Department of Computer Science and Engineering
The Fantasy Football Analytics project provides users with statistics for football players, and fantasy football point projections using machine learning. Each football player and their respective team’s statistics are scraped from publicly available football statistic websites using Python language. After gathering relevant statistical information on players, a prediction algorithm utilizing machine learning projects expected fantasy football points for each player.
The python scripts responsible for web scraping and gathering available statistics on players gather data from http://www.sports-reference.com. A public API that provides data from sports-reference is used to gather most of the statistics, while the rest are gathered by web scraping the site. The prediction algorithm uses linear regression and gradient descent models of machine learning to analyze the data and make fantasy football point projections for each of the four position categories.
The statistics and projections for each player are displayed on a publicly accessible website and divided into four main positions that consist of quarterbacks, running backs, wide receivers, and tight ends. Users can access the website and find relevant statistics on each of the positions for any football player that is playing in the current season. The table is sortable by columns and also contains a search function that allows users to quickly find a specific player that allows for easy access to player statistics and fantasy football point projections.