Prices as of Wednesday, September 2, 2026, 8 a.m. ET, from Kalshi's live order books and the OddsShopper live odds screen.
At 6:32 p.m. ET on Tuesday, September 1, somebody on Kalshi bought about 9,400 contracts of "Seattle wins by more than 3.5 points" in a single order, clearing the exchange's price from 49 cents to 51. Three minutes later a second order of roughly the same size cleared it again, from 51 up to a print of 53. By 8 p.m., 20,106 contracts had traded in 47 fills across two hours, buyers took the "yes" side in 38 of them, and the price settled at 50 to 51 cents. The contract had closed every one of the previous eight days between 47 and 49 cents.
That is the biggest move any price on the NFL Kickoff Game made this week, measured by the money behind it, and it happened without a headline. The Patriots and Seahawks open the 2026 NFL season on Wednesday, September 9, at Lumen Field, a rematch of Super Bowl LX, which Seattle won 29-13 in February. The spread on that game has been a fight over one number, 3.5, since the schedule came out in May. On Tuesday night the exchange picked a side of it. The sportsbooks, including the sharpest ones, did not follow, and the contract that pays if Seattle simply wins never budged from 63 cents. The gap that opened is about a point and a half of implied probability, real money for anyone shopping the number, but not a full point of spread. Below: what actually traded, what the books did with it, why the number keeps trying to be 4 and cannot get there, what happened the last time this market saw a burst like this, and where six models that were never shown a price put the line.
A note before the numbers. On Kalshi a price in cents is the crowd's probability: a 51-cent contract is the market saying about a 51% chance, and it pays $1 if the thing happens and nothing if it does not. Prices on this page are the live bid (what buyers will pay) and ask (what sellers want) pulled early on September 2, 2026, and the trade-by-trade history is the exchange's own tape. Treat them as one moment's estimate, not a prediction of fact and not financial advice. Kalshi is a CFTC-regulated event-contract exchange, open to users 18+ in eligible states; availability varies by state.
Free: The Weekly PM Market Brief
The 8-model panel's graded record, the week's biggest market-vs-model gaps, and what's spiking next. One email, Sundays.
Unsubscribe anytime, one click. We never share your email.
The odds screen behind every book quote on this page is free for a week with OddsShopper Pro, and code FIELDGOAL20 takes 20% off your first month after that. The button is at the bottom.
What Actually Traded On Tuesday Night
Kalshi's spread market for this game is not one line. It is a ladder of separate yes-or-no contracts, one per margin, and the rule on the rung that moved is exact: "If Seattle wins by more than 3.5 points in the New England vs Seattle Pro Football game originally scheduled for Sep 9, 2026, then the market resolves to Yes." Seattle by 4 pays. Seattle by 3 does not. Holding "no" on that contract is the same position as taking the Patriots +3.5 at a sportsbook, which is why this rung, and not the winner market, is where the game's real argument lives. (If the mechanics are new to you, how to bet NFL on Kalshi covers the ladder, and our Week 1 board prices all 16 games.)
This rung is also, by a wide margin, the busiest spread contract in the game. About 108,000 contracts have traded on it since it listed on August 3, with roughly 74,500 still open, and 24,500 of those trades came in the 24 hours ending Wednesday morning. Roughly half of everything that traded on the whole spread ladder this week traded on this one rung.
Here are the two hours after 6 p.m., from the exchange's trade history, in Eastern time:
| Time (ET), Tuesday Sept. 1 | What happened | Price of "Seattle by more than 3.5" |
|---|---|---|
| 6:15 P.m. | First trade of the burst | 48 cents |
| 6:32 P.m. | One buyer sweeps about 9,400 contracts through the book | 49 to 51 cents |
| 6:35 P.m. | A second sweep of roughly 9,400 contracts | 51 cents to a high print of 53 |
| 7:06 P.m. | Largest sell of the evening, about 970 contracts | 50 cents |
| 8 P.m. Onward | Trading returns to two- and three-digit fills | 50 cents bid, 51 ask |
The rows I keep coming back to are the second and third. Two sweeps three minutes apart, each about 9,400 contracts, together nearly 18,800 of the evening's 20,106, is not a crowd reacting to news. That is one or two accounts deciding the price was wrong and paying up to fix it. I looked for anything stamped near 6:32 p.m. Tuesday and did not find it: no injury news, no transaction, and no practice report, because the first official ones of the week had not been issued yet. What the tape shows is size, not a catalyst.
