Nowadays, data science is applied in several areas of daily life. There have been many applications to sports. In this context, the attention will be focused on football (i.e. 'soccer' for Americans): the making of strategic choices, whether by the scouting department of the football club, or the technical staff, up to the management, is crucial. It has been measured and monitored football players' performance in the season 2018/2019, for the top five European Leagues, using data provided by Electronic Arts (EA) experts and available on the Kaggle data science platform. For this purpose, with the help of football experts, a third-order partial least-squares path model (PLS-PM) approach was adopted to the sofifa key performance indices in order to compute a composite indicator differentiated by role and compare it with the well-known overall indicator from EA Sports. It has been taken into account players' observed heterogeneity (i.e. roles and leagues), since often experts refer to differences in these features, and so the objective is to verify their importance scientifically. The results are very consistent with this because they underline how some sub-areas of performance have different significance weights depending on the role.
Observed heterogeneity in players' football performance analysis using PLS-PM
Mattia Cefis
2022-01-01
Abstract
Nowadays, data science is applied in several areas of daily life. There have been many applications to sports. In this context, the attention will be focused on football (i.e. 'soccer' for Americans): the making of strategic choices, whether by the scouting department of the football club, or the technical staff, up to the management, is crucial. It has been measured and monitored football players' performance in the season 2018/2019, for the top five European Leagues, using data provided by Electronic Arts (EA) experts and available on the Kaggle data science platform. For this purpose, with the help of football experts, a third-order partial least-squares path model (PLS-PM) approach was adopted to the sofifa key performance indices in order to compute a composite indicator differentiated by role and compare it with the well-known overall indicator from EA Sports. It has been taken into account players' observed heterogeneity (i.e. roles and leagues), since often experts refer to differences in these features, and so the objective is to verify their importance scientifically. The results are very consistent with this because they underline how some sub-areas of performance have different significance weights depending on the role.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
