Machine Learning · Sports Analytics
NBA Player Points Analytics
Collaborator: Gabriel Castillo
Built a machine learning project to predict how many points an NBA player will score in a given game by combining player offensive statistics with opponent defensive metrics. The model moves beyond season averages by accounting for matchup-specific context.
- Developed a regression-based prediction pipeline using publicly available NBA data
- Compared standard linear regression and ridge regression
- Used standardized numerical features across thousands of player-game instances
- Benchmarked results against season averages and FanDuel player prop lines