mlbAnalysis

LAD ML @ SF: Dodgers vs. Giants Pick for Sept. 27, 2026

The model gives the Dodgers 80% at Oracle Park behind Jack Dreyer. The reported 44% market figure conflicts with the -235 listing, and we explain why that matters.

The regular season ends Sunday at Oracle Park. The Los Angeles Dodgers (99-62) face the San Francisco Giants (65-96). Los Angeles has already clinched the National League West, the NL's No. 2 seed and a first-round bye. San Francisco has lost four straight. Our engine lists Jack Dreyer for the Dodgers and Carson Seymour for the Giants. One ticket listing still showed both starters as TBA, and reporting indicates the Dodgers plan a bullpen game. Those two accounts can both be true if Dreyer opens and relievers follow him. Either way, the pick is LAD moneyline.

The number

- Model probability (LAD win): 80% - Market-implied probability, as supplied to our engine: 44% - Edge, as reported by our engine: 83.2 - Model confidence score: 79.3

The 44% market-implied figure conflicts with the price we found. The Dodgers are listed at -235 and the Giants at +195.

- A -235 price implies roughly 70% for Los Angeles before removing the bookmaker's margin. - After removing the margin, it implies roughly 67–68%.

We cannot reconcile the 44% input or the 83.2 edge figure with that listing. We report both exactly as our engine produced them, but we do not treat them as reliable.

Measured against the -235 price, the gap is about 10–13 percentage points (80% model vs. roughly 67–70% market). That is still a meaningful difference. It is also far smaller than a 36-point gap would suggest. Readers checking our public ledger should use the -235 comparison as the realistic edge.

Why the model leans this way

Projected result. The model projects Los Angeles 5, San Francisco 3, a margin of 2.27 runs.

Offense. This is the largest gap in the inputs. The Dodgers carry an offense rating of 61.5 against the Giants' 36.1. The weekend so far fits that difference:

- Friday (LAD 2-0): Teoscar Hernández and Andy Pages homered in the second inning. Tarik Skubal allowed three hits over seven innings and struck out 10. He is 5-2 with a 2.61 ERA since arriving from Detroit at the deadline. - Saturday (LAD 4-3): Mookie Betts homered, and Kiké Hernández led off the ninth with a tiebreaking home run.

Totals and park. The model total is 8.0 against a market total of 8.5, a lean to the Under. Oracle Park's park factor is 0.95, which suppresses scoring slightly. A lower-scoring game usually adds variance for a favorite. Here, the model still expects Los Angeles to produce more of the limited offense because of the ratings gap.

Form. The Dodgers have won two straight and have taken both games of this series. The Giants have lost four straight. The model reads recent form as a modest supporting factor, not the main driver.

Pitching structure. A bullpen game led by Dreyer means no single starter has to get deep into the game. The model's projection already accounts for that structure.

The case against

At an 80% model probability, this pick loses about one time in five. Against the market's roughly 67–70%, it loses about one time in three. Several factors could push the result toward the Giants.

- Nothing at stake for Los Angeles. The Dodgers' seed and bye are locked. Snell's recent groin issue shows how much the club cares about its postseason rotation. That gives the Dodgers reason to limit exposure for key players, shorten appearances and use depth arms late in a close game. In a bullpen game, a lineup or innings plan that favors rest can change the matchup in ways a pre-game projection may not fully capture. - Late-inning margins. The first two games were decided by 2 runs and 1 run. Saturday's game turned on a leadoff home run in the ninth. Close games are close to coin flips at the finish, and a late swing could just as easily go to San Francisco. - Seymour and home-field variance. A one-time start from Carson Seymour at home can outperform a season-long rating. Oracle Park can mute the power that won the first two games. - Uncertain inputs. The TBA listing and the unreconciled market figure mean some pre-game information is incomplete. If the Dodgers' final lineup is heavily rested, the real edge shrinks. - Ordinary variance. A 65-96 team still wins about 40% of its games. Single MLB games are noisy.

The model's lean holds, but the realistic edge is roughly 10–13 points against -235. Treat it with that level of confidence.

21+ only. Set limits before you play, and never stake more than you can afford to lose. If it stops being fun, call 1-800-GAMBLER.

See every pick on the record

Model-driven picks, published before lock and settled on a public, hash-chained ledger. The record is free to check.

Informational only, not financial advice. 21+. If you or someone you know has a gambling problem, call 1-800-GAMBLER.