Dynamic research of tennis match based on machine learning

Authors

  • Xiangyu Wu
  • Haoxuan Xu
  • Xuan Zeng

DOI:

https://doi.org/10.62051/smfmby95

Keywords:

AdaBoost; Catboost; Wald-Wolfowitz Runs Test; Tennis; Momentum.

Abstract

The study employed the Adaboost algorithm to identify optimal indicators of momentum, demonstrating that momentum could significantly predict the likelihood of a player's subsequent wins or losses. This study offers a novel exploration into the dynamics of tennis matches, focusing on the concept of "momentum" and its impact on match outcomes. Through an analysis of the 2023 Wimbledon Men’s Singles final, the research demonstrates how momentum significantly influences the likelihood of a player's success, challenging traditional perceptions of match progression as merely a sequence of random events. By employing advanced statistical methods, including the Adaboost algorithm for identifying key indicators of momentum and the Wald-Wolfowitz Runs Test to examine the randomness of winning streaks, the findings reveal a substantial effect of momentum on the game's flow. The analysis further utilizes the CatBoost algorithm to pinpoint factors critical in shifting momentum between players, such as set scores, serving dynamics, and return depths. These indicators provide concrete evidence against the randomness of match dynamics, suggesting that strategic adjustments based on these factors can potentially alter the course of a match. The implications of this research extend beyond academic interest, offering practical insights for players and coaches in strategizing to leverage momentum for competitive advantage. By debunking the myth of randomness in tennis match outcomes, this study underscores the importance of psychological and strategic elements in sports performance, providing a foundation for further research and practical applications in the field of sports science and coaching.

Downloads

Download data is not yet available.

References

[1] Giles B, Kovalchik S, Reid M. A machine learning approach for automatic detection and classification of changes of direction from player tracking data in professional tennis [J]. Journal of sports sciences, 2020, 38 (1): 106 - 113. DOI: https://doi.org/10.1080/02640414.2019.1684132

[2] Asra T, Setiadi A, Safudin M, et al. Implementation of AdaBoost Algorithm in Prediction of Chronic Kidney Disease[C]//2021 7th International Conference on Engineering, Applied Sciences and Technology (ICEAST). IEEE, 2021: 264 - 268. DOI: https://doi.org/10.1109/ICEAST52143.2021.9426291

[3] Ke J, Zhengxuan Z, Zhe Y, et al. Intelligent islanding detection method for photovoltaic power system based on Adaboost algorithm[J]. IET Generation, Transmission & Distribution, 2020, 14 (18): 3630 - 3640. DOI: https://doi.org/10.1049/iet-gtd.2018.6841

[4] De La Rubia J M. Calculation of Two-Tailed Exact Probability in the Wald-Wolfowitz One-Sample Runs Test [J]. Journal of Data Analysis and Information Processing, 2024, 12 (1): 89 - 114. DOI: https://doi.org/10.4236/jdaip.2024.121006

[5] Abdullahi A. Ibrahim, Raheem L. Ridwan, Muhammed M. Muhammed, Rabiat O. Abdulaziz and Ganiyu A. Saheed, “Comparison of the CatBoost Classifier with other Machine Learning Methods” International Journal of Advanced Computer Science and Applications (IJACSA), 11 (11), 2020. DOI: https://doi.org/10.14569/IJACSA.2020.0111190

[6] Shahri N, Lai S B S, Mohamad M B, et al. Comparing the performance of AdaBoost, XGBoost, and logistic regression for imbalanced data [J]. Math Stat, 2021, 9: 379 - 85. DOI: https://doi.org/10.13189/ms.2021.090320

[7] Goyal A, Simonoff J S. Hot Racquet or Not an Exploration of Momentum in Grand Slam Tennis Matches [J]. arXiv preprint arXiv:2009.05830, 2020.

[8] Rusdiana A, Syahid A M, Kurniawan T. Momentum transfer analysis of upper body during backhand drive stroke in tennis [C]//AIP Conference Proceedings. AIP Publishing, 2023, 2646 (1). DOI: https://doi.org/10.1063/5.0113766

[9] Depken C A, Gandar J M, Shapiro D A. Set-level strategic and psychological momentum in best-of-three-set professional tennis matches[J]. Journal of Sports Economics, 2022, 23 (5): 598 - 623. DOI: https://doi.org/10.1177/15270025221085715

[10] Ma K. A real time artificial intelligent system for tennis swing classification[C]//2021 IEEE 19th World Symposium on Applied Machine Intelligence and Informatics (SAMI). IEEE, 2021: 000021 - 000026. DOI: https://doi.org/10.1109/SAMI50585.2021.9378695

Downloads

Published

12-08-2024

How to Cite

Wu, X., Xu, H. and Zeng, X. (2024) “Dynamic research of tennis match based on machine learning”, Transactions on Computer Science and Intelligent Systems Research, 5, pp. 1503–1511. doi:10.62051/smfmby95.