Investigating and Forecasting Momentum Shift Effects on Strategy Development in Sports Competitions

Wenqian Sun; Heng Hua

doi:10.62051/avjdhz14

Authors

Wenqian Sun
Heng Hua

DOI:

https://doi.org/10.62051/avjdhz14

Keywords:

Momentum-Shifts prediction; Correlation metrics; Light Gradient Boosting Machine algorithm; Error tests; Features importance analysis.

Abstract

Momentum shifts in competitive sports have a significant impact on game outcomes and strategy development, and despite their widely recognized importance, there is a relative lack of relevant systematic research. In this regard, this study quantified the relationship between momentum shifts and player success, using the metrics analysis method to analyze their correlation and stochasticity. The relevant metrics mainly comprised Spearman rank correlation analysis, and P-value test calculations, which yielded a strong correlation of 0.84 and a P-value less than 0.05. Furthermore, to predict and visualize possible shifts, our study used the LightGBM algorithm to build a predictive model, where we considered shifts to represent changes in strengths and weaknesses between players i.e. moments of change. Afterward, the researchers did an error test for our model, which had a low mean squared error of 0.0019. Finally, this research performed a features importance analysis to identify the most relevant factors such as GPCI, and SGD, and based on the results four ways were suggested for players to deal with new opponents.

Downloads

Download data is not yet available.

References

[1] Iso-Ahola, S.E. and W. J. Blanchard, Psychological momentum and competitive sport performance: A field study. Perceptual and Motor Skills, 1986. 62 (3): p. 763 - 768.

[2] Ke, G., et al., Lightgbm: A highly efficient gradient boosting decision tree. Advances in neural information processing systems, 2017. 30.

[3] Wang, Y. and T. Wang, Application of improved LightGBM model in blood glucose prediction. Applied Sciences, 2020. 10 (9): p. 3227.

[4] Chen, C., et al., LightGBM-PPI: Predicting protein-protein interactions through LightGBM with multi-information fusion. Chemometrics and Intelligent Laboratory Systems, 2019. 191: p. 54 -64.

[5] Sun, X., M. Liu and Z. Sima, A novel cryptocurrency price trend forecasting model based on LightGBM. Finance Research Letters, 2020. 32: p. 101084.

[6] Razali, M.N., et al., Football matches outcomes prediction based on gradient boosting algorithms and football rating system. Human Factors in Software and Systems Engineering, 2022. 61: p. 57.

[7] Vu, Q.H., et al. A combination of general and specific models to predict victories in video games. in 2021 IEEE International Conference on Big Data (Big Data). 2021: IEEE.

[8] Palut, Y. and P. Zanone, A dynamical analysis of tennis: Concepts and data. Journal of sports sciences, 2005. 23 (10): p. 1021 - 1032.

[9] Tang, C., N. Luktarhan and Y. Zhao, an efficient intrusion detection method based on LightGBM and autoencoder. Symmetry, 2020. 12 (9): p. 1458.

[10] Briki, W., et al., How psychological momentum changes in athletes during a sport competition. Psychology of Sport and Exercise, 2013. 14 (3): p. 389 - 396.