MLB Predictor
Baseball prediction and calibration project.
problem
Sports models are easy to make overconfident. The useful work is less about the prediction itself and more about calibration, data hygiene, and learning when a model is simply guessing loudly.
context
This is a weekend project, not a serious product. Baseball is a convenient sandbox because the data is rich, the season is long, and small edges tend to disappear fast.
what I built
I built a small modeling workflow around public baseball data for win probability, first-inning run predictions, and prop-style edges. I check outputs against real results and calibration plots.
- Predicts win probability, first-inning runs, and a few prop-style outcomes
- Uses public game data and calibration checks rather than vibes
- Kept as a practice project for probability, modeling, and evaluation
what I learned
Most of the work is data plumbing. The model is only interesting after the inputs, evaluation, and calibration story are honest.
status
Weekend ML project and notes, kept deliberately low-stakes.