MLB · model accountability
Every prediction we've made on MLB games, with every result. Pre-game probabilities are frozen at the moment of prediction — the backtest never reads from a future rating.
Hit rate
56.7%
2370 / 4182 games
Home-always baseline
53.5%
Naive: pick home every time
Edge over naive
+3.2pp
What the model adds
Calibration gap
0.1pp
Avg pred 53.3% vs actual 53.5%
By confidence bucket
Buckets defined by edge from 50%: tossup < 4pp, lean 4–8pp, edge 8–15pp, lock ≥ 15pp.
| Bucket | Games | Wins | Hit rate |
|---|---|---|---|
| tossup | 1514 | 803 | 53% |
| lean | 1186 | 647 | 54.6% |
| edge | 1143 | 684 | 59.8% |
| ★ Lock | 339 | 236 | 69.6% |
Best calls
Highest-confidence games where the model was right. The receipts.
Worst misses
Highest-confidence games where the model was wrong. Honest accounting.
Recent results
BAL pick 61%
LAA 1 – 3 BAL
Aug 4, 26 · vs LAA
CIN pick 63%
ATH 4 – 5 CIN
Aug 4, 26 · vs ATH
CLE pick 63%
NYM 6 – 2 CLE
Aug 4, 26 · vs NYM
PHI pick 61%
WSH 0 – 5 PHI
Aug 4, 26 · vs WSH
NYY pick 59%
STL 0 – 2 NYY
Aug 4, 26 · vs STL
BOS pick 61%
CHW 2 – 14 BOS
Aug 4, 26 · vs CHW
ATL pick 58%
MIA 2 – 4 ATL
Aug 4, 26 · vs MIA
MIN pick 53%
MIN 2 – 8 KC
Aug 4, 26 · vs KC
MIL pick 58%
PIT 2 – 4 MIL
Aug 4, 26 · vs PIT
CHC pick 55%
LAD 1 – 5 CHC
Aug 4, 26 · vs LAD
TEX pick 58%
SF 4 – 5 TEX
Aug 4, 26 · vs SF
HOU pick 69%
TOR 2 – 7 HOU
Aug 4, 26 · vs TOR
How we report this
Every row above carries the probability we'd have shown you before first pitch — never a backfilled rating. A game counted as "model right" if the team we favored won, regardless of margin. Hit rate is total wins divided by total finals. No selection bias, no excluded results.
← Tonight's slate