What is it about?

This study analyzes the effect of momentum on the outcome of tennis matches, using the 2023 Wimbledon men’s singles final as an example. The match data was preprocessed, including invalid sample treatment, outlier treatment, data normalization, feature screening and the innovative combination of wavelet analysis and Bayesian Optimization are used in data preprocessing. An ARIMA model was constructed to analyze the effect of match duration on the outcome. The study found that the longer the duration of the match, the more difficult it is to predict. A Pearson correlation analysis model was used to determine the correlation between the distance moved by players and wins and losses. The random forest model analysis showed that serve order, match point, and set point were the key factors influencing the match outcome. Additionally, Djokovic’s momentum changes were evaluated through hierarchical analysis, and the model’s accuracy was verified using the 2022 Qatar World Cup Final as an example. The study proposes a coaching strategy based on risky decision-making to help coaches effectively coach athletes during matches.

Featured Image

Why is it important?

This study examined the effects of athletes’ mental states, particularly their changes in momentum dominance during competition,on the outcomes of sports competitions. A variety of data preprocessing methods were used prior to constructing the mathematical model. These methods included the innovative use of wavelet analysis combined with Bayesian optimization on a random forest training set to ensure the accuracy of the data and conclusions.In this paper, the correlation between match duration, distance covered, serve side, depth of serve, and match outcome was investigated using a variety of statistical methods such as ARIMA,Pearson’s correlation, analysis of covariance, and random forest regression. The results of the study showed that as the duration of the match increased, the match became more intense, players covered more distance and scored more points. The study showed that the side of the serve had a significant effect on the outcome of the match. The random forest model identified key factors such as points earned, match wins and losses. The paper also introduced momentum analysis and confirmed the findings through hierarchical analysis. Finally, Markov chains were used as a basis for risky decision making, thus completing the conclusions of the thesis. This study quantitatively investigated the effects of athletes’ mental states and motivation on athletic performance, providing valuable information for coaches and athletes.

Perspectives

In recent years, the importance of an athlete’s mental state in sports competition has received increasing attention. A study by Patrick R. Young and Erin L. Knight [1] found that adventurous athletes use mental skills to cope with challenges during competition. On the other hand, a study by Ming-Hong Lin and Mei-Hua Huang [2] explored the effects of athletes’ decision-making styles, mental skills, and sport anxiety on the outcome of competition. However, there is a lack of comprehensive research on the effects of athletes’ mental states on competition outcomes. Based on the existing studies, this paper attempts to explore the effects of athletes’ mental states, race motivation and individual players’ motivation on race outcomes, and to establish a corresponding prediction model. This study aims to provide reference for coaches and athletes’ psychological training and game decision-making through empirical analyses, and at the same time to lay the foundation for an in-depth study of the relationship between athletes’ psychology and sports performance. The innovative combination of wavelet analysis, Bayesian Optimization and random forest algorithm is used in data preprocessing. The paper considers special factors when treating enormous datasets, thus data is accurate and the conclusions are rigorous. In this paper, ARIMA model, Pearson correlation analysis model, covariance test model and random forest regression model are used to explore the effects of game time, running distance of both sides, serving height and serving depth, as well as both sides’ scores and wins and losses on the results of the game, respectively. Finally, the results of the models were combined and the effects of match momentum and individual momentum on match outcomes were explored according to Markov chain theory

Yuefan Wang
Guangzhou City University of Technology

Read the Original

This page is a summary of: A study of multifactor quantitative analysis of sports game trends, May 2024, ACM (Association for Computing Machinery),
DOI: 10.1145/3685088.3685179.
You can read the full text:

Read

Resources

Contributors

The following have contributed to this page