Introduction: To date, much of the research on coordination dynamics in team sport has benefitted from concepts and tools from Dynamical Systems Theory (DST). However, while DST provides suitable techniques for modeling living systems, it makes no direct claims about their status nor provides a theoretical basis for understanding the principles of their behavior. The aim of this study was to develop a method of analysis to measure the self-organization of teams based on a recently developed ecological ontology of perception and action in soccer [1]. This framework applies DST to the functional semantics of the game, where, for example, self-organizing behaviors in the defensive team are defined by the relations between the position of the possessor of the ball relative to the goal as a function of time. Method: Twenty-two elite soccer players from La Liga played a practice game during a training session. Player position data were collected via GPS (10 Hz) for an entire match (two halves of thirty minutes). We analyzed the degree of synchronization in the defensive team in terms of the Kuramoto order parameter (cluster amplitude) r, (high synchronization = 1) by measuring location data (x,y coordinates on the pitch) over time [2]. Results: Team synchronization was measured in 4 blocks of 5 minute time-epochs. For block 1 (r = 0.978 ± 0.014); block 2 (0.962 ± 0.033); block 3 (0.981 ± 0.026); and block 4 (0.968 ± 0.046). However, while mean synchronization across blocks was fairly consistent, F(1, 8) = 0.22, p = 0.65, the overall degree of synchronization ranged between 0.84 and 0.99. To account for this drastic changes we compared our synchronization time series of all blocks to corresponding events in the match. Visual inspection of these time series revealed five instances when r decreased by more than 0.20. These dramatic decreases in the defensive team`s synchrony corresponded with scoring opportunities of their opponent. Discussion: Our analysis by means of the Kuramoto order parameter revealed that the synchronization measure was able to distinguish between periods of low and high group synchronization. Using this approach as a foundation, we are building a preliminary model that captures systematic changes in coordination between defensive players at the local (one-on-one) and global (team-wide) scales—quantifying breakdowns in the stability of the defensive team`s coordination and showing how they anticipate scoring opportunities for an opponent. This approach promises to address gaps in this literature by taking DST measures related to group coordination and synchrony and integrating them with measures that capture the functional, qualitative aspects of the game.
© Copyright 2016 21st Annual Congress of the European College of Sport Science (ECSS), Vienna, 6. -9. July 2016. Published by University of Vienna. All rights reserved.