Skip to content

mattflaherty97/cfb_stats

Repository files navigation

Data

I am using data from the cfbscrapR package in this repository.

This folder contains code and a report that analyzes the two contenders for the 2021 College Football Playoff National Championship between Alabama and Ohio State. The report includes bootstrap sampling distributions to measure how both teams perform at the 95% confidence level.

This folder analyzes how field goals affect the win probability of a team on different parts of the field. For example, it analyzes how a field goal on 4th & 2 between the 11 yardline and 20 yardline affects the win probability of the kicking team as opposed to said team going for the first down.

I analyze the 2020 Memphis Tigers' team comparing this team to all other Memphis Tigers teams from the 2014 season to the most recent 2020 season in this folder.

I wanted to try dimensionality reduction that lead to a cluster analysis. I used PCA as my dimensionality reduction technique and was able to reduce my dataframe from 79 dimensions to 4 dimensions. I then clustered all 127 NCAA division-I college football teams using the 4 new dimensions. I was able to cluster the teams into 4 groups using k-means clustering.

About

Playing around with data from the cfbscrapR package.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages