In this chapter, we have presented consideration of sports games as selforganizing systems on the basis of coupled oscillator dynamics. Relative phase was introduced as a metric for investigating the behavioural dynamics of both net games (e.g. tennis and squash) and invasive games different (e.g. basketball and football), with in-phase and anti-phase in some instances representing attractor states for game behaviours at the expense of other possible phase relations. The notion of behavioural perturbations serving to destabilize the system from these phase attractions was introduced and considered important far game description. Moreover, the presence ofbehavioural perturbations in game sports has been validated from visual inspection although their predicted corresponding associations with relative phase variability await future demonstration for the most part. Beyond considerations of perturbations, relative phase and transitions between phase attractors, artificial neural networks were introduced as a means of identifying game structure from large datasets using football data as example. As with earlier demonstrations using relative phase, various unique phase structures (playing configurations) were identified by the neural networks, together with time-dependent phase changes indicating the underlying game dynamies. In closing, the approaches outlined in this article go some way to describing and explaining the behavioural structure of sports games and their underpinning dynamies.
© Copyright 2014 Complex systems in Sport. Published by Routledge. All rights reserved.