Abstract

Purpose: This study aims to model and predict team performance outcomes during the 2023 FIFA Women’s World Cup by integrating a supply chain perspective into sports analytics. The research addresses a gap in the literature by focusing on the distribution and flow of player and team performance data, which is critical for optimizing decision-making in modern football management, media reporting, and logistics. Research design, data and methodology: Data were collected from 60 matches involving 16 national teams, incorporating detailed player-level performance metrics. Five feature engineering strategies and seven machine learning algorithms were compared through rigorous nested cross-validation techniques. This analytical pipeline represents a structured supply chain of data processing and model evaluation. Results: Among the tested algorithms, logistic regression consistently outperformed others, achieving 99% accuracy under nested cross-validation and 95.8% accuracy on an independent test set. This indicates robust generalizability and practical reliability in predicting match outcomes. Conclusions: This study contributes a comprehensive machine learning framework tailored for women’s football, emphasizing the importance of data distribution efficiency. The results offer practical implications for enhancing performance forecasting, supporting coaching strategies, streamlining media reporting, and improving event supply chain operations in international sports tournaments