During my time at General Assembly, I decided to try a kaggle competition for some extra practice. This repository is a collection of the work that was completed for that project.
The goal of this project was to see how well you could predict whether or not Kobe Bryant made or missed a shot given certain information about the shot attempt. This was a nice project to sharpen my skills on. The data was relatively clean, but it still required some feature engineering. My best model ended up being a simple logistic regression. I was surprised, but it performed much better than any of the other models I tested.
One neat visualization that came out of this project was a chart showing all the makes and misses for 25,000 of Kobe Bryant's career shots. It's printed below.
