MoneyballVo

Steven Treadway | MLB Analytics | Building Statcast prediction models |

[MoneyballVo @MoneyballVo]

⚾️ MLB Predictions | 2026.06.14 ⚾️

[MoneyballVo @MoneyballVo]

⚾️ MLB Predictions | 2026.06.13 ⚾️

[MoneyballVo @MoneyballVo]

⚾️ MLB Predictions | 2026.06.09 ⚾️

[MoneyballVo @MoneyballVo]

⚾️ MLB Predictions | 2026.06.08 ⚾️

[MoneyballVo @MoneyballVo]

⚾️ MLB Predictions | 2026.06.08 ⚾️

[MoneyballVo @MoneyballVo]

⚾️ MLB Predictions | 2026.06.07 ⚾️

[MoneyballVo @MoneyballVo]

⚾️ MLB Predictions | 2026.06.06 ⚾️

[MoneyballVo @MoneyballVo]

⚾️ MLB Predictions | 2026.06.05 ⚾️

[MoneyballVo @MoneyballVo]

⚾️ MLB Results | 2026.06.04 ⚾️

Buxton HR (highest HR probability) 🚀
Pirates ✅
DBacks ✅

[MoneyballVo @MoneyballVo]

⚾️ MLB Predictions | 2026.06.04 ⚾️

[MoneyballVo @MoneyballVo]

⚾️ MLB Predictions | 2026.06.03 ⚾️

[MoneyballVo @MoneyballVo]

⚾️ MLB Predictions | 2026.06.03 ⚾️

[MoneyballVo @MoneyballVo]

Yesterday marked the first day of predictions from our latest model. Here's a look at where the previous model stood:

📊 Model Performance Summary (28 games)
Win rate: 50.0% (14/28)
Total P&L: +0.02U
ROI: 0.001U/game

By edge bucket:
4-8%: 9/22 (40.9%) | -4.69U
8-15%: 3/4 (75.0%) | +1.87U
15-100%: 2/2 (100.0%) | +2.84U

The 4-8% bucket was dragging everything down. I retrained the model with some tuning, but more importantly discovered that pitching data for several teams wasn't properly backfilled, which I believe was hurting performance more than the model itself.

Both issues are now fixed. Will be monitoring closely going forward. Still a very small sample size — the real test starts now.

[MoneyballVo @MoneyballVo]

⚾️ MLB Predictions | 2026.06.02 ⚾️

* sorting through some missing bullpen data issues. these are updated predictions after fix.

[MoneyballVo @MoneyballVo]

⚾️ MLB Predictions | 2026.06.02 ⚾️

[MoneyballVo @MoneyballVo]

⚾️ MLB Predictions | 2026.06.01 ⚾️

[MoneyballVo @MoneyballVo]

⚾️ MLB Predictions | 2026.05.31 ⚾️