Athletics vs. Astros: Pitching Matchup, Injury Report and Storylines for Saturday's Game 3
Setting the Board
Game 3 of a four-game series from Sutter Health Park, and the board has already moved twice since it opened. Houston arrived as a -146 favorite with a total of 8.5. By Saturday night the Astros sit at -162 and the total has climbed a full run to 9.5. That's a rare full-point jump on the total at this stage of the season, and it tells you exactly what the market is doing: chasing offense in a small ballpark to close out the year.
Earl's model isn't buying it.
Our numbers project Houston 4.5, Oakland 4.4 — a 0.11-run game. That's a coin flip dressed up as a two-run favorite. The model's strongest play on this board isn't the side at all. It's Oakland +1.5 (EV 15.26), the biggest edge the model has generated on this slate, with a secondary lean to the Under 9.5 (EV 1.99). The straight moneyline pick is Houston, but at -162 it carries a negative expected value of -8.25. That tension is the whole story of this game.
Perkins vs. Wesneski: The ERA Is Lying to You
| Pitcher | Record | ERA | WHIP | xERA | K/9 | BB/9 |
|---|---|---|---|---|---|---|
| Jack Perkins (R) | 3-12 | 5.90 | 1.45 | 4.09 | 10.16 | 3.72 |
| Hayden Wesneski (R) | 5-2 | 3.50 | 1.15 | 3.81 | 6.83 | 3.17 |
Two pitchers whose ERAs sit 2.4 runs apart — and whose advanced metrics narrow that gap to almost nothing.
Jack Perkins' 5.90 ERA is the number on the back of the baseball card, but his expected ERA is 4.09. That 2.01-run gap is one of the widest on the slate, and it's backed by the contact data: hitters have slugged .462 against him against an expected slugging of .390, and his wOBA allowed (.351) sits 40 points above his expected mark (.311). He's surrendered 20 home runs in 116 innings, and 8.5 percent of his batted balls have been barrels — the kind of damage rate that tends to regress rather than repeat.
Hayden Wesneski's profile runs the other way. His 3.50 ERA flatters him slightly: expected ERA 3.81, expected wOBA .301 against a .281 actual. He's been good, not great, and 6.83 strikeouts per nine does not describe a pitcher who misses bats when he needs one.
Go a layer deeper into the arsenals and the real matchup emerges.
- Perkins' 4-seam fastball is a problem: .291 average, .591 slugging, .432 wOBA allowed, and a minus-12 run value across 607 pitches. His cutter (.441 wOBA) and sinker (.481 wOBA) are worse in small samples.
- His sweeper and changeup are the reason he's still in the rotation — .308 and .235 wOBA allowed, with whiff rates of 37.9 and 37.5 percent.
- Wesneski's sweeper is his weapon: .193 wOBA allowed, 35.5 percent strikeout rate. His changeup (.167 wOBA) plays too.
- But Wesneski's 4-seam is his vulnerability — .503 expected slugging and a 51 percent hard-hit rate allowed.
Translation: Perkins has to survive on two pitches and stay away from the fastball. Wesneski's fastball is the pitch to attack. The gap between these two is real, but it's a one-pitch gap, not a 2.4-run gap.
Run Support: The Hidden Number in Perkins' Record
Perkins is 3-12. That looks like a disaster until you check what his offense has done for him. In his last five starts, Oakland has scored 0, 1, 2, 7 and 3 runs — 13 runs across five games, or 2.6 per start. No pitcher wins 12 games when he's handed fewer than three runs of support.
That's the systemic issue here. Oakland averages 4.36 runs per game on the season, and the lineup is thinner than that number implies. The Athletics' injury report includes Nick Kurtz, Brent Rooker, Jacob Wilson (all 60-day), plus Max Muncy and Tyler Soderstrom (10-day) — alongside six pitchers on the 60-day list. This is not a lineup built to punish a mistake pitcher.
Houston's list is real too — Carlos Correa (60-day) plus a center-field logjam of Daulton Varsho, Jake Meyers and Brice Matthews — but the Astros have weathered it to 79-81, and they're 40-39 on the road against Oakland's 32-47 at home.
Why the Run Line Is the Real Play
Here's the math that matters. The model projects a 0.11-run Houston win. Houston must win by two to cash -1.5. Oakland cashes +1.5 if it wins outright or loses by one run.
With a projected total of 8.88 and a projected margin of 0.11, the outcome distribution is heavily concentrated around one-run games in a low-scoring environment. Oakland +1.5 doesn't require Perkins to out-pitch Wesneski. It requires the Athletics to lose by one or better — which is precisely what the model's most likely outcome is.
An EV of 15.26 on the run line against -8.25 on the moneyline is the clearest signal on the board. Same game, same side, radically different prices. When a model's projected margin is a tenth of a run, betting the favorite at -1.5 and -162 means paying premium prices for two separate things that aren't there.
The Total: Don't Chase the Move
The line has moved from 8.5 to 9.5, and the market's logic is easy to follow. Sutter Health Park seats just 14,014 — the smallest capacity in MLB — and it plays to a 102 park factor with a 105 home run factor. Add a 5.90 ERA starter and the over writes itself.
But that over is being priced on reputation, not form. Oakland averages 4.36 runs per game. Houston averages 4.45. Neither is an offensive juggernaut, both are missing bats, and the model's projected total of 8.88 sits more than half a run under the current number. The Under 9.5 at EV 1.99 is a modest edge, and it's the natural pairing with an Oakland +1.5 ticket — the same low-scoring, close-game script produces both.
Public Side vs. Sharp Side
The moneyline move from -146 to -162 tells you where the public money went. A division contender against a 64-96 team starting a 3-12 pitcher is the easiest bet on the board for a casual player, which is exactly why the price has gotten worse. You're paying 16 cents of extra juice for the privilege of backing the popular side in a game our model calls a near coin flip.
Our model still sides with Houston to win. The Astros' 4.45 runs per game against Perkins' fastball problems is a legitimate edge. But "who wins" and "who covers" are different bets, and the price on Houston to win outright has crossed into negative-value territory. There's also a human variable worth filing away: this is the closing stretch of Houston's season with a sentimental clubhouse storyline — Justin Verlander's final start — dominating the news cycle, and the Astros just blew a 4-0 lead in this series. End-of-season teams with nothing settled emotionally are unpredictable favorites.
The Verdict
Final card, straight from the model:
- Oakland +1.5 — top play, EV 15.26
- Under 9.5 — model total 8.88, EV 1.99
- Houston moneyline — model side, but negative EV (-8.25) at -162. Pass at this price; -146 was the number.
The edge here isn't picking the winner. It's recognizing that a 0.11-run projection and a -162 price are fundamentally incompatible. Take the run, take the under, and let the market keep paying for the popular side.