Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
479207 | European Journal of Operational Research | 2016 | 11 Pages |
•Scoring efficiency of each team in a football match is computed.•Two parallel processes (home team and away team), with crossed inputs, are considered.•Average team efficiency for home and away matches and for the whole season is computed.•Target scores can also be used to estimate a virtual league table.
Data Envelopment Analysis (DEA) has been widely used for efficiency assessment in many sports and specifically in football. This paper presents a novel approach to measuring the scoring efficiency of football teams in a match. It uses a parallel process network DEA model that takes into account the defensive and attacking variables of the two teams, their respective percentage of possession of the ball and also the team's economic value. The proposed approach provides estimates of how many more goals each team should have scored in each of the matches played in the season. By averaging the scoring efficiency in the different matches played, team scoring efficiencies for the season can be computed. Moreover, the proposed DEA model can be run after each match so that the changes in efficiency can be monitored throughout the season. This can help coaches and managers assess their team's past performance and plan ahead for each match. The proposed approach has been applied to Spanish First Division teams for the season 2013/2014 and a virtual final league table has been computed.