NBA Win Probability model

This repo contains code for modeling NBA win probability and various notebooks for exploration and analysis of the modeling approach.

• dataloader.py: functions for loading play-by-play data into pandas
• plotting.py: functions for displaying specific plots inside of notebooks
• transformations.py: dataframe transformations required for building model
• win_probability: code for win_probability model

• Part 1 - ValidateNormality: notebook for confirming model assumptions of normality in the data
• Part 2 - ModelFit: notebook showing basic model usage and results
• Part 3 - EvaluateModel: model evaluation code comparing training and testing code

• https://eightthirtyfour.com/data

• A Brownian Motion Model for the Progress of Sports Scores