A method to assess the influence of individual player performance distribution on match outcome in team sports

Abstract

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.

Multifactorial analysis of factors influencing elite australian football match outcomes: a machine learning approach

“…For the calculation of team characteristic variables, individual player game data was sourced from Champion Data (CD, Pty Ltd, Melbourne, Australia), the official statistics provider of the AFL, which is a valid and reliable source (Robertson, Gupta, & McIntosh, 2016). For every season (2013-2018) each team's official team list was collated, this included a player's height, mass and date of birth.…”

“…A general solution is to exclude opposition data in the analysis in a long format with each game (and variable) being represented from the perspective of both teams (Leicht, Gomez, & Woods, 2017;Robertson, Gupta, & McIntosh, 2016;Robertson & Joyce, 2015;Woods, Sinclair, & Robertson, 2017). Nevertheless, this approach still presents challenges, as for any given game two different predictions can be made from the analysis (i.e., one for each team).…”

Team performance indicators that predict match outcome and points difference in professional rugby league

“…Robertson, Back and Bartlett (2016) advocated this method for preparing for matches by including the opposition in the analysis; although it is more common for research papers to use absolute values (e.g. Casamichana, 2015;Castellano, Casamichana & Lago 2012;Graham & Mayberry, 2014;Higham, Hopkins, Pyne & Anson, 2014a;Higham, Hopkins, Anson & Pyne, 2014b;Lago-Penas, Lago-Ballesteros & Rey, 2011;Meletakos, Vagenas & Bavios, 2011;Najdan, Robins & Glazier, 2014;Vahed, Kraak & Venter, 2014) or correctly labelled (Robertson, Back & Bartlett, 2016;Robertson, Gupta & McIntosh, 2016). O'Donoghue (2008) suggested key performance indicators had higher correlations with principal components; Bremner, Robinson & Williams (2013) suggested they were more significantly related to success and Shafizadeh, Taylor & Penas (2013) provided no definition or evidence.…”

Is Football players’ performance influenced by the quality of opposition? Application of the Golden Index formula in Club Atlético de Madrid 2016/2017

“…Quantifying the individual performance of football players within teams is always a challenging task due to the variability and complexity of game dynamics. Nevertheless, understanding the impact of the individual contribution of players to the overall team performance 1 , is paramount for aiding performance analysts and coaches to better optimize teams' training and selection, or even to enhance scouting methods 2,3,4 .…”