How the 2026-09-25 predictions scored against actual results
Cumulative Δlog-loss vs always-pick-home: -0.0216 ± 0.0063 per game over 444 games (paired; negative = beating the baseline - currently ahead).
| Yesterday | Correct | Log-loss |
|---|---|---|
| model | 10/17 | 0.669 |
| always-pick-home | 11/17 | 0.675 |
Daily detail: 59% hit rate · Brier 0.239 · avg margin error 3.0 · avg total-runs error 4.5. Running: 263/444 (59.2%) · cumulative log-loss 0.670 · cumulative Brier 0.239
| Game | Pick | Predicted | Actual | |
|---|---|---|---|---|
| ❌ | Astros @ Athletics | Astros (56%) | 5.1–4.1 | 6–5 (home first) |
| ✅ | Cubs @ Red Sox | Red Sox (52%) | 4.5–4.4 | 4–3 (home first) |
| ✅ | Rockies @ White Sox | White Sox (66%) | 5.6–4.1 | 6–1 (home first) |
| ✅ | Pirates @ Tigers | Tigers (55%) | 4.4–4.1 | 8–7 (home first) |
| ✅ | Guardians @ Royals | Guardians (52%) | 4.6–4.2 | 9–12 (home first) |
| ❌ | Braves @ Marlins | Braves (50%) | 4.5–4.2 | 3–0 (home first) |
| ✅ | Cardinals @ Brewers | Brewers (66%) | 5.3–3.8 | 8–4 (home first) |
| ❌ | Rangers @ Twins | Rangers (51%) | 4.6–4.2 | 10–2 (home first) |
| ❌ | Orioles @ Yankees | Yankees (62%) | 4.7–3.7 | 2–10 (home first) |
| ❌ | Rays @ Phillies | Phillies (55%) | 4.4–4.1 | 0–2 (home first) |
| ❌ | Diamondbacks @ Padres | Padres (57%) | 4.7–4.2 | 4–11 (home first) |
| ❌ | Angels @ Mariners | Mariners (59%) | 4.7–4 | 5–7 (home first) |
| ✅ | Dodgers @ Giants | Dodgers (61%) | 5.2–3.7 | 0–2 (home first) |
| ✅ | Reds @ Blue Jays | Blue Jays (56%) | 4.6–4.3 | 6–5 (home first) |
| ✅ | Mets @ Nationals | Nationals (52%) | 4.6–4.5 | 7–6 (home first) |
| ✅ | Cubs @ Red Sox | Red Sox (52%) | 4.5–4.4 | 2–0 (home first) |
| ✅ | Orioles @ Yankees | Yankees (58%) | 4.6–3.9 | 6–3 (home first) |
The headline is the paired per-game log-loss difference against a fixed always-pick-home forecast (p=0.534) on the same games, ± one standard error; daily hit rate is noisy and demoted on purpose. Reference points: 0.693 is a coin flip; the model's walk-forward backtest runs 0.680 (research/SP-BACKTEST.md).
Betting-blind Elo model (K=3, +24 home, MOV-weighted, 60% season carryover) · Can Tre Beat Vegas · whosyurgoat.app/vegas/mlb
This email online: 2026-09-25 grade · Previous: 2026-09-24 · All MLB emails