137-150.
- Choi, H. J. (2004). A study of formations and roles of gender between winning and losing team in an international level of mixed double badmintons. MSc. Performance Analysis Dissertation, University of Wales Institute, Cardiff, Caridff.
- Choi, H. J., Kim, J. H., Kim, J. H., Hong, S. J., & Hughes, M. (2006a). “A study of valid contents for an evaluation of team performance in Soccer” World Congress of Performance Analysis of Sport VII. (In process)
- Choi, H. J., Kim, J. H., & Lee, W. H. (2004). “The analysing system of broadcasting in real time” P. O’Donoghue & M. Hughes (Eds.) Performance Analysis of Sport VI (pp. 245-252). Cardiff: Centre for Performance Analysis, UWIC.
- Choi, H. J., O’Donoghue, P., & Hughes, M. (2006b). “A Study of team performance indicators by separated time scale using a real-time analysis techniques within English national basketball league” World Congress of Performance Analysis of Sport VII, (In process).
- Choi, H. J., Reed, D., O’Donoghue, P., & Hughes, M. (2006c). The valid numbers of performance indicators for real-time analysis using prediction models within men singles in 2005 Wimbledon Tennis Championship. In World Congress of Performance Analysis of Sport VII. (In Process). Evangelos, T., Alexandros, K., & Nikolaos, A. (2005). “Analysis of fast breaks in basketball” eIJPAS, 5, 17-22.
- Hong, Y. & Tong, Y. M. (2000). “The playing pattern of the world’s top single badminton players in competition – a notation analysis” Journal of Human Movement Studies, 38, 185-200.
- Hughes, M. & Bartlett, R. (2002). “The use of performance indicators in performance analysis” Journal of Sports Sciences, 20, 739-754.
- Hughes, M. & Franks, I. M. (2004). Notational Analysis of Sport; Second Edition. London: Routledge. IBM Corp. (2005). The Championship, Wimbledon 2005 – Grand Slam Tennis – Official site by IBM. http://www.wimbledon.org [On-line].
- O’Donoghue, P. (1998). “Notational analysis of rallies in European club championship badminton” In M. Hughes & F. Tavares (Eds.), Notational Analysis of Sport IV (pp. 225-228). Portugal: Centre for Team Sports Studies, University of Porto.
- Tavares, F. & Gomes, N. (2003). “The offensive process in basketball – a study in high performance junior teams” eIJPAS, 3, 34-39.
- Tina, P. (1998). “Performance Indicators of basketball – a study of their impact for winnings” M. Hughes & F. Tavares (Eds.), Notational Analysis of Sport IV (pp. 123-126). Porto: Centre for Team Sports Studies. University of Porto.