Wild Card Wednesday: Two Leans, One Pass, and a Starter Sample the Size of a Grapefruit

Four Game 2s on the wild-card Wednesday slate (MLB), three of them priceable. The desk's ledger is still dark — the season feed died at Game 1 — so every number below is a hand-rebuild of the factor recipe on pulled probables and published odds, the same rebuild that graded Tuesday's card 2-0 with a correct pass before a pitch was thrown today. Philadelphia at Atlanta starts at noon MT and goes untracked for a second straight day; that miss is logged, not papered over.

The board

The tickets, as placed

The card above is the model's opinion. The tickets are the owner's, and the desk logs the difference in the open rather than absorbing it silently — the same deviation discipline the survivor vertical runs on. Two moneylines were placed this morning, and neither is a card lean:

• Braves −106 against Philadelphia, the noon game the model never priced — a coverage bet with no ledger number behind it (PickDawgz).
• Yankees −139 against Boston — the game the rebuild passed at +0.6 percent and the house played anyway (TheSpread).

Stated plainly: one of these is a bet the model cannot grade, and the other is a coin flip it graded and passed. Both facts are published in the same place, on purpose. Both legs grade tomorrow against the prices on this page, on the same terms as every other leg — wins, losses, and untracked games logged separately, nothing retroactively claimed.

* Market column is no-vigged from TheSpread's published Game 2 prices (CWS–HOU) (BOS–NYY) (CHC–SD), not the PRO•LINE app — the desk's own scrape is dead for the postseason. Re-check every price before anything goes on the card.

The rebuild, in the open

The recipe, applied by hand: 50 plus home field (+4), a starter term built from ERA, WHIP, and K/9 differentials (weight 8/10/0.8, capped at ±12), a bullpen term (4× season bullpen ERA differential, capped at ±5), a record term (1.5× win-percentage differential), and the park. Probables and pitcher lines came off the pulled game pages (CWS–HOU) (BOS–NYY) (CHC–SD); bullpen ERAs off the season leaderboard (InsideThePen) — Yankees 3.15, Boston 3.27, San Diego 3.31, the White Sox 3.86, Houston 4.14 — with the Cubs' 4.06 from a second tracker (Screwball) because the primary table would not give the number up. Park terms: Yankee Stadium graded league-average and Petco a run suppressor in the mid-90s on a 100 scale (TheSpread) (TheSpread); Daikin Park treated as neutral, which costs at most ±0.02 either way.

The lean that rhymes with yesterday's pass

White Sox at Astros is the same matchup the ledger passed Tuesday — Houston 0.512 against a 0.521 market — and the pass graded when Chicago won Game 1, 6-3 (box). Today the market has moved the other way: the Astros opened −144 and are bet to −152 with 65 percent of tickets (TheSpread), and the rebuild says that is the wrong direction. Sean Burke (3.34 ERA, 1.195 WHIP, 9.7 K/9) is the steadier arm against Hunter Brown (3.19, 1.27, 10.4 — the 1.27 walks-plus-hits confirmed (ESPN)), Chicago owns the better record and a plus-56 run differential against Houston's 81-81 and minus-32, and the Astros bullpen — 4.14 on the season, 16th in baseball — is missing four of the eight arms it used Tuesday, with Blubaugh, King, Ullola, and Hendrickson all listed unavailable and Okert already out (TheSpread). The rebuild lands Houston 53.6 percent. The market asks 57.8. That 4.2-point gap is the card's cleanest edge, worth +5.3 percent expected value at +127.

The honest counterweights: Josh Hader (1.01 ERA, 26 consecutive saves) is very available, and Houston won the season series 5-2. Burke's Game 1 counterpart Grant Taylor threw a career-high 54 pitches Tuesday and is likely gone for today. And the line moving from −144 to −152 with the public is drift, not information — the desk's read is on the numbers, and the numbers say the wrong team is favored.

