MLB Slate — 2026-09-03

Predictions for every game on 2026-09-03 (ET)

Matchup (home in bold)SP (away vs. home)SP adj (Elo, away/home)PickWin probExp. runs (winner first)Exp. total
Red Sox @ OriolesJake Bennett vs. Brandon Young+16.6 / +1.0Red Sox55%4.9–4.19.0
Brewers @ CubsLogan Henderson vs. Kevin Gausman+30.2 / +6.4Brewers54%5.2–4.59.6
Blue Jays @ GuardiansJosé Soriano vs. Tanner Bibee-5.8 / -0.7Guardians54%4.3–4.28.5
White Sox @ AstrosLuis Castillo vs. Hunter Brown+5.4 / +20.9Astros55%4.6–4.38.9
Marlins @ RoyalsSandy Alcantara vs. Michael Wacha+5.7 / +14.5Royals52%4.6–4.58.9
Cardinals @ DodgersQuinn Mathews vs. Tarik Skubal+1.9 / +20.9Dodgers64%5.2–4.09.2
Giants @ PiratesBlade Tidwell vs. Lake Bachar-15.9 / +10.0Pirates60%4.9–4.19.0
Athletics @ MarinersJack Perkins vs. Kade Anderson-8.7 / +2.4Mariners64%5.4–4.29.6
Rays @ RangersShane McClanahan vs. Cal Quantrill+3.1 / -7.9Rays54%4.7–4.18.8

Starter identity IS a model input: each probable's rolling game score (exponentially weighted, 20-start half-life) is compared to his team's staff average and the difference enters the team's Elo at 3.0 points per game-score point (the SP adj column, so the effect is auditable per game). Rookies/thin history fall back to a shrunk staff rating; TBD uses the staff rating. Rest and travel add up to a few Elo points each. Win probability from Elo (+24 home advantage). Expected total is matchup-specific: each club's recent runs scored/allowed (exponentially weighted, ~20-game half-life, shrunk to league average) sets the run environment, and the Elo-implied margin is carved out of it. Expected runs are shown at one decimal; read them as averages, not a literal final score.

Betting-blind Elo model (K=3, +24 home, MOV-weighted, 60% season carryover) · Can Tre Beat Vegas · whosyurgoat.app/vegas/mlb