Citation:

Harish S, Abishek Kevin A, Harsha Vardhan U, Sharon Femi P, 2023. "Expected Goals Prediction in Football using XGBoost" ESP Journal of Engineering & Technology Advancements  3(1): 21-26.

Abstract:

Expected goals of a football match determine whether a team have won or lost. When considering the expected goal results, a team may appear to have lost the game but actually win it, and vice versa. The expected goal is the amount of goals a team should have scored based on the information available for that particular game. Numerous machine learning algorithms are employed to measure the effectiveness of shots in football. In this paper, we develop a gradient Boosting model to evaluate the scoring opportunities using event data collected from live football matches. This method can be used to show the players who are most likely to score at any given time throughout the game as well as where on the field they are most likely to do so. Experimental results demonstrate that they recognise and assess significant match opportunities and analyze the players in a football match depending upon their performance using Expected goals(xG) as an evaluation metric.

References:

[1] Spearman, W., 2018, February. Beyond expected goals. In Proceedings of the 12th MIT sloan sports analytics conference (pp. 1-17).
[2] Baboota, R. and Kaur, H., 2019. Predictive analysis and modelling football results using machine learning approach for English Premier League. International Journal of Forecasting, 35(2), pp.741-755.
[3] Partida, A., Martinez, A., Durrer, C., Gutierrez, O. and Posta, F., 2021. Modeling of Football Match Outcomes with Expected Goals Statistic. Journal of Student Research, 10(1).
[4] Rodrigues, F. and Pinto, Â., 2022. Prediction of football match results with Machine Learning. Procedia Computer Science, 204, pp.463- 470.
[5] Berrar, D., Lopes, P. and Dubitzky, W., 2019. Incorporating domain knowledge in machine learning for football outcome prediction. Machine learning, 108(1), pp.97-126.
[6] Bauer, P. and Anzer, G., 2021. Data-driven detection of counterpressing in professional football. Data Mining and Knowledge Discovery, 35(5), pp.2009-2049.
[7] Link, D., Lang, S. and Seidenschwarz, P., 2016. Real time quantification of dangerousity in football using spatiotemporal tracking data. PloS one, 11(12), p.e0168768.
[8] Staudemeyer, R.C. and Morris, E.R., 2019. Understanding LSTM--a tutorial into long short-term memory recurrent neural networks. arXiv preprint arXiv:1909.09586.
[9] Merhej, C., Beal, R.J., Matthews, T. and Ramchurn, S., 2021, August. What Happened Next? Using Deep Learning to Value Defensive Actions in Football Event-Data. In Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining (pp.3394-3403).
[10] Zambom-Ferraresi, F., Rios, V. and Lera-López, F., 2018. Determinants of sport performance in European football: What can we learn from the data?. Decision Support Systems, 114, pp.18-28.
[11] Joseph, A., Fenton, N.E. and Neil, M., 2006. Predicting football results using Bayesian nets and other machine learning techniques. Knowledge-Based Systems, 19(7), pp.544-553.
[12] Femi, P.S. and Vaidyanathan, S.G., 2022. An efficient ensemble framework for outlier detection using bio-inspired algorithm. International Journal of Bio-Inspired Computation, 19(2), pp.67-76.

Keywords:

Expected Goals, Extreme Gradient Boosting, Logistic Regression