The aim of this study was to examine the relationship between playing style variability and within-season team match performance outcomes. We examined 21,708 matches played in the top five leagues in Europe from the 2014/15 to 2019/20 season. Playing styles used by each team in each match were derived through a Gaussian-Mixture Model clustering of 31 match actions. The relationship between playing style choice and match outcome was also analysed using Multinomial Logistic Regression. However, team and opposition playing style predicted match outcome with low precision (39.78%). Measures of playing style variability, namely flexibility, stability, and adaptability, were derived using the coefficient of unalikeability, and relationship with performance outcomes were computed using Pearson`s correlation. Playing style flexibility was positively correlated with both offensive and defensive performance outcomes and win frequency (p < .01). Conversely, teams displaying higher playing style stability conceded less shots in the penalty box but demonstrated poorer offensive performance outcomes (p < .01). Playing style adaptability, indicative of high flexibility and stability, was positively associated with defensive performance outcomes and win frequency (p < .01). Our results indicate that playing style variability may be a significant indicator of team performance.
© Copyright 2022 Proceedings of the 9th International Performance Analysis Workshop and Conference & 5th IACSS Conference. PACSS 2021. Advances in Intelligent Systems and Computing, vol 1426. Veröffentlicht von Springer. Alle Rechte vorbehalten.