How the 2026-09-01 predictions scored against actual results
Cumulative Δlog-loss vs always-pick-home: -0.0194 ± 0.0084 per game over 240 games (paired; negative = beating the baseline - currently ahead).
| Yesterday | Correct | Log-loss |
|---|---|---|
| model | 5/15 | 0.707 |
| always-pick-home | 8/15 | 0.691 |
Daily detail: 33% hit rate · Brier 0.257 · avg margin error 4.2 · avg total-runs error 4.6. Running: 141/240 (58.8%) · cumulative log-loss 0.671 · cumulative Brier 0.239
| Game | Pick | Predicted | Actual | |
|---|---|---|---|---|
| ❌ | Phillies @ Diamondbacks | Diamondbacks (50%) | 4.7–4.6 | 1–7 (home first) |
| ❌ | Mariners @ Red Sox | Red Sox (59%) | 4.9–4.2 | 6–9 (home first) |
| ❌ | Brewers @ Cubs | Cubs (53%) | 4.8–4.7 | 4–9 (home first) |
| ❌ | Padres @ Reds | Padres (52%) | 4.7–4.2 | 4–3 (home first) |
| ✅ | Blue Jays @ Guardians | Guardians (60%) | 4.6–3.8 | 6–1 (home first) |
| ❌ | Orioles @ Rockies | Orioles (61%) | 5.4–3.9 | 4–2 (home first) |
| ❌ | White Sox @ Astros | Astros (53%) | 4.6–4.5 | 1–5 (home first) |
| ❌ | Marlins @ Royals | Royals (54%) | 4.5–4.3 | 3–6 (home first) |
| ✅ | Yankees @ Angels | Yankees (63%) | 5–3.2 | 3–7 (home first) |
| ❌ | Cardinals @ Dodgers | Dodgers (60%) | 4.8–4 | 8–13 (home first) |
| ❌ | Tigers @ Twins | Tigers (53%) | 4.7–4.1 | 15–2 (home first) |
| ✅ | Giants @ Pirates | Pirates (58%) | 4.6–4 | 13–12 (home first) |
| ✅ | Mets @ Rays | Rays (57%) | 4.5–4 | 6–2 (home first) |
| ✅ | Athletics @ Rangers | Rangers (61%) | 5.1–4.2 | 8–5 (home first) |
| ❌ | Braves @ Nationals | Braves (55%) | 4.7–3.9 | 9–5 (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