The Winner Price Did Not Move, And That Is The Tell
Here is the part that makes Tuesday interesting rather than just large. Kalshi's plain "Seattle wins" contract, the most-traded market in the game with about 577,000 contracts across both sides, sat at 62 cents bid and 63 ask on Tuesday afternoon, sat there through the burst, and sits there Wednesday morning. Its own 24-hour tape ran past 31,000 contracts on the Seattle side without changing the price by a penny.
A spread and a moneyline are two views of the same distribution of final margins, and they are supposed to move together. If the crowd had learned something that made Seattle more likely to win by 4, it should also have made Seattle a little more likely to win at all. It did not. The only thing that changed on Tuesday was the probability mass the market assigns to Seattle winning by exactly 1, 2 or 3 points. On Kalshi that gap is the difference between the 63-cent winner contract and the 51-cent "by more than 3.5" contract: 12 cents, or 12% of outcomes, riding on margins of exactly 1, 2 or 3, with three points doing most of that work. Three is the single most frequent winning margin in the NFL, which is why a half point on either side of it is the most expensive half point in the sport. Tuesday's buyers paid to be on the far side of it.
The Week On The NFL Opener's Board
I said Tuesday's move was the biggest of the week by the money behind it, and I want to show my work, because the honest ranking depends on what you measure. Over the last week, from the close on Tuesday, August 25, to Wednesday morning, five thinner rungs moved more in raw cents than the 3.5 rung's two: "Seattle by more than 16.5" climbed from 12 to about 17.5 cents on 149 contracts all week, "by more than 9.5" went from 24 to 28.5 on 689, "by more than 13.5" went from 16 to 20.5 on about 6,700, and the two nearest New England rungs, "by more than 1.5" and "by more than 2.5," each rose about 3.5 cents, one on 1,532 contracts and the other on 199. Those are real prices, and the 13.5 rung is not thin, but none of them had a day like the 3.5 rung's Tuesday, and none of them moves the line everybody actually bets.
| Contract | Kalshi price to buy, Sept. 2 | Aug. 25 close | Move this week | Contracts traded this week |
|---|---|---|---|---|
| Seattle Wins | 63¢ | 63¢ | flat | about 198,000 both sides |
| Seattle By More Than 3.5 | 51¢ | 49¢ | up 2 cents (up 3 from Monday's close) | about 47,500 |
| Seattle By More Than 2.5 | 59¢ | 57¢ | up 2 cents | about 3,500 |
| Seattle By More Than 6.5 | 39¢ | 38¢ | up 2 cents | about 9,800 |
| Seattle By More Than 13.5 | 21¢ | 19¢ | up 4.5 cents | about 6,700 |
| Over 44.5 Total Points | 48¢ | 48¢ | flat at the ask, up half a cent at the midpoint | about 5,900 |
| New England Wins | 39¢ | 39¢ | flat | (counted above) |
| New England By More Than 1.5 | 34¢ | 34¢ | up 3.5 cents, and 1,492 of its 1,532 weekly contracts traded in one fill Tuesday afternoon | 1,532 |
Both price columns are Kalshi asks, the price to buy "yes," as of September 2, 2026, 8 a.m. ET and at the August 25 close; the move column is measured midpoint to midpoint, which is why an ask can sit still while the midpoint moves half a cent.
The last row is the one that tells you the argument is not over. The contract that pays if New England wins by 2 or more, the nearest thing on the Patriots side of the ladder to the same question, had traded 32 contracts all week until 2:29 p.m. Tuesday, when one fill of 1,492 came in at 33 cents. That was four hours before the Seattle sweep. Somebody bought the Patriots side of the margin in the afternoon, and somebody else paid up for the Seattle side that evening. The board is not a consensus. What it shows is two views of the margin, priced four hours apart.
