MLB September 8 Full Slate

The September 8 card is a pitcher card disguised as a Tuesday. Fifteen games, twelve with confirmed lineups by late evening, and the board's real edges are almost all on the mound: the model's biggest legitimate strikeout edges sit in Philadelphia, Los Angeles, and Kansas City, while the sim's biggest edge on the whole slate is also its biggest lie - Corbin Burnes priced as prime Burnes despite a 464-day absence.

Three tool patterns drive the evening's reads. The sim anchored Burnes to his historical profile (a 6.15 strikeout projection for a pitcher on a 65-80 pitch return cap). The Texas-Seattle game was simulated with neither starter in it. The win model prices Alcantara and Misiorowski on name rather than current shape. And the K%-heavy matchups inflate totals. Every place we go against a tool tonight traces to one of those four.

Data Freshness

• Schedule: fresh, synced Sep 8 (15 games)
• Odds: FD board synced ~3 PM ET; DK sweep ~8:55 PM ET (29 credits left on key)
• Props: K props fresh on FD and DK; batter boards from the DK sweep
• Lineups: 12 of 15 games confirmed as of 10:15 PM ET (CHC, SF, SEA posted late; WSH@SD still open)
• Weather: clean slate - worst is 9% precip in San Francisco; domes in MIL, MIA, SD
• News: re-synced ~8:45 PM ET; nothing material on card legs
• Simulation: rerun on confirmed lineups ~10:30 PM ET

Slate Overview

The board splits into a pitcher night (Wesneski, Wacha, Painter) and a price board (Caminero plus money, Bell's boring edge, the HR bonus). One game stays fully gated (TEX@SEA at publish time) and one total is flagged as research rather than a play (LAA@BOS). The DK 50% SGP is the premium anchor - all three legs verified against the fresh sims.

Game-by-Game Breakdown

CLE @ BAL (10:35 PM EDT, CLE +104, BAL -121, O/U 8.5)

Tanner Bibee against Brandon Young with the lineups in. Bibee's over 5.5 at plus-125 prices a number the K-distribution says is dead (60.8% vs a 61.7% break-even), and Young's over 3.5 at -129 is thin against Cleveland's 23.2% K%. The only real lever is Baltimore's 28.8% team K% feeding Bibee's under at plus EV - but under props at -161 are not our lane.

Targets: no stake. Bibee under 5.5 as a track note at -161.

HOU @ PHI (10:40 PM EDT, HOU +120, PHI -142, O/U 8.5)

The best play on the slate. Wesneski's 0.89 K/IP is trending up on an 87-pitch leash, the fresh confirmed-lineup sim puts P(4+ K) at 76.1%, and the K-distribution's 69.6% agrees. DK's -122 is the best price of the books. The honest risk is the attack-K lane's recent misses, which share exactly this shape - projections clearing 3.5 by a full strikeout landing on 1-3 K nights.

Painter over 4.5 at -125 is the secondary angle with a 15-point K-distribution edge, legit but half the Wesneski edge on a worse price. Philadelphia's ML at -142 prices at 58.9% break-even against a 56.1% sim - a thin leg that only works inside the boosted SGP.

Targets: Wesneski over 3.5 K (DK -122, 0.5u, core). PHI ML only inside the DK 50% SGP, never solo.

NYM @ MIA (10:40 PM EDT, NYM +108, MIA -126, O/U 8.0)

The model says Mets 60.5%, the ensemble says 50.5%, and the market has New York at plus-108. The disagreement is real, but one signal is not a stake - the model's edge here leans on exactly the kind of stale-input artifacts we are catching elsewhere. Alcantara's K-distribution row is negative in both directions (his 0.44 K/IP is the collapse the ensemble is pricing). Manaea's under at -166 prices a number the sim puts at 79% - the wrong side of the price.

Targets: pass. The conflict resolves at first pitch.

MIN @ DET (10:40 PM EDT, MIN +120, DET -142, O/U 8.5)

Josh Bell from the 5-hole against righty Drew Anderson is the boring play that pays: 72.2% on the hit board, 69.8% in the sim, 66.9% break-even at -202. Three sources, room to spare, and 58 cents cheaper than the leadoff name with the identical edge.

Kremer and Anderson K props are coin flips at their lines (both distributions within 3 points of break-even). Detroit's fresher pen and Minnesota's cold pulse tilt the game, but -142 is a price, not an edge - the model edge is only 2.9 points.

Targets: Bell over 0.5 hits (DK -202, 0.25u). Pass on the game side.

