Visualisation of Athlete Performance Based on Markov Chains and EWMs

Abstract

The aim of this study is to synthesize the effects of several factors on the performance of sports players, such as skill level, score differences, and psychological factors, and to establish a corresponding evaluation index system. To address this goal, this paper propose a performance evaluation framework to quantitatively assess the state of sports players. In this paper, this paper take the performance scores and winning rate in the game flow as the evaluation indexes, and firstly, this paper measure the performance scores of sports players by calculating nine indexes such as the winning rate of the first serve, the winning rate of the second serve, etc., and use the entropy weighting method to determine the weights of each index. Then, this paper used Markov model to calculate the winning rate of each point, combined with the performance score and the winning rate to establish the regression equation, the results show that the coefficient of determination are close to 1, indicating that the model has a good fitting effect. In this paper, this paper continue to calculate the winning percentage of a player at a certain point of winning this game through Markov chain. This study provides a comprehensive quantitative method for evaluating the performance of sports players, which helps to more accurately determine the players' status and level, and provides a scientific basis for training and competition.