About ✦
About First Pitch
A daily digest of simulated MLB game predictions.
First Pitch publishes simulated MLB game predictions every day. Each morning before the slate starts, a handful of computational models run on every game, and their picks combine into the consensus prediction you see across the site.
The models pull from team batting stats, starting pitcher ERA, and announced lineups when MLB has released them. Each model weighs those inputs differently, which is why they sometimes disagree — and why we publish a consensus rather than any single take.
— The Simulations —
- BASELINE
- Monte Carlo simulation using career batting probabilities only.
- PITCHER-WEIGHTED
- Monte Carlo with pitcher-quality adjustment based on ERA.
- PLATOON-WEIGHTED
- Monte Carlo with handedness platoon adjustments.
- PITCHER + PLATOON
- Monte Carlo combining pitcher-quality and platoon adjustments.
- LOGISTIC
- Logistic regression model trained on historical game data.
- RANDOM FOREST
- Random forest model trained on historical game data.
The consensus combines these as a weighted average. ML models currently carry slightly more weight than the Monte Carlo variants — that's tunable and will shift as each model earns or loses trust over the season. Individual sim results stay visible on each game's detail page so you can see when the models disagree.
— Caveats —
These are simulations and they have limits. Early in the season, models lean toward league averages because there isn't much current-year data yet. When MLB hasn't published a starting lineup, the prediction falls back to a best guess — those picks are tracked separately under Lineup impact on the Track Record page.
Injuries, weather, and bullpen state aren't factored in beyond what surfaces in the basic stats. Predictions are also deterministic for a given slate: once they're posted in the morning, they don't update mid-day.
— For Fun, Not For Betting —
Baseball is hard to predict. First Pitch is built for friends and family to follow along — please don't bet money on what you see here.