Modeling team compatibility factors using a semi-Markov decision process: a data-driven approach to player selection in soccer

Abstract

Player selection is one of the great challenges of professional soccer clubs. Despite extensive use of performance data, a large number of player transfers at the highest level of club soccer have less than satisfactory outcome. This study uses player performance and decision making data to estimate team performance in terms of goal differential and model the effects of team compatibility on player and team performance. In this methodology, players' attributes are assessed with respect to the potential contribution to team performance, given the attributes of surrounding players. The study is using a semi-Markov decision process to model game flow. Performance data from the English Premier League between seasons 2008/2009 and 2011/2012 is used to predict the outcome of 69 transfers. The model provides an average error of 7.86 in predicting teams' goal differential in a season with current squad and 18.91 in estimating the effect of a future transfer on team performance.

Multiple Choice Method With Genetic Algorithm for the Formation of Soccer Teams

“…In many cases, decision-making about transferring athletes is questionable since the low performance of them. It is often associated the way in which players are evaluated, disregarding the contribution of their attributes to the collective performance of the contracting team (Al-Shboul et al, 2017;Jarvandi et al, 2013;Tavana et al, 2013).…”

Psychological assessment in sport psychology.

“…The rapidly evolving science of predictive modeling has many techniques, with artificial intelligence/ machine learning (i.e., procedures that find patterns in data that are not obvious) at its core. As an example in the sport context, simulations can model an entire season and can deduce optimal lineups, substitution patterns, and scoring potential of players (Aldrich, 2009;Jarvani, Sarkani, & Mazzuchi, 2014). When unforeseen events occur (e.g., long-term athletic injury), additional simulations can be performed to develop new predictions based on changes in available data.…”

A novel hybrid method for selecting soccer players during the transfer season

“…Ballı and Korukoğlu (2014) applied a fuzzy multiattribute decision-making framework based on the fuzzy analytic hierarchy process and the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) to the selection problem of basketball players in Turkey. Jarvandi, Sarkani, and Mazzuchi (2013) presented a model where the skills of each player were evaluated considering their potential role in the performance of the team. Vaeyens et al (2006) determined the relationship existing between physical factors, skill level, and the performance of young soccer players aged 12 to 16.…”