Loading...

Prediction of Penalty Kick Direction with the Help of Data Mining Algorithms

Velayati Asgari, Mohammad | 2025

26 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 58700 (01)
  4. University: Sharif University of Technology
  5. Department: Industrial Engineering
  6. Advisor(s): Rafiee, Mjid
  7. Abstract:
  8. Football, as the world’s most popular sport, has always been a stage for critical decisions and decisive moments. One such moment is the penalty kick, where the penalty taker and goalkeeper face each other directly, and the outcome of this duel can determine the result of the match or even a championship. Predicting the direction of a penalty kick is not only fascinating and practical for players and coaches but also for sports analysts and fans. In this study, data mining and machine learning algorithms were employed to identify the key factors influencing the direction choice of penalty kicks. A dataset containing real-world information on player and goalkeeper characteristics, match conditions, and behavioral patterns was collected and processed. After data preprocessing and feature selection, various models—including Decision Tree, Random Forest, Support Vector Machine, Gradient Boosting, K-Nearest Neighbors (KNN), Naive Bayes, XGBoost, CatBoost, and the Stacking Ensemble model—were implemented and evaluated. The results show that advanced models such as XGBoost and CatBoost have delivered more accurate and stable performance compared to classical models. These models have proven particularly effective in identifying different behavioral patterns of players under high psychological pressure and across various match scenarios (e.g., tied score, trailing, or leading situations). The outcomes of this research can be applied to design targeted training sessions for players, enhance tactical decision-making for coaches, and develop specific strategies to confront certain penalty takers. Moreover, the findings indicate that combining data-driven analysis with expert coaching knowledge can significantly increase the success rate in penalty kick situations
  9. Keywords:
  10. Soccer Math ; Machine Learning ; Data Mining ; Penalty Kick ; Prediction

 Digital Object List

 Bookmark

No TOC