Unveiling Momentum Dynamics in Tennis

Abstract

This study investigates the impact of momentum in tennis competition through a multifaceted approach. We introduce the Momentum Balance Index (MBI), derived via Principal Component Analysis (PCA) from 42 indicators, identifying 13 significant factors influencing player momentum. MBI analysis reveals a strong correlation between peak momentum and score leads, indicative of player condition and winning likelihood. Spearman correlation analysis validates momentum's link to enhanced performance and victory probabilities, challenging randomness in match outcomes. Additionally, an ARIMA-based prediction model, refined through logistic regression, forecasts momentum shifts within matches, offering real-time tactical insights for coaches. Receiver Operating Characteristic (ROC) analysis confirms the model's robust predictive capability, with AUC values exceeding 0.5 across three matches. This comprehensive approach sheds light on momentum dynamics in tennis, providing valuable insights for coaching and strategic decision-making.