Affiliation:
1. Reviewers: László Csató (Corvinus University of Budapest, Hungary) Filipe Clemente (Polytechnic University of Viana do Castelo, Portugal)
Seife Dendir (Radford University, USA)
2. MTA-BME Information Systems Research Group, Department of Telecommunications and Media Informatics, Faculty of Electrical Engineering and Informatics, Budapest University of Technology and Economics, Budapest, Hungary
Abstract
The success of a soccer player is not entirely pre-destined by their physical ability, talent, and motivation. There are certain decisions along the way that greatly affect the arc of their career: which skills to develop, and which club to sign a contract with. In this paper, we identify the optimal strategic choices toward multiple potential aims a soccer player can have and we seek the knowledge of what made the greatest soccer players in terms of those decisions. Our two main data sources are Transfermarkt and Sofifa from which we collect data for the period between 2007 and 2021 with 29,231 players. We perform time series analysis on skill features of soccer players, and network analysis of the players’ acquaintance graph, i.e., a graph that indicates whether two given players have ever been teammates before. Finally, we create key performance indicators to check the differences in certain features, i.e., individual player skills and connectivity attributes, between top-tier and the rest of the players, and use dynamic time warping for validation. The outcome of this work is a recommendation tool that helps players to find what needs to be improved in order to achieve their desired goals. The source code and the career advisor tool for soccer players that we have implemented are available online.
Funder
Nemzeti Kutatási Fejlesztési és Innovációs Hivatal
Subject
Social Sciences (miscellaneous)