| Summary: | Due to the fast growth of advanced technology, the volume of sports data has increased, which has led to researchers' interest in studying the tennis data set. Tennis is an example of an individual sport where the opponent's skills and anticipation are just as significant as the player's personal qualities and abilities in determining the outcome of a match. Winning is greatly influenced by the experience as well as the comparative talent of the opponents. Hence, this study has two aims: first, to calculate the Relative Quality of each player and then classify the similar patterns of the female single players' performance based on their experience and Relative Quality, and second, to determine which weightings of the variables best discriminate between winners and losers for each cluster. The match statistics for four 2018 Grand Slam main-draw women's singles matches were collected from separate official tournament websites: the Australian Open, the French Open, Wimbledon, and the US Open, which consist of 1016 observations. The methods used in this study are hierarchical-based clustering analysis, which groups the players based on these two attributes, and discriminant analysis, which governs which attributes are best to discriminate among the winners and losers for each of the four clusters. Based on the cluster analysis results, four clusters were identified and designated as LEHRQ, LELRQ, HEHRQ, and HEHRQ. The outcome of the discriminant analysis shows that the performance of a tennis player is most likely reliant on holding their serve and remaining powerful to break their opponent's serve. Moreover, there are variations in the components among clusters. Identifying to which group a player belongs provides valuable assistance to both players and coaches in optimizing their training strategies and augmenting their performance by refining their serving and return abilities. © 2024 Author(s).
|
|---|