Miami Dolphins 10, Minnesota Vikings 15

Opening and closing line

DraftKings via ESPN, one book. Not the Kalshi price — that block reads our own tape.

Spread

Vikings −7.5 → Vikings −10

Total

43.5 → 38.5

To win

MIA +310 / MIN −395 → MIA +400 / MIN −535

The spread moved toward the side that won. The total fell 5 to 38.5; the game landed on 25.

How it happened

What this recap does not know

• —This recap was written on 2026-10-09, five days after the game. The ledger graded the Dolphins game winner (pnl_cents −30) and both Bell PLAY rows (wins) on 2026-10-06. The WATCH tallies come from a script run against ESPN’s box score (handoffs/artifacts/NFL_W4_LATE_AND_W3_SNF_RECAPS_2026-10-09/props-w04-late.csv), not from the ledger. If the ledger ever disagrees with this page, the ledger is right.
• —The model’s lean is graded on its own schedule, against the line in points, and that grade is not shown here. The sentence above says only that Minnesota won by five against a model margin of 5.1.
• —entry_price on an UNDER row is the YES price, so that side cost 100 minus it. No P/L is summed on this page.
• —This recap does not know what the Kalshi book did during the game. The entry and close quoted are the ledger’s captures.

What this read does not know

• —Our injury feed is practice participation, not game-status designations. The slate behind this page was run Friday at 16:51 ET with Jefferson marked out; every other status is a practice line, and anything either club files on Saturday is not in it.
• —Who takes Jefferson’s snaps is reported, not confirmed. Addison, Hockenson, Jennings and Felton are the names in Minnesota’s own reporting; Jennings had one catch through two games and Felton’s role is unverified. Our own slate over-corrected the other way — it handed Jefferson’s entire target share to Addison, which is why the Addison receptions call is held at WATCH rather than published.
• —Miami’s right tackle is a three-way question. One report has Austin Jackson benched for Caedan Wallace, Week 3 snap counts had Kadyn Proctor at the position, and the depth chart still lists Jackson. Against this blitz rate, that is the most important unresolved lineup spot on either side.
• —The Dolphins’ backfield split without De’Von Achane is unknown. Ollie Gordon II carried 17 times for 41 yards in Week 3; Jaylen Wright is back to full practice and listed first on the depth chart. Minnesota allows 3.5 yards a carry and has not allowed a rushing touchdown.
• —The power ratings behind the model line are this week’s board (as-of 2026-09-30, written by the Week 4 pipeline run). Minnesota’s offensive rating still leans on the games Carson Wentz played; Murray’s own sample is one quarter and one start.
• —Our model’s lean on this game is large and we do not headline it. The model has Minnesota well short of the market’s margin, and Miami is the lowest-rated team on our board. A big disagreement that depends on the worst-rated team’s rating is a ratings question before it is a read, so no game-line position is published here.
• —Should you take Malik Washington under 4.5 catches? We hold it at WATCH, and here is the fork. Our model projects 4.0 receptions and has him over 4.5 about 35% of the time; he has caught 3, 4 and 5 in his three games on 8, 6 and 10 targets, with the trend running up since Achane went down. Under wins if Flores’s pressure ends drives before Willis gets to his third read; over wins if Miami answers the blitz with quick throws to its slot, which is exactly what it did against Kansas City.