Factors associated with match outcomes in elite European football – insights from machine learning models
Settembre M, Buchheit M, Hader K, Hamill R, Tarascon A, Verheijen R, McHugh D. Factors associated with match outcomes in elite European football – insights from machine learning models. Journal of Sports Analytics. In press
Full text here
🚀 Unveiling the Dynamics of Football Match Outcomes: Insights from Machine Learning
Our latest research paper takes a groundbreaking look at the top factors influencing match results in European football. By analyzing over 61,000 matches with the power of XGBoost, we’ve uncovered compelling insights:
- 🚌 Travel Impact: The journey matters just as much as the destination. Travel distance has a more significant effect on match outcomes than the advantage of playing at home, especially when teams travel over 100 km.
- 🏆 Elo Rankings: Beyond general team rankings, the relative differences in Elo between teams play a crucial role, revealing the importance of team quality and the potential for “easier” matches against lower-ranked opponents.
- 🔥 Recent Form: A team’s performance in their last five matches and overall season performance contributes significantly to match outcomes, underlining the momentum a team carries into each game.
- 🧠 Managerial Influence: The duration a manager has been with a team can affect performance, with a sweet spot existing before the onset of diminishing returns—highlighting the strategic impact of leadership duration.
- ⏳ Rest and Recovery: Rest days are more than a pause; they are a strategic component of performance. Our findings show that less than three days of rest can adversely affect top teams’ competitive edge.
- 🔄 Squad Rotation: Stability wins the game. Frequent player rotations may hinder match outcomes due to reduced player communication and team cohesion, pointing to the intricate balance coaches must strike between rotation and consistency.