Comparing Predictive Models For English Premier League Games

Data science has become an important aspect of modern day society. While the term was first coined in 1960 by Peter Naur, over the past decade, it has been applied to many different fields, one of which is sports. Over the past years, many ranking methods and rating systems have been developed for different sports; the Massey Ranking method, the Elo-rating system, and the Pomeroy ranking method are just a few examples of such models. However, there has been a lack of research in the area of accurate predictive modeling in soccer. The goal of this thesis is to compare and contrast a set of predictive models for determining the outcome of English Premier League (EPL) games.