Identifying professional career path patterns is an important topic in sports analytics. It helps teams and coaches make the best transfers and team compositions. It also helps players find out what skills and how they need to improve to achieve their career goals. In this paper, we seek the player characteristics that mostly affect a player`s evaluation. To this end, we first created three-year-long career path segments from the time series data of 4204 players, then we created clusters from each segment based on the market value change over the examined period. After the clustering we searched for professional career path patterns where the market value growth was outstanding. Then we identified the 5 most important features with dynamic time warping and calculated how these should change over the years to achieve this career path. Finally we validated our findings with binary classification. We found that it is possible to explain real life professional career path patterns based on outstanding market value growth with the information collected from the FIFA video game series data collection. We managed to evaluate the extent of how these characteristics should change over the years to achieve the desired career.
© Copyright 2022 Machine Learning and Data Mining for Sports Analytics. MLSA 2021. Communications in Computer and Information Science. Veröffentlicht von Springer. Alle Rechte vorbehalten.