LAA @ BOS (10:45 PM EDT, LAA +122, BOS -144, O/U 7.5)

The sim total is 10.2 with 72% over 7.5 - and that number is inflated by the Angels' league-high 28.8% K% feeding the sim's contact model. Both starters' K/IP trends are flat to down. This is the K%-overreaction pattern running in reverse, and it is the next pattern to systematize - but totals are not our lane yet, so it is a flag, not a play.

Targets: no stake. Over 7.5 flagged for the research ledger.

COL @ NYY (11:05 PM EDT, COL +280, NYY -344, O/U 8.0)

Hughes over 4.5 is dead at the price (51.4% vs a 61.7% break-even) and Schlittler's over 7.5 carries a thin 6-point edge at -129. The game side is -344. Nothing here survives the prices.

Targets: pass both sides.

TB @ ATL (11:15 PM EDT, TB -102, ATL -116, O/U 8.5)

Peralta against Smith-Shawver is the slate's biggest starter gap at this price band - Peralta's sim puts P(4+ K) at 75% while Smith-Shawver projects 2.79. But Peralta's line is 4.5 at -117 with a negative distribution edge, so the prop is dead even where the matchup lives. The Rays ML at -102 prices at 52.4% break-even against a 51.5% sim - no edge.

The play here is the batter board: Caminero over 1.5 total bases at +187. The sim's 2+ hit rate (32.1%) plus his 25.2% HR rate implies about 45% for 2+ TB against a 34.8% break-even. The best plus-money on the board, gated on price verification - the DK sweep snapshot is three hours old and this market moves.

Targets: Caminero over 1.5 TB (DK +187, 0.25u, verify live price; dead below +160).

AZ @ KC (11:40 PM EDT, AZ -120, KC +102, O/U 9.5)

The sim's biggest artifact. Arizona is priced at 54.1% on a Burnes projection (6.15 K) that is impossible for a 65-80 pitch cap of rust after 464 days away. The model's own board flags DEBUT and IL-GAP - and its +60-point Burnes over EV is exactly the artifact class we pass on. Burnes over 2.5 at -175 is a pass regardless of the board number.

The live side is Kansas City at plus-102 against a pen game for Arizona after the opener: AZ's pen covered 7.1 innings yesterday and 4.3 per game over three days, against a rested Royals order with Jensen at .43 HR/g. Wacha over 4.5 at +129 is the model's top legitimate board edge (60.1% vs 41.5% break-even), haircut for Arizona's 22.1% contact profile.

Targets: Wacha over 4.5 K (+129, 0.25u). KC ML at +102 as the pick-against-the-sim frame (0.25u). Burnes anything: pass.

CHC @ MIL (11:40 PM EDT, CHC +184, MIL -222, O/U 7.5)

The HR bonus game. The sim crowns PCA at .512, but the matchup layer flips it: PCA is 2-for-13 with seven strikeouts against Misiorowski, while Bregman is 3-for-11 with a home run and .545 slugging. Both orders are confirmed (PCA leads off, Bregman third). Fair prices run about -105 for PCA and +155 for Bregman - buy whichever book pays past fair.

The K sides are dead: Peterson's over has a 21-point distribution edge against a 6-point independent negative (direct conflict, pass per rule), and Misiorowski's 8.5 line prices his K/IP decline. The Cubs ML at +184 is the contrarian frame the win models cannot see - name-priced home favorite against a K/IP trending down - but it is a research frame, not a stake.

Targets: HR bonus Bregman over PCA (fair ~+155, 0.25u each, both books bonus-eligible). Devers third on the board.

PIT @ CWS (11:40 PM EDT, PIT +124, CWS -147, O/U 8.5)

A true coin flip (sim 50.2/49.8) with the South Side pen most taxed on the slate (5.1 IP/g over three days) and Pittsburgh's order finally whole. Chandler and Burke K props are both dead at their prices.

Targets: PIT +124 as the scout-dog frame (0.25u, structure lane).

WSH @ SD (1:40 AM EDT, WSH +158, SD -188, O/U 7.5)

Mize's over is dead (51.1% vs 61.7% break-even) and Cornelio has no FD line. Both lineups unconfirmed at publish. The only live number is San Diego's rested pen (75.6p against Washington's 34.8p), which supports the favorite but at -188 there is no edge to buy.

Targets: pass.

TEX @ SEA (1:40 AM EDT, TEX +116, SEA -133, O/U 8.0)

Fully gated at publish time: the sim ran with neither starter in the game - the same missing-input artifact class as the Burnes pricing. TEX's probable (Quantrill) and Seattle's side both posted after analysis. Re-run before any stake; no play until the artifact is corrected.