The pass, because the market got this one right

Red Sox at Yankees is the game the rebuild wishes it could bet — and cannot. Sonny Gray (2.72 ERA, 1.148 WHIP) and Max Fried (2.80, 0.994) are the best starting matchup of the day (TheSpread), and the recipe splits their differentials almost perfectly: Fried's ERA and K/9 edge, Gray's WHIP. New York owns the better record (93-68 against 87-75), the better bullpen (3.15 against 3.27), and home field. The rebuild lands the Yankees 56.3 percent. The market prices them at 55.7. A 0.6-point gap is not an edge; it is a rounding error with a commission attached. The line agrees with itself: it opened −129 and has hovered in the −136 to −142 band all morning (VegasInsider) with 80 percent of tickets on New York — drift, not a signal. Pass. Both bullpens are fresh, both lineups are short their best hitter — Judge for New York, Anthony for Boston (TheSpread) — and the desk's model is injury-blind anyway. If there is a sharp side in this game it is the total, and totals are not this card's business.

The lean the whole card hangs on

Cubs at Padres is where the discipline gets tested. The mechanical rebuild lands San Diego 69.1 percent against a 56.1 percent market — a 13-point gap, the largest the desk has published all postseason. The driver is the starter term, which pins at its +12 cap: Kevin Gausman (4.51 ERA, 1.27 WHIP, 9.3 K/9, and a left shoulder that was being tested at Petco two days ago) against Nick Pivetta (2.86 ERA in eight starts and 34.2 innings, an elbow that cost him five months, and a K rate that says the stuff never left) (plaintextsports) (TheSpread). The bullpen term adds three full points — San Diego's 3.31 and fresh off Michael King's seven shutout innings against Chicago's 4.06 and 25th-ranked corps — and Petco takes a sliver back.

The sensitivity the number deserves: regress Pivetta's 34.2 innings to league average and the starter term falls from +12 to +1.9; the same rebuild then lands San Diego 59.1 percent — a 3.4-point edge, thin on expected value, still over the desk's action line. The truth is inside that interval, and so is the bet-sizing answer. The counterweight is real: the Cubs scored more runs than anyone in baseball this season (850, a 114 wRC+) and took the season series 5-1 — though all six meetings came before July 25, and the Padres have gone 41-18 since (TheSpread). Both versions of the model clear the line. The card plays it small and says so.

The card

Two straight moneyline leans: White Sox +127, Padres −141. Combined at published prices it pays 3.88 decimal (+288), implied 25.8 percent; the rebuild's joint probability is 46.4 × 69.1 = 32.1 percent mechanical, or 27.4 percent with Pivetta regressed. The mechanical ticket carries roughly +6 points of expected value; the regressed one carries +1.6. The season's own calibration block says edges of 3–5 percent have been worth +18.4 percent ROI across 817 resolved bets, which sizes the straight leans; the stale-price caveat and a 34.2-inning starter sample size the parlay, if it is played at all, to a fraction of the straights.

Stated plainly, the risks against the card: Pivetta's elbow decides whether the Padres leg is a steal or a fair price, and no model reads elbows. Houston's Hader is the best reliever left in either league today and the White Sox lean assumes the middle innings, not the ninth. And every market number on this page was pulled from published odds articles this morning, not the app — prices in the app move, and a lean that was +4.2 percent at +127 is something less at +115. If the number moved, the edge moved.

Desk notes

• Phillies–Braves went untracked for the second consecutive day (noon MT start precedes the desk's pricing tick). Logged as a coverage miss. Philadelphia was a −114 road favorite in published listings (PickDawgz).
• All prices are from TheSpread's Game 2 pages plus alternates (BetMGM, VegasInsider, PickDawgz). PRO•LINE remains unscrapeable for the postseason; re-check before betting.
• Yesterday's card graded 2-0 on the leans and right on the pass; the grading piece is here.

Prices verified against published odds pages as of late morning MT, September 30. Re-check in the app before anything goes on the card.

Bring your own pencil.