Evaluating the Utility of Hierarchical Multiple Regression and Quantile Regression in Determining Critical Factors for Success in Elite Men's Basketball

This research investigates the effectiveness of stratified multiple regression (MLR) and quantile regression (QR) in identifying the key performance indicators (KPIs) that impact the outcomes of elite men's basketball games. Using performance data from the Paris 2024 Olympic Games, the study compares MLR and QR across various quartiles to explore both general trends and specific variations within different distributions. Important predictors such as inside-out scoring, centre-back scoring, three-point shooting, free-throw percentage, and pace of play were found to be significant in both models. However, QR uncovered additional insights not captured by MLR, including the relevance of Q50 Offensive Rebounds and Q75 Assists and Caps. QR also demonstrated a higher sensitivity in revealing the intricate, context-dependent relationships between KPIs and game outcomes, offering a more detailed understanding of how these factors fluctuate across different levels of competition. While MLR provided stable results, it was less effective at capturing this variation. This study underscores the importance of quantitative analysis in sports research, shedding light on subtle performance dynamics and offering valuable insights for optimizing team strategies, tactical choices, and training plans. By incorporating advanced statistical techniques, the research contributes to a deeper understanding of basketball performance and establishes a robust framework for future studies in sports analytics.