Writing. Data visualization. Analysis.
Empty Arenas: A Shooter's Dream?
The NBA had a unique 2020 season, with about half of the games being played in arenas with no fans. Shri and I use this singular setting to run some (somewhat controlled) studies on the impact of crowd size on shooting open 3-point shots
The Great MVP Race
Building a machine learning model to predict the future and evaluate the past of the NBA's most valuable players
The Hustle Index
Shri Tanna and I put together an interactive tool for creating your very own NBA hustle metric using stats like average speed, deflections, box outs, screen assists, contested shots, and charges drawn.
A Guide to Winning March Madness
Some non-obvious tips on winning a March Madness prediction pool. Like changing your picks based on the number of people in your pool and the best single-number metrics out there.
Why churn prediction ≠ churn reduction, and what to do instead
On the inherent shortcomings of churn prediction, and how customer retention can be improved with uplift modeling and dynamic point of cancellation offers.
Investors be Warned: Snapchat's IPO and Future
Snap Inc is preparing for its IPO this week. Technical shortcomings, poor monetization, and the looming giant Facebook will create problems for Snapchat if it doesn’t evolve quickly.