Author Identifiers

Bradley Keller
ORCID: 0000-0003-2684-8926

Date of Award


Degree Type


Degree Name

Doctor of Philosophy


School of Medical and Health Sciences

First Advisor

Associate Professor Annette Raynor

Second Advisor

Doctor Lyndell Bruce

Third Advisor

Fiona Iredale


The attainment of expertise has been the focus of research in many domains including music, chess and sport. This research has progressed with many theories detailing the best way to develop expertise and nurture talent in sport. Soccer is a multifaceted sport which requires a number of physical, technical and tactical skills to be successful, making it difficult to achieve expertise. Although Australia’s performance on the international stage is improving, there is a lack of evidence to inform the most effective development pathways to support the next wave of talented youth soccer players. Therefore, the aim of the thesis was to understand what is required to be an expert in Australian youth soccer, and which environmental factors can influence the development of expertise in youth soccer players.

To enhance our understanding of the development of expertise in Australian soccer, the current thesis was guided by the Expert Performance Approach (Ericsson & Smith, 1991) and included three individual studies which captured expert performance, identified underlying mechanisms and examined how expertise was developed. Sixty-two male soccer players (17.0 ± 0.61 y) who represented three cohorts in Australian youth soccer; national elite (Australian Institute of Sport), state elite (state institute) and sub-elite (state league) participated in this study.

Study One captured expert performance through an in-depth analysis of the match characteristics of the three levels of expertise. A total of 24 matches across the three levels of expertise in Australian youth soccer were analysed, with each match videoed and manually coded using SportsCode according to frequently used match characteristics from the literature. A hierarchical cluster analysis was used to see if teams with similar technical characteristics could be grouped together in order to make inferences about distinctive tactics and game styles. There were three game styles identified across the cohorts, with the state and national elite cohorts forming two distinct clusters, whilst the sub-elite teams clustered together based on technical output. More specifically, the two elite cohorts executed two different possession styles of play, while the sub-elite cohort played a direct style of game. Although it was clear that technical output and game styles differed across cohorts, it was not clear which underlying mechanisms allowed teams to play this way.

The aim of Study Two was to identify which skills could distinguish the three levels of Australian youth soccer players and contribute to an explanation of the different game styles identified in Study One. This was done using a multifaceted testing battery including physical, technical and tactical tests. The physical tests included intermittent endurance, sprinting, change of direction and vertical jumps, the technical tests included short and long passing, dribbling and shooting, while the tactical test was a perceptual-cognitive decision-making task which required players to choose the correct option in a video-based task. There were a number of physical, technical and tactical outcome measures that could distinguish between cohorts based on the Receiver Operating Characteristic curves. The most prominent tests included the Yo-Yo Intermittent Recovery Test Level 1, 30m sprint and 20m flying start, height, Loughborough Soccer Passing Test, long passing test, ball control, shooting test and perceptual-cognitive decision-making task. Furthermore, the multidimensional analysis could clearly differentiate players from each cohort based on a Cumulative Total Score for each player. It is evident that the underlying mechanisms for expert performance in Australian youth soccer included elements of physical, technical and tactical prowess which may contribute to the differences in game styles observed in Study One.

Study Three examined how expertise is developed in Australian youth soccer players. The participants completed the Development History of Athletes Questionnaire (DHAQ) (Hopwood, Baker, MacMahon, & Farrow, 2010). A decision tree induction analysis was used to determine which developmental factors contributed most to the predictor variable, the Cumulative Total Score. The amount of sport specific practice distinguished the two highest skilled groups from the lower skilled players. There were then two distinct pathways taken by the elite Australian youth soccer players. The first pathway included players who were later born in their family and had older siblings that participated in other sport, which contributed to their development in soccer. The second pathway included those players who were born early in their family (first or second), with this group specialising later in soccer (after the age of 13), compared to the second tier of athletes.

Overall it was clear that there were distinguishing game styles for various levels of Australian youth soccer players. The elite players had underlying physical, technical and tactical attributes that allowed them to execute a possession-based game style. This thesis has provided evidence that the national elite players had followed a different pathway and been exposed to different environmental influences compared to the sub-elite players, factors that had contributed to their current level of expertise and success. This work provides Football Federation Australia and associated personnel with a strong framework upon which to base their talent identification and development programs given this thesis was able to provide evidence of distinct game styles, physical, technical and tactical skills distinguishing playing levels and differing pathways exhibited by the athlete cohorts.


Paper Location