Super Bowl Analytics with polars
Key Takeaways:- Learn how to use the polars Python package for efficient data analysis.
- Solve real-world analytical questions using Super Bowl data like has the NFL evolved into a passing league, do player ratings from organizations like Madden correlate with performance on the field, and do defenses really win championships?
- Understand how to apply calculations and basic ML techniques to sports analytics problems and make your own model to predict this year’s super bowl winner.
Description
The Super Bowl isn’t just a spectacle—it’s also a rich source of data for uncovering patterns, testing hypotheses, and practicing your analytics skills. In this hands-on session, you’ll use the polars Python package to analyze historical Super Bowl data, answering real questions about team performance, scoring trends, and more.
In this code-along, Will Bracken, a Data Science Manager at MongoDB, will guide you through solving data analysis problems using polars, a lightning-fast DataFrame library built for modern analytics. You’ll explore how to manipulate and calculate metrics on large datasets, and even apply simple machine learning models to American football data.
Presenter Bio
Will runs the data team supporting Sales and Go-To-Market Operations at MongoDB. He has spent a decade as a data scientist, with stints at PwC and AIB. His favorite football team is a closely guarded secret.