Targets: no stake. Re-sim before lock.

TOR @ ATH (1:40 AM EDT, TOR -178, ATH +152, O/U 9.5)

Soriano's K row is negative in both directions and Perkins' over 3.5 at -156 carries a 16-point edge that the 21.2% Toronto K% argues against - and the price sits at the -160 limit. The total is the slate's highest at 9.5 with a 65% over sim on an 11.6 projection, but totals are not our lane.

Targets: pass. Perkins over 3.5 track-only at the price limit.

STL @ SF (1:45 AM EDT, STL -104, SF -112, O/U 7.5)

Lineups unconfirmed at publish, which gates the Devers HR bonus play. Mathews and Roupp K rows are both negative. The sim/ensemble split (STL 52.5 vs SF 51.2) resolves on profiled-pitcher logic: Roupp's 0.78 K/IP over a 4-walk sim shape from the unprofiled Mathews, with Devers home at .5 HR/g L20. No stake until nines post.

Targets: pass. Devers HR stays bonus-board only.

CIN @ LAD (2:10 AM EDT, CIN +270, LAD -333, O/U 8.0)

The Skubal under 8.5 at -121 is the second play on the card: K-distribution 65.0% under, sim projects 6.46 against the 8.5 line, and the game log [6, 6, 7, 11, 7, 6] shows one 11-K outlier carrying the market's price. Cincinnati's 28.7% K% is the counterweight - that is why it is a quarter unit and not more. Lodolo's side is a direct conflict (distribution +18 points, independent -2) and passes per rule.

Targets: Skubal under 8.5 K (FD -121, 0.25u).

Show Your Work

The picks below are not a black box. Here is the model's number, the band around it, what we are assuming, the variables that actually decide it, and how we will grade it after first pitch.

Wesneski over 3.5 K (DK -122, 0.5u) — attack

• Model: K-distribution 69.6% over, +23.5 EV per $100, blended mean 4.71. Fresh confirmed-lineup sim P(4+ K) 76.1% on a 5.19 projection. Independent pitch-chain: 0.89 K/IP on an 87-pitch leash → ~4.8.
• Confidence band: 62–76% over. The floor is the attack-K lane's recent misses (projections clearing 3.5 by a full strikeout that landed 1–3 K); the ceiling is the sim's 76.1%.
• Assumptions: No early hook (Philly's pen is rested, so no early-exit tailwind either way). Wesneski holds his 0.89 K/IP trend rather than regressing to his floor games.
• Key variables: Pitch count (~87 leash), PHI team K% (24.7, mid), and whether he clears 5 innings.
• Grade it by: 4+ K is a win. If he lands exactly 3, the hook/floor was the miss — that is the shape we flagged, not a model failure.

Skubal under 8.5 K (FD -121, 0.25u) — lean

• Model: K-distribution 65.0% under, +18.8 EV; sim projects 6.46 against the 8.5 line. Game log [6, 6, 7, 11, 7, 6] — cleared 7 once in six starts.
• Confidence band: 60–70% under. The 11-K outlier is the market's anchor; the honest range is wide because Cincinnati's 28.7% K% is a real over-threat.
• Assumptions: The 11-K game is the outlier, not the trend. Skubal does not repeat it against a high-K lineup.
• Key variables: CIN team K% (28.7, second-highest), Skubal's swing-and-miss rate on the night, and whether he works deep enough to threaten 8.
• Grade it by: 7 or fewer K is a win. An 8+ K night means the CIN K% counterweight won — that is why this is a quarter unit, not a core.

Bell over 0.5 hits (DK -202, 0.25u) — lean

• Model: Hit board 72.2% with 4.1 expected AB; confirmed-lineup sim 69.8% from the 5-hole. Break-even at -202 is 66.9%.
• Confidence band: 68–74%. Tight band — this is a low-variance, thin-edge shape, not a swing.
• Assumptions: Bell stays in the confirmed 5-hole against righty Drew Anderson. No late scratch.
• Key variables: Lineup slot (5th, confirmed), opposing hand (RHP), and 4+ plate appearances.
• Grade it by: 1+ hit is a win. A 0-for is a variance miss on a correctly-priced play — the edge was real, the outcome wasn't.

