Journal of Sports Science and Medicine
Journal of Sports Science and Medicine
ISSN: 1303 - 2968   
Ios-APP Journal of Sports Science and Medicine
Androit-APP Journal of Sports Science and Medicine
from September 2014
©Journal of Sports Science and Medicine (2013) 12, 116 - 121

Research article
Evaluating Australian Football League Player Contributions Using Interactive Network Simulation
Jonathan Sargent , Anthony Bedford
Author Information
School of Mathematical & Geospatial Sciences, RMIT University, Australia.

Jonathan Sargent
✉ RMIT University, Plenty Road, Bundoora East, VIC, Australia, 3083
Publish Date
Received: 19-11-2012
Accepted: 10-01-2013
Published (online): 01-03-2013
Share this article

This paper focuses on the contribution of Australian Football League (AFL) players to their team’s on-field network by simulating player interactions within a chosen team list and estimating the net effect on final score margin. A Visual Basic computer program was written, firstly, to isolate the effective interactions between players from a particular team in all 2011 season matches and, secondly, to generate a symmetric interaction matrix for each match. Negative binomial distributions were fitted to each player pairing in the Geelong Football Club for the 2011 season, enabling an interactive match simulation model given the 22 chosen players. Dynamic player ratings were calculated from the simulated network using eigenvector centrality, a method that recognises and rewards interactions with more prominent players in the team network. The centrality ratings were recorded after every network simulation and then applied in final score margin predictions so that each player’s match contribution-and, hence, an optimal team-could be estimated. The paper ultimately demonstrates that the presence of highly rated players, such as Geelong’s Jimmy Bartel, provides the most utility within a simulated team network. It is anticipated that these findings will facilitate optimal AFL team selection and player substitutions, which are key areas of interest to coaches. Network simulations are also attractive for use within betting markets, specifically to provide information on the likelihood of a chosen AFL team list “covering the line ”.

Key words: Interaction matrix, negative binomial distribution, eigenvector centrality, player ratings

           Key Points
  • A simulated interaction matrix for Australian Rules football players is proposed
  • The simulations were carried out by fitting unique negative binomial distributions to each player pairing in a side
  • Eigenvector centrality was calculated for each player in a simulated matrix, then for the team
  • The team centrality measure adequately predicted the team’s winning margin
  • A player’s net effect on margin could hence be estimated by replacing him in the simulated side with another player
Home Issues About Authors
Contact Current Editorial board Authors instructions
Email alerts In Press Mission For Reviewers
Archive Scope
Supplements Statistics
Most Read Articles
  Most Cited Articles
JSSM | Copyright 2001-2020 | All rights reserved. | LEGAL NOTICES | Publisher

It is forbidden the total or partial reproduction of this web site and the published materials, the treatment of its database, any kind of transition and for any means, either electronic, mechanic or other methods, without the previous written permission of the JSSM.

This work is licensed under a Creative Commons License Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.