How well can NFL Home Game Wins be predicted using gameday statistics, weather, and positional salary data from the last 10 seasons?
This project will focus on predicting the winner of NFL home games using machine learning methods. My motivation for doing this project is my enjoyment of watching football and my interest in connecting sports statistics to outcomes. The data to be used will be the following:
• NFL stats from a JSON feed on NFL.com’s live GameCenter page from the 2009 -2018 seasons
• NFL positional spending by team from Spotrac.com
• Gameday weather and conditions from NFLWeather.com
My motivation for doing this project is my enjoyment of watching football and my interest in connecting sports statistics to outcomes. I am a regular visitor to Nate Silver’s FiveThirtyEight, and I am obsessed with the numerous sports predictors on the site. I want to create similar models that I can build upon with other features, such as player-specific data.
• Pablo Bosch and Sandjai Bhulai. 2018. “Predicting the winner of NFL-games using Machine and Deep Learning”
• Jim Warner. 2010. "Predicting Margin of Victory in NFL Games: Machine Learning vs Deep Learning"