Data is the day job. Curiosity is the common thread.
Connect with me on LinkedIn if you want to follow along or get in touch.
My professional background sits at the intersection of data analytics, marketing analytics, and data engineering. I've worked with SQL and Python, built reporting workflows and dashboards, connected data from different platforms, and helped make tracking and attribution more useful for decision-making.
That work has shaped how I approach this project. I like understanding where data comes from, how it can be made more reliable, and what it can actually tell us. A model is only as useful as the data and assumptions underneath it, so I care about the whole process, not just the final prediction.
Why college basketball?
I've always enjoyed the mix of strategy, uncertainty, and personality that makes college basketball fun to follow. The numbers tell part of the story, but every season is full of odd matchups, unexpected runs, and games that make you rethink what you thought you knew.
This project gives me a way to bring my professional skills into a personal interest: build a repeatable data pipeline, test a model, share the predictions, and be honest about the results.
A little friendly competition
The project started when I saw my brother still manually entering game results into a spreadsheet that was about ten years old, relying on a limited set of features to make his predictions. I wanted to build a more automated pipeline, bring in a broader set of inputs, and make it easier to track how the predictions actually perform.
What I'm interested in
Outside the model, I'm interested in sports analytics, building useful data products, and finding ways to make complex information easier to act on. This site is a home for that ongoing experiment, including the wins, the misses, and the lessons between them.