This project explores the use of machine learning algorithms to predict the outcome of football matches in the English Premier League (EPL). It intends to create a working model, using common match statistics such as xG (expected goals), shots on target (sot), and more in order to make these predictions. It is successfully predicted historical match results to high accuracy and precision, and also was able to predict the results of unplayed matches also with good accuracy and precision. This model provides a sound base that can be expanded to improve predictions with the addition of more complex features (e.g. player stats).