Since 2015, the NBA has released a detailed report of foul calls and non-calls that occur in the final two minutes of close games. This talk is a case study in using open source Python packages to analyze these reports in order to understand the relationship between game dynamics, player abilities, and foul calls. Our main goal is to quantify the relationship between player ability and foul calls. Since intentional fouls are a ubiquitous part of the NBA endgame, this data set also contains rich information about the relationship between game dynamics and intentional fouls for us to model. This case study showcases the role of open source Python pacakges in a modern statistical modeling workflow:

• pandas is used to load, explore, and transform the foul call data into a format suitable for modeling,
• Matplotlib is used to visually explore trends in the foul call data,
• PyMC3 is used to build hierarchical Bayesian models of how game dynamics and player abilities affect foul calls. We highlight the power of easy access to state-of-the-art Hamiltonian Monte Carlo and the freedom this access gives modelers,
• seaborn is used to explore and validate the model's predictions. seaborn's FacetGrid is particularly well-suited to hierarchical models,
• Dash is used to interactively visualize the model's inferences.

This case study in open source data analysis with Python based on previously published modeling work.