The use of machine learning in sport outcome prediction: A review

Abstract

The increase in the volume of structured and unstructured data related to more than just sport events leads to the development and increased use of techniques that extract information and employ machine-learning algorithms in predicting process outcomes based on input but not necessarily output data. Taking sports into consideration, predicting outcomes, and extracting valuable information has become appealing not only to sports workers but also to the wider audience, particularly in the areas of team management and sports betting. The aim of this article is to review the existing machine learning (ML) algorithms in predicting sport outcomes. Over 100 papers were analyzed and only some of these papers were taken into consideration. Almost all of the analyzed papers use some sort of feature selection and feature extraction, most often prior to using the machine-learning algorithm. As an evaluation method of ML algorithms, researchers, in most cases, use data segmentation with data being chronologically distributed. In addition to data segmentation, researchers also use the k-cross-evaluation method. Sport predictions are usually treated as a classification problem with one class being predicted and rare cases being predicted as numerical values. Mostly used ML models are neural networks using data segmentation.

Who Will Score? A Machine Learning Approach to Supporting Football Team Building and Transfers

“…Nevertheless, as mentioned in the Section 5 , there are some threats to validity that can affect the obtained results and the whole approach. Additionally, as mentioned in [ 27 ], sport is unpredictable. Hence, the indications coming from ML should be treated solely as the pieces of advice.…”

Understanding gender differences in professional European football through machine learning interpretability and match actions data

“…Professional players are now evaluated and compared using a comprehensive list of performance indicators to determine their value and contribution to the team 1 – 3 . These indicators have also shown the ability to determine the style of play of specific teams 4 and numerous applications for sports forecasting 5 . Despite the undoubtedly increasing data-based culture in football, there are still certain factors in this sport that are based on subjective and biased criteria.…”

Use of Machine Learning and Deep Learning to Predict the Outcomes of Major League Baseball Matches

“…Horvat and Job [17] reported on the accuracies of sports outcome prediction. The highest accuracy was noted for the outcome prediction of American football matches (>90%), followed by cricket matches (highest accuracy rate >85%).…”