The Last Burst Reverted
This game has done this once before. On Sunday, August 23, the winner market traded 80,655 contracts in a day, more than it had traded in the previous two weeks combined, and the Seattle price printed as low as 56 cents before closing the day back at 63. The 3.5 rung printed as low as 40 cents that day on 4,020 contracts and closed at 47. Open interest in the winner market jumped by about 27,000 contracts, so the money stayed, but the price did not. That was the weekend Vrabel called his own roster not very deep, which if anything argues the other way, so I do not read it as a news day either. The lesson is narrower than "bursts always revert": on this particular game, size has arrived twice without a news hook, and the first time the price went back to where it started within hours. Tuesday's move has held for about 14 hours so far. That is the fact to re-check before you treat 51 cents as the new number.
Where The Sportsbooks Sit On 3.5
Now the other half of the title. The live odds screen carries this game at more than 30 sportsbooks and exchanges, and on Wednesday morning all but two of them post Seattle -3.5; Bovada is at 4 and PointsBet at 4.5. DraftKings was at 3.5 on Monday, per Dave Loughran's writeup, and 3.5 is where all but those two books sit on Wednesday morning. What differs is the juice on each side of 3.5, and juice is how a sportsbook says which side it would rather not take. Strip the book's margin out of each pair of prices and you get a fair probability that Seattle covers, the same unit as the Kalshi cent price, which makes the whole board comparable in one column.
| Venue | Seattle -3.5 | New England +3.5 | Fair chance Seattle covers, vig removed |
|---|---|---|---|
| Kalshi (Bid 50 / Ask 51) | n/a | n/a | 50.5% |
| FanDuel | -115 | -111 | 50.4% |
| BetMGM | -110 | -108 | 50.2% |
| Caesars | -110 | -109 | 50.1% |
| DraftKings | -110 | -110 | 50.0% |
| Circa | -110 | -110 | 50.0% |
| NoVig | +100 | -102 | 49.8% |
| BetRivers | -109 | -112 | 49.7% |
| Pinnacle | -101 | -111 | 48.9% |
| Bet105 | -101 | -111 | 48.9% |
| WynnBET | -101 | -116 | 48.3% |
Sportsbook prices from the OddsShopper live odds screen, September 2, 2026, 8 a.m. ET; 33 venues posted 3.5, and the de-vigged fair number across the sportsbooks and other exchanges, Kalshi excluded, ran from 47.8% to 50.4% with a median of 49.8%. Kalshi's figure is the midpoint of its bid and ask, the gap that is the exchange's version of the vig.
Read the column top to bottom and the split is plain. The recreational books that see the most public money on a Super Bowl rematch, FanDuel, BetMGM and Caesars, shade a hair toward Seattle covering. DraftKings and Circa sit dead even at -110 both ways. And the books that sharp bettors price against, Pinnacle, Bet105 and the low-hold exchange NoVig, shade the other way: at Pinnacle you pay -111 to take the Patriots +3.5 and only -101 to lay Seattle, which is the book telling you it thinks the true number is a little under 3.5, closer to 3 than to 4. Kalshi, at 50.5%, now sits above every sportsbook on the board. Tuesday did not just move the exchange. It put Kalshi at the top of the distribution, past the public books, while the sharp books lean toward 3.
A Worked Example: What 51 Cents Buys
In payout terms, the two sides of Tuesday's argument are close to even money with a small premium on the champions. Buying "yes" on Seattle by more than 3.5 at the 51-cent ask turns $100 into about $196 if Seattle wins by 4 or more, and into nothing if it wins by 3, wins by less, or loses. Buying "no" at 50 cents, since the "no" price is a dollar minus the "yes" bid, is the position that matches the Patriots +3.5, and it turns $100 into about $200 if Seattle fails to cover. At Pinnacle the same Patriots side costs -111, so $100 returns about $190, while Seattle -3.5 at -101 returns about $199. The exchange is the cheaper venue for the Patriots side by about $10 per $100 on identical risk, before Kalshi's trading fee, which is largest on a contract priced near 50 cents and takes a few dollars of that back, and the dearer one for Seattle, which is exactly what "Kalshi leans toward 4, Pinnacle toward 3" means in money. The tape adds one more line to that ledger: the 1,492 contracts of "New England by 2 or more" bought Tuesday afternoon at 33 cents paid for a strictly narrower version of the same Patriots opinion, a contract that loses if New England wins by a single point. The contract settles on the official final score; Kalshi lists it to expire within hours of the final whistle on the night of September 9, so this is a one-week hold with no way out except the game.
