Created a project that applies linear regression to NBA player data to identify the most overvalued and undervalued players based on their contracts. Using the HoopR package, I retrieved ESPN player stats, including salary information, and filtered out players with insufficient games or missing salary data.

I then built a linear model, using key performance metrics—such as points, assists, rebounds, shooting percentages, and PER—to predict player salaries. By comparing the predicted salary to the actual contract amount, I identified potential bargains and overpays. Finally, I created a ranked table to highlight the top 10 most overvalued and undervalued players based on the model’s residuals.

This project showcases my data manipulation, modeling, and visualization skills in a real-world basketball analytics context, making it a strong addition to my resume or online portfolio.