MLB Grade — 2026-09-03

How the 2026-09-03 predictions scored against actual results

Cumulative Δlog-loss vs always-pick-home: -0.0160 ± 0.0081 per game over 264 games (paired; negative = beating the baseline - currently ahead).

YesterdayCorrectLog-loss
model5/90.682
always-pick-home6/90.673

Daily detail: 56% hit rate · Brier 0.244 · avg margin error 2.9 · avg total-runs error 2.4. Running: 151/264 (57.2%) · cumulative log-loss 0.676 · cumulative Brier 0.241

GamePickPredictedActual
✅Red Sox @ OriolesRed Sox (55%)4.9–4.15–6 (home first)
❌Brewers @ CubsBrewers (54%)5.2–4.52–1 (home first)
❌Blue Jays @ GuardiansGuardians (54%)4.3–4.23–6 (home first)
✅White Sox @ AstrosAstros (55%)4.6–4.36–2 (home first)
✅Marlins @ RoyalsRoyals (52%)4.6–4.57–3 (home first)
✅Cardinals @ DodgersDodgers (64%)5.2–43–2 (home first)
✅Giants @ PiratesPirates (60%)4.9–4.15–2 (home first)
❌Athletics @ MarinersMariners (64%)5.4–4.24–7 (home first)
❌Rays @ RangersRays (54%)4.7–4.16–0 (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