Minnesota Twins at San Francisco Giants
MLB · Mon Sep 21, 9:45 pm ET · 6 legs priced, 1 recommended.
The drop closed at kickoff, Mon Sep 21, 9:45 pm ET. Every leg below is settled.
Our model against the market
Ours is the model’s probability. Implied is the probability the posted price carries. Edge is the gap between them, in points.
| Market | Pick | Ours | Implied | Edge | Tier |
|---|---|---|---|---|---|
| Spread | SF +1.5 | 68.8% | 59.3% | +9.5 | A |
| Moneyline | SF to win | 48.4% | 46.9% | +1.6 | C |
| Moneyline | MIN to win | 51.6% | 53.1% | -1.6 | C |
| Total | Over 8 | 47.5% | 50.8% | -3.2 | C |
| Spread | MIN -1.5 | 31.2% | 40.7% | -9.5 | C |
| Total | Under 8 | 38.7% | 49.2% | -10.5 | C |
Swipe the table sideways for the remaining columns.
How it graded
| Pick | Market | Tier | Result | CLV |
|---|---|---|---|---|
| SF +1.5 | Spread | A | Win | — |
| SF to win | Moneyline | C | Win | +3.13 |
| MIN to win | Moneyline | C | Loss | -3.12 |
| Over 8 | Total | C | Loss | -0.76 |
| MIN -1.5 | Spread | C | Loss | — |
| Under 8 | Total | C | Win | +0.76 |
Swipe the table sideways for the remaining columns.
Closing-line value is the no-vig probability of the side we took at the close minus the same probability when the drop was built. Positive means the market moved onto our side. Two things keep it honest here. A leg whose drop was built after the game started has no closing line to beat, so its CLV is a dash rather than a zero. And the average is taken over recommended legs only: both sides of every market are priced and graded, and averaging a side against its own mirror image cancels to zero whatever the model did.
Injury report the model saw
MIN
- Mike Paredes — 60-Day-IL
- Mick Abel — 60-Day-IL
- Kaelen Culpepper — 10-Day-IL
- Trevor Larnach — 10-Day-IL
- Byron Buxton — 10-Day-IL
SF
- JT Brubaker — 15-Day-IL
- Matt Gage — 60-Day-IL
- Willy Adames — 10-Day-IL
- Nate Furman — 10-Day-IL
- Bryce Eldridge — 7-Day IL
Copied from the drop file as it stood when the drop was built. It is not updated afterwards.
How to read this page
Tier A is an edge of six points or more and tier B three to six; both are recommended. Tier C is below three points — it is shown for completeness and is never recommended. A prop market whose calibration gate is failing is marked watch: the model prices it, the record grades it, and nothing here recommends it.
Every leg on this page is graded in public on the record and on the Sep 21, 2026 results page. One leg a day is posted in full, free, on today’s free leg.