Multi-Factor Evaluation of Performance and Momentum Prediction Models in Tennis Matches

Abstract

With the increasing popularity and competitive nature of tennis, there is a growing demand for comprehensive analytical methods that can evaluate player performance and predict momentum shifts during matches. Traditional statistical approaches focus mainly on the result of the game, but they often fail to capture the complex dynamics that influence player performance throughout a game. In response to this need, our study proposes an integrated approach that combines the Analytic Hierarchy Process (AHP) and the Seasonal AutoRegressive Integrated Moving Average with exogenous factors (SARIMAX) model to comprehensively evaluate player performance and predict momentum. The AHP model considers critical factors such as won sets, won games, fatigue, and winning streaks, while the SARIMAX model is used to predict momentum changes based on match data. Our findings demonstrate a significant positive correlation between momentum and player performance, suggesting that momentum shifts can strongly influence the outcome of tennis matches. It also provides valuable insights into the dynamics of tennis performance and offers practical strategies for enhancing player performance through momentum analysis.