Uncovering the Secrets of Momentum Hidden in the Game of Tennis

Haoqian Huo
Proceedings of 2025 2nd International Conference on Machine Learning and Intelligent Computing, PMLR 278:717-725, 2025.

Abstract

Advances in sports technology have had a profound impact on the tennis game, not only improving the fairness and enjoyment of the game, but also changing the way players are trained and performance analyzed. This article builds a momentum evaluation model and deeply explores the impact of momentum on game results based on the game data set. Before building the model, we cleaned and standardized the given data and classified it into four parts: fatigue level, psychological state, personal technical ability, and real-time conditions. Preliminary preparations were made for the construction and solution of the model. We developed a comprehensive tennis player “momentum” evaluation model using Logistic-LGBM, employing point granularity and five-fold cross validation. This model dynamically assesses and captures real-time changes in player momentum. Our real-time visualization during the 2023 Wimbledon men’s singles final revealed observable momentum trends. However, due to the complexity of factors affecting player scores, not accounting for them introduced significant noise and disrupted player scores. This insight serves as a foundation for refining the model.

Cite this Paper

BibTeX
@InProceedings{pmlr-v278-huo25a, title = {Uncovering the Secrets of Momentum Hidden in the Game of Tennis}, author = {Huo, Haoqian}, booktitle = {Proceedings of 2025 2nd International Conference on Machine Learning and Intelligent Computing}, pages = {717--725}, year = {2025}, editor = {Zeng, Nianyin and Pachori, Ram Bilas and Wang, Dongshu}, volume = {278}, series = {Proceedings of Machine Learning Research}, month = {25--27 Apr}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v278/main/assets/huo25a/huo25a.pdf}, url = {https://proceedings.mlr.press/v278/huo25a.html}, abstract = {Advances in sports technology have had a profound impact on the tennis game, not only improving the fairness and enjoyment of the game, but also changing the way players are trained and performance analyzed. This article builds a momentum evaluation model and deeply explores the impact of momentum on game results based on the game data set. Before building the model, we cleaned and standardized the given data and classified it into four parts: fatigue level, psychological state, personal technical ability, and real-time conditions. Preliminary preparations were made for the construction and solution of the model. We developed a comprehensive tennis player “momentum” evaluation model using Logistic-LGBM, employing point granularity and five-fold cross validation. This model dynamically assesses and captures real-time changes in player momentum. Our real-time visualization during the 2023 Wimbledon men’s singles final revealed observable momentum trends. However, due to the complexity of factors affecting player scores, not accounting for them introduced significant noise and disrupted player scores. This insight serves as a foundation for refining the model.} }
Endnote
%0 Conference Paper %T Uncovering the Secrets of Momentum Hidden in the Game of Tennis %A Haoqian Huo %B Proceedings of 2025 2nd International Conference on Machine Learning and Intelligent Computing %C Proceedings of Machine Learning Research %D 2025 %E Nianyin Zeng %E Ram Bilas Pachori %E Dongshu Wang %F pmlr-v278-huo25a %I PMLR %P 717--725 %U https://proceedings.mlr.press/v278/huo25a.html %V 278 %X Advances in sports technology have had a profound impact on the tennis game, not only improving the fairness and enjoyment of the game, but also changing the way players are trained and performance analyzed. This article builds a momentum evaluation model and deeply explores the impact of momentum on game results based on the game data set. Before building the model, we cleaned and standardized the given data and classified it into four parts: fatigue level, psychological state, personal technical ability, and real-time conditions. Preliminary preparations were made for the construction and solution of the model. We developed a comprehensive tennis player “momentum” evaluation model using Logistic-LGBM, employing point granularity and five-fold cross validation. This model dynamically assesses and captures real-time changes in player momentum. Our real-time visualization during the 2023 Wimbledon men’s singles final revealed observable momentum trends. However, due to the complexity of factors affecting player scores, not accounting for them introduced significant noise and disrupted player scores. This insight serves as a foundation for refining the model.
APA
Huo, H.. (2025). Uncovering the Secrets of Momentum Hidden in the Game of Tennis. Proceedings of 2025 2nd International Conference on Machine Learning and Intelligent Computing, in Proceedings of Machine Learning Research 278:717-725 Available from https://proceedings.mlr.press/v278/huo25a.html.

Related Material