In this project, I focus on using Logistic Regression to understand which factors are the most important to win a game of League of Legends. After fitting a Logistic Regression model to my data, I found the regression coefficients to understand which conditions were most crucial to a win. I also compared the regression coefficients between the five most popular regions to see if there is a difference in win conditions. In order, getting the First Inhibitor, First Tower, and First Blood are the most impactful predictors of whether or not a team will win or lose. The number of tower kills is the most consistently impactful predictors of a win across the regions accounted for in my data.
Click here to view the code for the project! I've also linked the dataset I generated here.
Medium article to come soon!