Using machine learning, the technical effectiveness of table tennis players and their winning probability were modeled. This study adopted a novel algorithm, SHapley Additive exPlanation (SHAP), to analyze the important features based on the gradient boosting + categorical features-tree-structured parzen estimator (Catboost-TPE) with the four-phase evaluation theory, an analysis framework in a table tennis match. A total of 110 singles’ matches (9536 rallies) were analyzed, and 59 male players’ winning rates from 2018 to 2022 were categorized into three levels (high, medium, low) by k-means cluster analysis. The results showed that Catboost-TPE has the best performance (MSE = 7.5e-05, MAE = 0.006, RMSE = 0.008, = 0.99, and adjusted = 0.989) among six hybrid machine learning algorithms. Using Catboost-TPE to calculate the SHAP value of each feature, the global interpretation and multiple local interpretations found that the performance of receive-attack and serve-attack phases have essential impacts on the winning probabilities in current matches. Besides, four and three phase evaluation theory are all import framework in table tennis match analysis. To further deepen the theoretical and applied value of the four-phase evaluation theory, this study derived the mathematical equations for converting indicators from the four-phase evaluation theory into the new three-phase evaluation theory. These results provided quantitative references to table tennis matches’ characteristics and winning phases. These methods used in the study can be widely applied to other sports performance analyses, and the equations derived in this study are also instructive for relative sports.
JordanMIMitchellTM.Machine learning: trends, perspectives, and prospects. Science2015; 349: 255–260.
2.
AlpaydinE.Introduction to machine learning. Cambridge, MA: MIT Press, 2020.
3.
GalliLGalvanGLevatoT, et al. Football: discovering elapsing-time bias in the science of success. Chaos Solitons Fractals2021; 152: 111370.
4.
HorvatTJobJ.The use of machine learning in sport outcome prediction: a review. Wiley Interdiscip Rev Data Min Knowl Discov2020; 10: e1380.
5.
McGrathJWNevilleJStewartT, et al. Cricket fast bowling detection in a training setting using an inertial measurement unit and machine learning. J Sports Sci2019; 37: 1220–1226.
6.
BarthMGuellichA.Non-linear association of efficiency of practice of adult elite athletes with their youth multi-sport practice. J Sports Sci2021; 39: 915–925.
7.
SchlembachCSchmidtSLSchreyerD, et al. Forecasting the Olympic medal distribution: a socioeconomic machine learning model. Technol Forecast Soc Change2022; 175: 121314.
8.
WuHLiZ.Research on technical diagnosis method for table tennis players. Int J Table Tennis Sci1992; 1: 99–103.
9.
HuangWShiZ.Three-stage index evaluation about Ding Ning’s table tennis playing. China Sport Sci Technol2016; 52: 126–130.
10.
ZhangHYangQ.Study on the expanded application of technical and tactical analysis of “three-phase-index evaluation” in table tennis matches. Sport Sci Res2016; 37: 61–66.
11.
ZhangXXiaoDZhouX, et al. The construction and application of dynamic three-phase method on table tennis technique and tactics. China Sport Sci Technol2018; 54: 80–83.
12.
ZhangHZhouZ.How is table tennis in China successful?Ger J Exerc Sport Res2019; 49: 244–250.
13.
ZhangHZhouZYangQ.Match analyses of table tennis in China: a systematic review. J Sports Sci2018; 36: 2663–2674.
14.
YangQZhangH.Construction and application of “four phase evaluation theory” technique and tactics for table tennis. J Tianjin Univ Sport2014; 29: 439–442.
15.
YangQZhangH.Application of BP neural network and multiple regression in table tennis technical and tactical ability analysis. J Chengdu Sport Univ2016; 42: 78–82.
16.
HuangWLuMZengY, et al. Technical and tactical diagnosis model of table tennis matches based on BP neural network. BMC Sports Sci Med Rehabil2021; 13: 1–11.
17.
XiaoYZhangH.Research report on the preparations of Chinese table tennis team for the Olympics: on the diagnostic model of table tennis competition based on artificial neural network. Sport Sci Res2008; 29: 19–22.
