Identifying playing talent in professional football using artificial neural networks

Barron, D., Ball, G., Robins, M. T. and Sunderland, C. (2020) Identifying playing talent in professional football using artificial neural networks. Journal of Sports Sciences, 38 (11-12). pp. 1211-1220. ISSN 0264-0414

[thumbnail of This is an Accepted Manuscript of an article published by Taylor & Francis in Journal of Sports Sciences on 15/01/2020, available online: http://www.tandfonline.com/10.1080/02640414.2019.1708036]
Preview
Text (This is an Accepted Manuscript of an article published by Taylor & Francis in Journal of Sports Sciences on 15/01/2020, available online: http://www.tandfonline.com/10.1080/02640414.2019.1708036)
Identifying playing talent in professional football using artificial neural networks.pdf - Accepted Version
Available under License Creative Commons Attribution.

Download (281kB) | Preview
[thumbnail of Identifying playing talent in professional football using artificial neural networks.pdf] Text
Identifying playing talent in professional football using artificial neural networks.pdf - Published Version
Available under License Creative Commons Attribution.

Download (1MB)

Abstract

The aim of the current study was to identify key performance indicators in professional
football that predict out-field players league status. The sample consisted of 966 players who
completed the full 90 minutes during the 2008/09 or 2009/10 season in the Football League
Championship. Players were assigned to one of three categories based on where they
completed most of their match time in the following season, then split based on five positions
including full backs (n = 205), centre backs (n = 193), centre midfielders (n = 205), wide
midfielders (n = 168) and attackers (n = 195). 340 performance, biographical and esteem
variables were analysed using a Stepwise Artificial Neural Network approach. The models
correctly predicted between 72.7% and 100% of test cases (Mean prediction of models =
85.9%), the test error ranged from 1.0% to 9.8% (Mean test error of models = 6.3%).
Variables related to passing, shooting, regains and international appearances were key factors
in the predictive models. This is highly significant as objective position-specific predictors of
league status have not previously been published. The method could be used to aid the
identification and comparison of players as part of the due diligence process in professional
football.

Item Type: Articles
Additional Information: Issue 11-12: Talent Identification and Development in Soccer
Uncontrolled Keywords: Soccer, talent identification, Premier League, championship, artificial intelligence, physical therapy, sports therapy and rehabilitation, Orthopedics and Sports Medicine
Subjects: G Geography. Anthropology. Recreation > GV Recreation Leisure > GV557 Sports
Q Science > Q Science (General)
Q Science > QA Mathematics > QA76 Computer software
Divisions: Academic Areas > Institute of Sport > Area > Exercise Physiology
Depositing User: Matthew Robins
Date Deposited: 12 Sep 2019 08:56
Last Modified: 03 Mar 2021 15:58
URI: https://eprints.chi.ac.uk/id/eprint/4814

Actions (login required)

View Item
View Item

Our address

I’m looking for