Running Logistic Regression on NBA play-by-play data. The objective was to show that we can mathematically predict the probability of a basketball shot going in based on certain shot characteristics. Consult 'FINALMODEL' for final and clean logistic model.
The files also include a try to predict basketball scoring with different distributions: mainly exponential and Poisson. The results showed interesting insights and could be material for future studies.
Lastly, the file "court" helps drawing a basketball court in R and eventually plot real shots having x-y coordinates.