Betting on Machine Learning: Extracting Patterns from Football’s Anarchic Odds

Abstract

The application of machine learning in sports analytics has gained significant attention in recent years, particularly in the domain of football betting. Predictive models, driven by historical data, have the potential to offer deeper insights into match outcomes, providing both bookmakers and bettors with a competitive advantage. This study investigates the use of machine learning algorithms, including Deep Neural Networks (DNN), Decision Trees, and Random Forests, to predict football match results based on historical performance metrics. Emphasis is placed on the role of feature selection and dimensionality reduction techniques, such as Mutual Information (MI) and Principal Component Analysis (PCA), in enhancing model accuracy. The findings contribute to the growing field of sports analytics, offering a framework for improved prediction strategies in football.