Three games, three moneylines, one repeating 6% edge — the model isn't hiding behind totals today, it's just picking winners.
No totals today. No run lines either. All three top plays on the slate are straight moneyline bets, and all three carry the same grade (A), the same confidence score (80), and — this is the part that made me double-check my spreadsheet — the exact same 6.0% edge. That's not a coincidence worth overthinking, it's just where the model landed three separate times. What differs is the shape of each game underneath the number. One's a pitching mismatch with a bullpen gap you could drive a truck through. One's a coin-flip start with everything else running cold. One's a legitimately ugly, low-scoring affair where the visiting bullpen number is the worst figure on the page. Let's get into them.
WSH to win outright at +105, Grade A, confidence 80, 6.0% edge, and the model likes the Nats a lot more than plus money suggests — 66% to win this one straight up at Citi Field. Getting paid on a favorite that size is the entire play.
Andrew Alvarez takes the ball for Washington with a 2.98 FIP. Robert Stock counters for the Mets at 4.6 FIP. That's not a subtle gap, and the components back it up — pitching sits at 31, suppressed, which tracks with a start-quality mismatch tilted hard toward the road team. Where it gets loud is the bullpen number: 86, elevated, the single highest figure across all three plays today. That combination is why the model's total balloons to 9.9 against a market total of just 8.5, with a projected split of 6.1 runs for Washington against 3.8 for New York. Offense is neutral at 57, park/weather neutral at 47, rest neutral at 50 — park score 34, weather 37. None of that is doing the heavy lifting here. This is a starter-and-bullpen story, full stop.
HOU to win outright at -124, Grade A, confidence 80, 6.0% edge. Model gives the Astros 62% to beat Seattle at Daikin Park, and paying -124 for it clears the same bettable edge as the Nats play above — just a very different-looking game to get there.
George Kirby starts for Seattle at 3.62 FIP, Peter Lambert for Houston at 3.68 FIP. Basically a wash on paper, and the components agree there's nothing exceptional happening on either side of this one. Offense grades 32, pitching 30, bullpen 38 — all three suppressed, which paints this as a quiet game across the board rather than a game decided by one standout unit. Park/weather sits neutral at 53 (park score 47, weather 35). The model's total comes in at 8.7 against a market 8, tight, with Houston projected to edge it 4.8 to 3.9. This is a lean built on the aggregate, not on any single loud number.
CLE to win outright at -127, Grade A, confidence 80, 6.0% edge — same number a third straight time. Model has the Guardians at 62% to handle San Diego at Progressive Field.
Michael King starts for the Padres at 3.98 FIP, Gavin Williams for Cleveland at 3.37 FIP, edge to the home arm there. But the number that jumps off the page is the bullpen component: 5. That's not suppressed, that's buried — the lowest figure anywhere on today's board. Offense checks in at 28 and pitching at 30, both suppressed as well, so this profiles as a low-scoring affair from top to bottom. The model's total sits at 6.6, well under the market's 7.5, with Cleveland projected to win a defensive slog 3.7 to 2.9. Park score 38, weather 41 — Progressive Field isn't pushing this number either direction. This is a game about a thin bullpen and light bats, and the model's siding with the home team to survive it.
Three plays, three identical 6% edges, zero totals or run lines in the mix. When the model repeats a number that cleanly across unrelated games, it's not being lazy — it's telling you where its conviction threshold actually sits today. Bet accordingly, and remember a 6% edge is real money over a season, not a lock on any single night.