Using NFL Regular Season Data to Predict Super Bowl Winners with Logistic Regression

For our approach we used ROSE to balance our data before training our GLM models. Our models were trained with data from the 2009 to 2014 NFL seasons. Our models were tested with data from the 2015 to 2020 NFL seasons. Our models were validated with data from the 2021 NFL season. For more information about our approach, refer to Our Paper,

Data Pre-Processing

Run Code.R for data preprocessing.

Implementation:

Run modified_implemtation.RMD