San Francisco 49ers 32, Seattle Seahawks 29

Opening and closing line

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

Spread

Seahawks −3.5 → Seahawks −3

Total

46.5 → 46.5

Moneyline

SF +164 / SEA −198 → SF +140 / SEA −166

The spread moved toward the side that won.

How it happened

What this recap does not know

• —The ledger has not graded this game. There is no live nfl_weekly_edges row for it (the Seahawks moneyline was voided 2026-10-07 and counts for nothing), and the nfl_prop_edges rows on the final pre-kickoff board carried no result when this was written — grading runs Tue/Wed on the Mac-mini launchd schedule because Kalshi IP-blocks the GitHub runner. Every outcome above is derived from the ESPN box score and play-by-play; the WATCH grades come from a script run against the box score (handoffs/artifacts/NFL_W5_SUNDAY_RECAPS_2026-10-12/prop-reconciliation-w05-sunday.csv), never by hand. If the ledger ever disagrees, the ledger is right and this page is wrong.
• —The distance of Pineiro’s go-ahead field goal is left out: ESPN’s scoring summary has 44 yards and the AP recap has 43.
• —Tory Horton does not appear in the ESPN box score. Whether he was active and not targeted, or inactive, was not confirmed, so his WATCH row is left ungraded rather than called a miss.
• —Drake Thomas (listed questionable, ankle) was hurt in the second quarter and returned, and Cowing was hurt on a third-quarter kickoff return and returned, per the ESPN play-by-play; their condition after the game was not known when this was published. The reason for Greenlaw’s disqualification is the AP’s account (a head-butt of Ernest Jones IV after a late-hit flag on Darnold in the second quarter). Per AP, Deebo Samuel injured his hamstring on the play he fumbled and did not return; the severity was not known when this was published.
• —The DraftKings spread moved half a point toward San Francisco and the total did not move. Nothing is claimed about why.

What this read does not know

• —Nick Bosa’s availability is the one fact that would change this read and we do not have it. He did not practise Wednesday with a knee and calf issue and carries no designation. The entire Darnold question — 16.8 rating or 138.0 — routes through whether San Francisco can generate pressure, and their rate without Bosa over the past two weeks is 27th in the league. Friday’s designation is not a detail here; it is the page.
• —Several San Francisco non-participants are rest, not injury, and we have kept them separate. Christian McCaffrey, Trent Williams and Dre Greenlaw were all listed as not injury related on Wednesday, which for veterans in week five is ordinary maintenance and not news. Mike Evans (ribs) limited, Upton Stout (shoulder) out and Kyle Juszczyk (knee) out are the genuine ones. Conflating the two categories is how a Wednesday report turns into a false injury story, so we have not.
• —We agree with the market on this game and we are not manufacturing a disagreement. Our model line and the market line are inside about a point and a half of each other and the lean tier is NONE. In a week whose six largest leans all sit on underdogs and look like a shrinkage artefact, agreement is the more credible state. There is no position because there is nothing to take.
• —Nobody has a game-status designation yet, on either team. Today is Thursday; designations land Friday. That applies to Bosa, Charbonnet, Smith-Njigba and Leonard Williams alike.
• —The Purdy-versus-Macdonald history is five games and it may be measuring a roster rather than a matchup. Those five meetings span different San Francisco receiving rooms and different Seattle personnel. It is the most specific fact on this page and also the one most likely to be a coincidence of circumstance. We cite it as history, not as a forecast, and note that Purdy’s current form is the best of his career by some distance.
• —Neither run-defense side is graded here. The source puts the running-back section behind a paywall, and the only run-defense data points available were two single-game rushing figures for opposing quarterbacks. In a game where San Francisco is fourth in the league in rush rate, that is a significant gap in the read and not one we can close.
• —Seattle’s third-down rate and San Francisco’s pace both push toward a low-possession game, and low-possession games are the ones our margin distribution prices worst. Fewer drives means more of the outcome sits in the tails, and the key-number problem logged in the standing gate handoff §6.2 — the model carrying roughly 3% of its mass on a margin of exactly three where reality carries about 15% — bites hardest on exactly this kind of game.