This study developed a method to determine whether the distribution of individual player performances can be modelled to explain match outcome in team sports, using Australian Rules football as an example. Player-recorded values (converted to a percentage of team total) in 11 commonly reported performance indicators were obtained for all regular season matches played during the 2014 Australian Football League season, with team totals also recorded. Multiple features relating to heuristically determined percentiles for each performance indicator were then extracted for each team and match, along with the outcome (win/loss). A generalised estimating equation model comprising eight key features was developed, explaining match outcome at a median accuracy of 63.9% under 10-fold cross-validation. Lower 75th, 90th and 95th percentile values for team goals and higher 25th and 50th percentile values for disposals were linked with winning. Lower 95th and higher 25th percentile values for Inside 50s and Marks, respectively, were also important contributors. These results provide evidence supporting team strategies which aim to obtain an even spread of goal scorers in Australian Rules football. The method developed in this investigation could be used to quantify the importance of individual contributions to overall team performance in team sports.

This study investigated the validity of the official Australian Football League Player Ratings system. It also aimed to determine the extent to which the distribution of points across the 13 rating subcategories could explain Australian Football League match outcome. Ratings were obtained for each player from Australian Football League matches played during the 2013–2016 seasons, along with the corresponding match outcome (Win/Loss and score margin). The values for each of the 13 subcategories that comprise the ratings were also obtained for the 2016 season. Total team rating scores were derived as an objective team outcome for each match. Percentage agreement and Pearson correlational analyses revealed that winning teams displayed a higher total team rating in 94.2% of matches and an association of r = 0.96 (95% confidence interval = 0.95–0.96) between match score margin and total team rating differential, respectively. A Partial Decision Tree (PART) analysis resulted in seven rules capable of determining the extent to which relative contributions of rating subcategories explain Win/Loss at an accuracy of 79.3%. These models support the validity of the Australian Football League Player Ratings system and its use as a pertinent system for objective player analyses in the Australian Football League.

No abstract

Player evaluation plays a fundamental role in the decision-making processes of professional sporting organisations. In the Australian Football League, both subjective and objective evaluations of player match performance are commonplace. This study aimed to identify the extent to which performance indicators can explain subjective ratings of player performance. A secondary aim was to compare subjective and objective ratings of player performance. Inside Football Player Ratings (IFPR) and Australian Football League Player Ratings were collected as subjective and objective evaluations of player performance, respectively, for each player during all 1026 matches throughout the 2013–2017 Australian Football League seasons. Nine common player performance indicators, player role classification, player age and match outcomes were also collected. Standardised linear mixed model and recursive partitioning and regression tree models were undertaken across the whole dataset, as well as separately for each of the seven player roles. The mixed model analysis produced a model associating the performance indicators with IFPR at a root mean square error of 0.98. Random effects accounting for differences between seasons and players ranged by 0.09 and 1.73 IFPR each across the five seasons and 1052 players, respectively. The recursive partitioning and regression tree model explained IFPR exactly in 35.8% of instances, and to within 1.0 IFPR point in 81.0% of instances. When analysed separately by player role, exact explanation varied from 25.2% to 41.7%, and within 1.0 IFPR point from 70.3% to 88.6%. Overall, kicks and handballs were most associated with the IFPR. This study highlights that a select few features account for a majority of the variance when explaining subjective ratings of player performance, and that these vary by player role. Australian Football League organisations should utilise both subjective and objective assessments of performance to gain a better understanding of the differences associated with subjective performance assessment.