Traditional approaches to the quantification of team sports have proved to be limited in their ability to identify complex structural regularities that, despite being unobservable, nonetheless underlie the development of the sporting contest between opposing teams. This paper describes a method for detecting the dynamics of play in professional soccer through the analysis of temporal patterns (T-patterns). The observation instrument used was SOF-5, which is especially designed for studying the dynamics of the game in soccer. The recording consisted of within-session monitoring using the MATCH VISION STUDIO 3.0 software, while the THEME software was used to detect and analyse T-patterns. These T-patterns revealed regularities in the playing style of the observed team, FC Barcelona. The structures detected included a ball possession pattern, whereby the ball was first kept in the central zone before being played forward, through several moves, into the zones closest to the opposing team's goal in order to disrupt the latter's equilibrium. The results obtained show that it is possible to identify stable temporal structures that provide information about concurrent interaction contexts with respect to lateral position and zone. As such, the proposed methodology appears to be useful in detecting complex structures within the game of soccer, structures which may help coaches to design attacking and defensive strategies.

Abstract:The following paper introduces a new approach to the analysis of sports and motor skill performance. The approach, known as T-pattern detection, is explained and preliminary data analyses from soccer, boxing, basketball, swimming and motor skill analysis are presented. The data show that specific temporal patterns can be identified within sports performances. The temporal patterns can relate to performance of specific actions (e.g. passes) or movement patterns. Further analysis showed a significant correlation between coaches' ratings' of performance with the number of temporal patterns exhibited by athletes.