18.
ZhangXWuLLiZ, et al. A robust method to measure the global feature importance of complex prediction models. IEEE Access2021; 9: 7885–7893.
19.
PanGFuLChenQ, et al. Road safety performance function analysis with visual feature importance of deep neural nets. IEEE/CAA J Autom Sin2020; 7: 735–744.
20.
ParkMSSonHHyunC, et al. Explainability of machine learning models for bankruptcy prediction. IEEE Access2021; 9: 124887–124899.
21.
PalansooriyaKNLiJDissanayakePD, et al. Prediction of soil heavy metal immobilization by biochar using machine learning. Environ Sci Technol2022; 56: 4187–4198.
22.
RibeiroMTSinghSGuestrinC, et al. “Why should I trust you?” Explaining the predictions of any classifier. In: Kdd’16: Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining (KDD), San Francisco, CA, 13–17 August 2016, pp.1135–1144. New York, NY: ACM.
23.
Rodriguez-GalianoVFGhimireBRoganJ, et al. An assessment of the effectiveness of a random forest classifier for land-cover classification. ISPRS J Photogramm Remote Sens2012; 67: 93–104.
24.
LundbergSMLeeS-I. A unified approach to interpreting model predictions. In: 31st annual conference on neural information processing systems (NIPS), Long Beach, CA, 4–9 Decembe 2017, Advances in neural information processing systems 30 (NIPS 2017). New York, NY: ACM.
25.
SongHLiYZouX, et al. Elite male table tennis matches diagnosis using SHAP and a hybrid LSTM–BPNN algorithm. Sci Rep2023; 13: 11533.
26.
Beheshtian-ArdakaniASalehiMSharmaR.CMPN: modeling and analysis of soccer teams using complex multiplex passing network. Chaos Solitons Fractals2023; 174: 113778.
27.
RajbahadurGKWangSOlivaGA, et al. The impact of feature importance methods on the interpretation of defect classifiers. IEEE Trans Softw Eng2021; 48: 2245–2261.
28.
JamilMPhatakAMehtaS, et al. Using multiple machine learning algorithms to classify elite and sub-elite goalkeepers in professional men’s football. Sci Rep2021; 11: 22703.
29.
ChowdhuryASReehlSMKehn-HallK, et al. Better understanding and prediction of antiviral peptides through primary and secondary structure feature importance. Sci Rep2020; 10: 1–8.
30.
LeyCDominicyYBruneelW.Mutual point-winning probabilities (MPW): a new performance measure for table tennis. J Sports Sci2018; 36: 2684–2690.
31.
YuJGaoP.Interactive three-phase structure for table tennis performance analysis: application to Elite men’s singles matches. J Hum Kinet2022; 81: 177–188.
32.
GlazierPS.Game, set and match? Substantive issues and future directions in performance analysis. Sports Med2010; 40: 625–634.
33.
GómezMAGarcía-de-AlcarázAFurleyP.Analysis of contextual-related variables on serve and receiving performances in elite men’s and women’s table tennis players. Int J Perform Anal Sport2017; 17: 919–933.
34.
LvanekVĐukićBMikićB, et al. Effects of technical and tactical characteristics on the performance of the table tennis players. Facta Univ Ser Phys Ed Sport2018; 16: 157–166.
35.
ZhangHLiuWHuJ, et al. Evaluation of elite table tennis players’ technique effectiveness. J Sports Sci2014; 32: 70–77.
36.
HuangWZhangHLiuW.Evaluation of table tennis Olympic winner ZHANG Ji-ke’s technique effectiveness. China Sport Sci Technol2014; 50: 31–34+39.
37.
TamakiSYoshidaKYamadaK.A shot number based approach to performance analysis in table tennis. J Hum Kinet2017; 55: 7–18.
38.
ZhangHLiuWHuJ.Study on technique effectiveness of antagonistic sports event. China Sport Sci2015; 35: 44–49.
39.
KooTKLiMY.A guideline of selecting and reporting intraclass correlation coefficients for reliability research. J Chiropr Med2016; 15: 155–163.
