Fantasy Football Prediction

The ubiquity of professional sports and specifically the NFL have lead to an increase in popularity for Fantasy Football. Every week, millions of sports fans participate in their Fantasy Leagues. The main tasks for users are the draft, the round-based selection of players before each season, and setting the weekly line-up for their team. For the latter, users have many tools at their disposal: statistics, predictions, rankings of experts and even recommendations of peers. There are issues with all of these, though. Most users do not want to spend time reading statistics. The prediction of Fantasy Football has barely been studied and are fairly inaccurate. The experts judge mainly based on personal preferences instead of unbiased measurables. Finally, there are only few peers voting on line-up decisions such that the results are not representative of the general opinion. Especially since many people pay money to play, the prediction tools should be enhanced as they provide unbiased and easy-to-use assistance for users. This paper provides and discusses approaches to predict Fantasy Football scores of Quarterbacks with relatively limited data. In addition to that, it includes several suggestions on how the data could be enhanced to achieve better results. The dataset consists only of game data from the last six NFL seasons. I used two different methods to predict the Fantasy Football scores of NFL players: Support Vector Regression (SVR) and Neural Networks. The results of both are promising given the limited data that was used. After an overview of related work in Section 2, I present my solution. Afterwards, I describe the data set in greater detail before Section 5 explains the experiments and show the results. Finally, Section 6 discusses the findings and possible future work.