Abstract
As the sport of football has progressed, the NFL has become more aware of the safety issues the players face. They have implemented various rule changes and technological advancements in an attempt to help protect players. Concussion risk and subsequent risk of long-term brain damage are some of the main issues facing players today.
This paper examines concussions from two perspectives: common characteristics of players and weeks missed. I utilized an exploratory data analysis of player characteristics to identify relationships with concussion occurrence. I then built ordinal regression models, decision trees, and random forests to predict weeks missed due to concussions, examining if any variables are particularly predictive. Finally, I conducted a Kolmogorov-Smirnov Test to see if the concussion distribution is similar to a random distribution.
My research was generally inconclusive, except with the Kolmogorov-Smirnov Test, which revealed that concussions are not randomly distributed. This leads me to believe that there must be some variable out there strongly linked to concussions, but it's not found within our dataset of player characteristics.
Advisor
Long, Colby
Department
Mathematics
Recommended Citation
Pearson, Joshua, "An Analysis of NFL Concussions: What Causes them and What Influences a Player's Time Off" (2024). Senior Independent Study Theses. Paper 11216.
https://openworks.wooster.edu/independentstudy/11216
Keywords
NFL, Football, Concussions, Predictive Analysis, Ordinal Logistic Regression, Decision Tree, Random Forest Model, Kolmogorov-Smirnov Test
Publication Date
2024
Degree Granted
Bachelor of Arts
Document Type
Senior Independent Study Thesis
© Copyright 2024 Joshua Pearson
