This article is part of the SaskPoly story. Read how it all started.
NLCS Game 3 Desk Card: The Machine Says Coin Flip, the Recipe Says the Market Is Right, and Two Dingers Priced Where the Form Credit Barely Clears
NLCS Game 3: Milwaukee leads 2-0 and the series shifts to Los Angeles; the loop passes at −1.7pp, the desk factor recipe regresses Snell's 42.2 innings down to the market's own number, the frozen PRO•LINE line sits 8.1 points below the US consensus, and the only two verified dinger prices are Freeman +680 and Contreras +820 — both graded against the house's 0-for-4 October on carded home runs.
The Milwaukee Brewers are two games from the World Series, and the defending champions are two games from the furniture. Milwaukee took the first two at home. Tonight the series moves to Los Angeles for Game 3 of the National League Championship Series, 6:00 p.m. MT, Robert Gasser against Blake Snell (VSiN). One game on the board. The desk card has no side bet and two dingers, and the argument for the pass is worth printing.
The game
| Gasser (MIL) | Snell (LAD) | |
|---|---|---|
| Season | 3.87 ERA, 1.19 WHIP, 8.5 K/9 | 1.90 ERA, 1.10 WHIP, 11.6 K/9 |
| Innings | a full season's work | 42.2 — nine starts, after missing three months |
Everything in Snell's row is the best pitching on either side tonight. Everything in his innings column is the reason the desk doesn't take that row at face value. Nine starts is a sample, not a season, and the recipe says so in numbers below.
Three instruments, one answer
The machine: pass. This morning's sheet has Los Angeles at 51.2% against its recorded market of 52.9% — a −1.7% edge, under the floor, no bet. The loop's read is Elo reading season strength: Milwaukee won 103 games and the model believes the clubs are even.
Honesty note, printed because the house prints these: the loop's recorded price feed doesn't match the real board. The sheet's PRO•LINE line (Dodgers 1.84 on the 3-way, Milwaukee 1.68 on the run line) has been frozen since it posted on October 13. The actual market is Los Angeles −180, Milwaukee +150, total 7.5 (VSiN) (Action Network). The frozen local line implies 52.9%; the US consensus implies 61.0%. That 8.1-point gap is the widest cross-market seam on this morning's sheet. It is also, on the local book, a 3-way price — the home leg pays only on a two-run win, and a one-run game is its own tie selection. The seam cannot be bought clean at the book of record, and the house does not card wagers it cannot grade at a published price.
The desk recipe: the same factor arithmetic the desk has run all October, applied to the actual matchup — Gasser's 3.87 ERA, 1.19 WHIP, 8.5 K/9 against Snell's 1.90, 1.10, 11.6. The starting-pitcher term hits its +12 cap on Snell's line, and the raw recipe makes Los Angeles 65.1%. Then the regression the machine doesn't do: shrink Snell's nine-start sample toward a league-average arm, and the number settles at 60–62%. The market's price is −180, about 61% no-vig. That's the market's number. When all three instruments land in the same place, the place is the market.
No side bet. Taken at face value the loop's 51.2% would make Milwaukee +150 look like a steal. It isn't, because it's reading the season, not the pitching. The recipe reads the pitching, and the pitching agrees with the price.
The dinger board
The house recipe, stated in the open: season home runs per plate appearance, times 4.3 plate appearances, times a starter adjustment. Gasser's 1.23 home runs per nine is a real number — multiplier ×1.05. Snell's 0.00 home runs per nine in 42.2 innings cannot be used literally; a zero is a sample-size accident at nine starts, regressed to roughly 0.85, multiplier ×0.73. The full board, raw fair probabilities, best price verified in parentheses (SI):
| Batter | Raw fair | Verified price | Breakeven |
|---|---|---|---|
| Muncy (LAD) | 23.4% | none found | — |
| Ohtani (LAD) | 21.9% | none found | — |
| Betts (LAD) | 19.7% | none found | — |
| Pages (LAD) | 18.1% | none found | — |
| W. Smith (LAD) | 16.4% | none found | — |
| Bauers (MIL) | 14.1% | none found | — |
| Chourio (MIL) | 13.7% | none found | — |
| Freeman (LAD) | 11.0% | +680 | 12.8% |
| Contreras (MIL) | 8.8% | +820 | 10.9% |
Two prices verified, both from DraftKings via SI. The board's two favorites have no price at all. No price, no leg — the scrape-gap rule the house has graded under all month. The math behind the board is the standing recipe, printed where it lives per the house explainer.
The card
1. Freddie Freeman home run, +680, 0.25u. The raw fair, 11.0%, is short of the 12.8% breakeven — so the leg stands on its form credit, named in the open: two home runs in six postseason games, and Gasser is a pitch-to-contact arm and a quick-hook candidate, 56 pitches for 12 outs in his NLDS start, with Milwaukee's unused long-relief options the least scary page in the book (VSiN). The park leads baseball with a 124 home-run index (Statcast). Form-weighted fair, about 13%. A whisper of an edge at the house floor. That is what this leg is, and the card says so.
2. William Contreras home run, +820, 0.25u. Same form-credit basis, shakier floor. Three home runs this postseason, a .783 slugging, 9-for-23. But Snell's stuff is the best on either side all night — 2.17 xERA, 1.72 FIP — and the strict fair is 8.8% against a 10.9% breakeven. The full form credit gets you to roughly 11–12%. Playable only if you accept up front that you are paying for the form, not the recipe. The card accepts it, at a quarter unit, and prints the split so you can decline it.
3. Ohtani, pass at any sane price. The recipe loves him — 21.9%, second on the board — but he has not homered since August 18 and is hitting .150 this postseason (NY Post). That is the exact veto the board cannot see: the arithmetic reads the season, the slump is the present. The house has been burned by form credits twice this week. It is not paying a premium for a third on the same night.
Where this bet type stands
Said plainly, because the ledger rule says the type gets graded, not hidden. The house's carded dingers are 0-for-4 this October. Last night José Ramírez did not homer — he made the fourth-inning error that let Tampa score three unearned runs in a 3–1 Rays win, while Freddy Peralta threw six scoreless — and the form-credit dinger has now lost twice in three nights. The machine leg cashed (+0.32u at 1.64), so the ALCS Game 2 card netted +0.07u, but the dinger half of the desk's October is still looking for its first win. Two quarter-unit whispers tonight, or pass entirely, is a defensible read either way. The card plays them. The disclosure rides along.
The running numbers, verified against this morning's sheet: 46–52 on positive-edge bets, +13.91u, mean Brier 0.2366 across 207 graded games — the full ledger lives in the open.
Grading pre-commit. Both dinger legs grade tonight against the prices on this page (+680 and +820, DraftKings via SI). A fill at a sub-floor price pulls the leg, per the standing rule. Tonight's results fold into tomorrow's grading piece with the ALCS Game 3 preview.
Bring your own pencil.