40.
YangQLüY.Construction of the subsection theory for table tennis chop stroke. Sports Sci Res2020; 24: 44–52.
41.
BengioY.Practical recommendations for gradient-based training of deep architectures. In: MontavonGOrrGBMüllerKR (eds) Neural networks: tricks of the trade. 2nd ed.Berlin, Heidelberg: Springer, 2012, pp.437–478.
42.
Hoang-PhuongNLiuJZioE.A long-term prediction approach based on long short-term memory neural networks with automatic parameter optimization by Tree-structured Parzen Estimator and applied to time-series data of NPP steam generators. Appl Soft Comput2020; 89: 106116.
43.
HeFZhouJMoL, et al. Day-ahead short-term load probability density forecasting method with a decomposition-based quantile regression forest. Appl Energy2020; 262: 114396.
HancockJTKhoshgoftaarTM.CatBoost for big data: an interdisciplinary review. J Big Data2020; 7: 94.
46.
PunmiyaRChoeS.Energy theft detection using gradient boosting theft detector with feature engineering-based preprocessing. IEEE Trans Smart Grid2019; 10: 2326–2329.
47.
SongHLiYPanP, et al. Multilayer network framework and metrics for table tennis analysis: integrating network science, entropy, and machine learning. Chaos Solitons Fractals2025; 191: 115893.
48.
SongHLiYZhangZ, et al. Exploring the relationship among soccer-related knowledge, attitude, practice, and self-health in Chinese campus soccer education. iScience2024; 27: 109409.
49.
WenningerSLamesM.Performance analysis in table tennis-stochastic simulation by numerical derivation. Int J Comput Sci Sport2016; 15: 22–36.
50.
ZhangHHohmamA.Athletic diagnosis of table tennis matches through mathematic simulation. J Shanghai Univ Sport2004; 28: 68–72.
51.
ZhangHHohmamA.Theory and practice of performance diagnosis through mathematical simulation in ball game. China Sport Sci2005; 25: 39–44.
52.
PfeifferMZhangHHohmannA.A Markov chain model of elite table tennis competition. Int J Sports Sci Coa2010; 5: 205–222.
53.
YamamotoKNarizukaT.Examination of Markov-chain approximation in football games based on time evolution of ball-passing networks. Phys Rev E2018; 98: 052314.
54.
ZhouX.Explanation and verification of the rules of attack in table tennis tactics. BMC Sports Sci Med Rehabil2022; 14: 1–8.
55.
ĐokićZStraubGLanzoniIM, et al. Effects OF rule changes ON performance efficacy: differences between winners and losers table tennis players. Facta Univ Ser Phys Educ Sport2019; 17: 149–163.
56.
DjokicZMunivranaGLevajacD. Role of serve and return of serve at European games 2015 Table tennis tournament. In: 15th international table tennis sports science congress, Dusseldorf, Germany, 27–28 May 2017, Proceedings book of the 15th ITTF sports science congress, pp.242–246.
57.
DjokićZMalagoli LanzoniIKatsikadelisM, et al. Serve analyses of elite European table tennis matches. Int J Racket Sports Sci2020; 2: 1–8.
58.
SongHLiYFuC, et al. Using complex networks and multiple artificial intelligence algorithms for table tennis match action recognition and technical-tactical analysis. Chaos Solitons Fractals2024; 178: 114343.
Sáez MoralesGRuano AnocetoAGutiérrez PairolM. Las relaciones lógicas de oposición entre las acciones técnico-tácticas en el tenis de mesa. Podium2019; 14: 392–402.
61.
ChenJZhangH.Construction and application on “three phase evaluation method” in game analysis of elite badminton women’s singles. China Sport Sci Technol2015; 51: 63–67.
62.
LiLZhuJ. Construction and application of “3-stage index evaluation method” in men’s doubles badminton competition. J Anhui Sports Sci2022; 43: 31–35.
63.
GuoWZhaoG. Research on construction and application of tennis technical and tactical effectiveness evaluation model. J Henan Normal Univ2018; 46: 117–124.
