Applications of dynamical systems theory and 'complex' analyses to cricket fast bowling

The aims of this thesis were to: (i) increase understanding of the biomechanical and motor control processes that underpin proficient fast bowling performance using dynamical systems theory and 'complex' analyses; and (ii) demonstrate the application of dynamical systems theory and the utility of 'complex' analyses to performance-oriented sports biomechanics research using cricket fast bowling as a representative task vehicle. Prior to analysing within- and between-bowler differences in coordination patterns at different levels of analysis and their relationship to ball release speed, the suitability of manual coordinate digitising for analysing intra- and inter-individual variability was examined. Both the reliability of time-discrete and time-continuous kinematic variables was considered. Of the 33 time-discrete kinematic variables examined, 31 exhibited between-participant variances and re-digitisation variances that accounted for the largest and smallest portions of total variance, respectively. Furthermore, re-digitisation variance accounted for less than 5% of total variance in 29 of these variables with 15 of these exhibiting less than 1%. For the 45 time-continuous kinematic variables, measurement error accounted for 17.2% of movement variability (range 4.3-41.0%). When considered together, these results indicated that manual coordinate digitising was sufficiently sensitive to reliably measure differences in technique within and between bowlers. Kohonen Self-Organising Maps (SOMs) were used to analyse coordination patterns in cricket fast bowling at a global whole-body level of analysis. Qualitative differences in SOM trajectories between bowlers signified participant-specific coordination patterns, which were attributed to differences in organismic constraints and intrinsic dynamics. A theoretical argument against the common optimal movement pattern concept was constructed and the utility of SOMs was evaluated. Several issues currently limiting their practical application, including the difficulty in linking the SOM trajectory to aspects of technique and the inability of biomechanists to identify optimal sports techniques, were highlighted. A combination of 'complex' analytical techniques was then applied to quantify intersegmental coordination among key limb and torso segments. Cross-correlation functions showed that moderate (0.5+) to very strong (0.9+) coupling relationships existed for the four segment couplings (NBA vs. FL, BA vs. NBA, BA vs. FL, UT vs.P) with the majority of these moving in synchrony. Statistically significant mean differences in both cross-correlation coefficients and average coupling angle for the four segment couplings throughout (0-100%), and during different phases (0-24%, 25-49%, 50-74%, 75-99%) of, the delivery stride provided further evidence of participant-specific coordination patterns. However, no associations between coupling relationships and ball release speed could be identified either within or between bowlers. This study further highlighted the difficulties in making associations between technique and outcomes.It was concluded that, based on the reported research findings, dynamical systems theory and its associated 'complex' analyses could make a substantive contribution to the enhancement of knowledge of cricket fast bowling techniques and also advance applied sports biomechanics research more generally. Further investigations into cricket fast bowling performance, focusing on the link between technique and outcomes using a combination of kinetic, energetic and coordination analyses, were identified as a research priority.