Kansas City Chiefs 24, Miami Dolphins 10

Opening and closing line

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

Spread

Chiefs −7 → Chiefs −10

Total

44.5 → 45.5

Moneyline

KC −340 / MIA +270 → KC −600 / MIA +440

The spread moved toward the side that won. The total rose 1 to 45.5; the game landed on 34.

How it happened

What this recap does not know

• —The ledger has graded this game’s position: 2026_03_KC_MIA_moneyline graded LOSS (minus 18c) at 2026-09-28 00:55 UTC (20:55 ET Sunday), via the canonical settle pass (gridiron_edge/scripts/capture_edge_clv.py --season 2026 --week 3 --mode settle --tables game), run by Cowork once Kalshi had finalized the contracts — the same code the Tue/Wed launchd schedule runs. That grade matches the outcome stated above; the P/L figure is the ledger’s number, not this page’s. The prop rows on this page are not yet graded by the ledger — the Tue/Wed launchd pass owns them (--tables prop was not run) — so every prop grade above is derived from the ESPN box score by script (handoffs/artifacts/NFL_W3_SUNDAY_2026/prop-reconciliation-w03-sunday.csv), never by hand; entry_price on a NO or UNDER row is the YES price, so that side’s cost is 100 minus it, and no P/L is summed on this page. If the ledger ever disagrees with anything above, the ledger is right and this page is wrong.
• —The Miami moneyline row was created 2026-09-16, and Saturday’s article said the game page withheld its moneyline call while the model sat at REVIEW. Whether the MODERATE ledger row rendered anywhere a reader could see it before kickoff was not seen directly. It is a ledger loss regardless.
• —Johnson’s zero receptions is an absence from ESPN’s receiving table, not a printed zero; he appears in the rushing and kick-return tables, so he played. The strike is graded a loss on that basis and is exposed to an official-gamebook correction like any box-score grade.
• —Achane’s injury had no confirmed severity when this was published beyond Hafley’s remark. Gordon left with cramps late in the third and returned.
• —The Willis fumble is placed by ESPN’s drive chart as ending a Miami possession at the Kansas City 31; the AP does not give a spot. The Rodriguez interception spot is the drive chart’s.
• —ESPN’s summary carried no weather block. Attendance (59,216) is ESPN’s figure and is not independently confirmed.
• —This recap does not know what the Kalshi book did during the game. Nothing above is a live price; the entry and close quoted are the ledger’s captures, the DraftKings numbers are ESPN’s archived open and close, and the prop grade is a resolution against the box score, not a settlement read off the exchange.

What this read does not know

• —Malik Willis’s absence from the injury report is not a medical statement. Last week he carried a toe; this week he is simply not listed. That is consistent with fully healthy and also with an issue no longer worth listing. No source found stated it either way.
• —Kansas City’s line is the open question on its side. In the latest official practice report Josh Simmons did not practise, Chris Jones and L’Jarius Sneed were limited, and Patrick Mahomes and Rashee Rice were full. Miami gets Chop Robinson back from the concussion protocol.
• —Our model’s disagreement with the market on this game is the widest on the slate, and we do not believe it. It is a power-ratings artefact rather than a read on the football, it is flagged internally as such, and it is the reason this game is kept off our headline boards. Nothing on this page should be taken as our number on this game.
• —Two games is not a rate. “Zero sacks” and “78.8 percent completions allowed” are both real and both drawn from two games against two specific opponents.
• —We publish no position on this game — the model rows for it sit in held stat families on a run with a systemic one-way lean. See the week’s handoff.