Optimizing wheelchair basketball lineups: A statistical approach to coaching strategies

Abstract

This study was designed to support the tactical decisions of wheelchair basketball (WB) coaches in identifying the best players to form winning lineups. Data related to a complete regular season of a top-level WB Championship were examined. By analyzing game-related statistics from the first round, two clusters were identified that accounted for approximately 35% of the total variance. Cluster 1 was composed of low-performing athletes, while Cluster 2 was composed of high-performing athletes. Based on data related to the second round of the Championship, we conducted a two-fold evaluation of the clusters identified in the first round with the team’s net performance as the outcome variable. The results showed that teams where players belonging to Cluster 2 had played more time during the second round of the championship were also those with the better team performance (R-squared = 0.48, p = 0.035), while increasing the playing time for players from Classes III and IV does not necessarily improve team performance (r2 = -0.14, p = 0.59). These results of the present study suggest that a collaborative approach between coaches and data scientists would significantly advance this Paralympic sport.