Ball possession an important key performance indicator (KPIs) in men`s soccer; thus, coaches and analysts have been interested in the technical and physical parameters resulting in different possession statuses. In this regard, this study aims to examine the relationship between match KPIs and ball possession in elite men`s soccer, and identify the suitable analysis method (multiple linear regression, MLR vs. quantile regression, QR) to model this relationship in the Chinese Super League (CSL). A total of 474 performance records from 237 matches during the 2019 competitive season in the CSL were selected, and analysed using MLR and QR using the 10th, 25th, 50th, 75th, and 90th quantiles. Several technical (foul, pass, air duel, tackle, shot, and corner kick) and physical (sprint and high-speed running distance) KPIs were associated with ball possession. Compared to the MLR model, the QR model identified additional physical KPIs, which are sprint, high-speed running, and high-intensity running efforts, that influenced ball possession. For the situational variables, the quality of the opposition affects the ball possession from the 25th to 75th quantiles. Therefore, QR modelling is a potential tool for performance analysts and coaches to strategise and monitor different parameters in a match.
© Copyright 2021 International Journal of Performance Analysis in Sport. Taylor & Francis. Alle Rechte vorbehalten.