Wacha over 4.5 K (FD +129, 0.25u) — lean

• Model: K-distribution 60.1% vs 41.5% break-even, +36 EV per $100. K/IP trend 0.52 → 1.09 over his last six.
• Confidence band: 55–65% over. The haircut is Arizona's 22.1% team K% (third-lowest in baseball) — a contact team that caps the ceiling.
• Assumptions: Wacha's K/IP uptrend is real, not a small-sample blip. The +129 pays for the contact-team risk.
• Key variables: AZ team K% (22.1), Wacha's K/IP trend, and whether he gets 5+ innings.
• Grade it by: 5+ K is a win. Exactly 4 K is his median — the +129 is what makes that a coin-flip worth taking.

Caminero over 1.5 TB (DK +187, 0.25u) — lean

• Model: Sim 2+ hits 32.1% plus 25.2% HR rate → ~45% for 2+ TB. Break-even at +187 is 34.8%.
• Confidence band: 40–50%. The band is wide because the price is a three-hour-old snapshot and total-bases markets move.
• Assumptions: The +187 price holds at lock. Under +160 the edge is gone and the play is dead.
• Key variables: Live line at lock, Caminero's lineup slot, and the opposing starter's handedness.
• Grade it by: 2+ total bases is a win. If the line moved under +160 before lock, the play was correctly voided — that is the gate, not a loss.

HR bonus: Bregman over PCA (fair ~+155 / ~-105, 0.25u each) — watch

• Model: Sim crowns PCA at .512, Bregman at .354. Head-to-head flips it: PCA 2-for-13 with 7 K vs Misiorowski; Bregman 3-for-11 with a HR and .545 SLG.
• Confidence band: Sub-52% events even at their sims. This is lottery pricing with a matchup edge, not a model edge.
• Assumptions: The head-to-head sample (13 and 11 AB) is meaningful, not noise. Both orders hold (PCA leadoff, Bregman third).
• Key variables: Real app price vs fair marker, and whether the book pays past fair on either name.
• Grade it by: A HR is a win. A miss is expected — the play is the price, not the probability.

Slate Shortlist

Best strikeout props: Wesneski over 3.5 (DK -122, three models), Painter over 4.5 (-125, secondary), Wacha over 4.5 (+129, top board edge), Skubal under 8.5 (-121).

Best 2+ hit candidates: Bell over 0.5 hits (DK -202, three sources), Jeffers (confirmed cleanup), Caminero over 1.5 TB (+187, price-verify).

Best correlated builds: DK 50% SGP - Wesneski o3.5 K + PHI ML + Schwarber o1.5 TB (+540 computed, all legs sim-verified).

Notes Before Lock

• Caminero's total-bases price is from the ~9 PM DK sweep - verify the live line at lock; dead below +160
• Burnes workload remains unsourced by design - no K stakes on AZ@KC until a sourced pitch cap exists
• HR bonus plays need real app prices on both books before staking; fair markers are PCA ~-105, Bregman ~+155
• TEX@SEA was simulated with no starters - re-sim before trusting any number from that game
• The LAA@BOS total is a research flag, not a play - totals are not our lane yet
• All K props priced from FD/DK syncs ~9 PM ET; late scratches void the under-market reads

TL;DR

• Wesneski over 3.5 K (DK -122, 0.5u) — three models converge, best price of the books
• Skubal under 8.5 K (FD -121, 0.25u) — market prices the name, the log says under
• Bell over 0.5 hits (DK -202, 0.25u) — three sources, confirmed 5-hole slot
• Wacha over 4.5 K (+129, 0.25u) — model's top legit edge, contact-team haircut
• Caminero over 1.5 TB (DK +187, 0.25u) — best plus-money on the board, verify the line
• HR bonus: Bregman over PCA, Devers third — matchup beats sim; fair ~+155 / ~-105
• DK 50% SGP: Wesneski o3.5 + PHI ML + Schwarber o1.5 TB (+540) — every leg sim-verified
• KC ML +102 — the pick against the sim's Burnes artifact (frame, 0.25u)
• Pass: Burnes anything, TEX@SEA (no-starter sim artifact), NYM@MIA (model conflict)

Updates

• Revision 2 (11:45 PM EDT, September 8): Added Show Your Work — confidence bands, assumptions, key variables, and grading criteria for every staked play. No picks, prices, or units changed.
• Revision 1 (11:10 PM EDT, September 8): Initial evening look. 12 of 15 lineups confirmed (CHC, SF, SEA posted late; WSH@SD open). Full model/sim/packet pipeline run; fresh confirmed-lineup sims at ~10:30 PM ET. TEX@SEA flagged as no-starter sim artifact - game gated. Five staked plays (Wesneski core 0.5u, four 0.25u), HR bonus board priced to fair-value markers, DK 50% SGP verified leg-by-leg.