On the moneyline the split is a little wider and runs the other way. DraftKings at -180 and +150 de-vigs to about 61.6% Seattle, Pinnacle at -181 and +158 to about 62.4%, Circa at -185 and +160 to about 62.8%, FanDuel at -198 and +166 to about 63.9%, and Kalshi's 62-to-63-cent quote to about 61.9%. The books sit within about two points of each other on how often Seattle wins, and the exchange is at the low end of that range on the winner while sitting at the top of the range on the cover. Whatever the exchange is saying, it is not saying Seattle wins more often. It is saying that when Seattle wins, it wins by more than a field goal more often than the books think.
Why The Number Wants To Be 4, And Why It Cannot Get There
The case for Seattle by 4 or more is the same case that won the Super Bowl, and it is intact. Seattle's defense produced 56 sacks last season, sacked Drake Maye six times in February, and, per ESPN's Bill Barnwell, pressured him 28 times that night with 14 of those pressures charged to left tackle Will Campbell, the most by any player in any game all season. Maye was sacked 68 times across 21 games in 2025. Seattle kept the front, re-signed Leonard Williams for three years and $90 million, and paid Devon Witherspoon $132 million over four years on August 15. Maye, meanwhile, will play his first game since the Super Bowl with zero preseason snaps; per ESPN, Vrabel estimated about 84 full-team reps for him across three joint practices. And it is Vrabel himself who said on August 23, per Boston.com, "I do not think we have a very deep team." A loud home building against a quarterback whose protection failed in this exact matchup is exactly the shape of game that produces a two-score margin.
The case against 4 is why DraftKings has not posted one since the schedule came out on May 14, and Dave Loughran laid it out in full on Monday. The short version: Seattle is not the roster that won. Kenneth Walker III, the Super Bowl MVP, left for Kansas City in March; Klint Kubiak took the Raiders' head-coaching job, so Brian Fleury calls plays for the first time on September 9; Zach Charbonnet is on reserve/PUP, leaving first-round rookie Jadarian Price to carry the backfield. The secondary that held New England to 48 first-half passing yards in February lost Coby Bryant, Riq Woolen and rookie safety Bud Clark, cut Trevon Diggs and Noah Igbinoghene, and cannot use Terrion Arnold, signed on August 29 and parked on the commissioner's exempt list. Against that group New England lines up A.J. Brown, acquired from Philadelphia on June 1 for a 2028 first-round pick and a 2027 fifth, and Romeo Doubs, while Christian Gonzalez, the Patriots' best corner, returned to practice on August 24 after an August 10 injury, per ESPN, still without the extension the two sides are negotiating.
What the ladder adds to that list is where each fact lands. The Seattle pass rush against Maye's protection is a case for the far rungs, "by more than 6.5" at 39 cents and "by more than 9.5" at 30, because pressure produces turnovers and turnovers produce two-score margins. The gutted cornerback room against Brown is a case for the total and for New England keeping it inside a score, which is the 3.5 rung's "no" side. Neither fact says much about Seattle winning by exactly 4, 5 or 6 rather than exactly 1, 2 or 3, and that band is the only thing Tuesday's buyers paid for. The 12 cents between "wins" and "wins by 4" is the one number on this page nobody on the sportsbook board has tested in public, and until someone does, the honest answer to whether Tuesday's buyers overpaid is that they bought the far side of the most crowded margin in football at a price the sharpest books would not offer them.
The whole tension in one sentence: Seattle's advantage is a pass rush against a bad protection unit, and Seattle's weakness is a rebuilt secondary against the best receiver New England has had in a decade. A pass rush wins by 7. A shootout is decided by 3. The market has spent the summer unable to decide which game it is watching, and the number has sat at 3.5 because 3.5 is the price of not deciding.
Dave Loughran wrote it up on Monday for this site and took the Patriots +3.5, specifically "no" on this exact Kalshi contract at 52 cents, on the grounds that a favorite that cannot become a 4 in three and a half months of public betting is telling you something. His line for getting out was the number moving to 3. It moved the other way instead: the "no" side he priced at 52 cents on Monday cost 50 on Wednesday morning. Whatever you make of his read, it is worth saying plainly that nothing on Tuesday's tape was information. It was size.
Six Models, Shown No Prices: Where They Put The Number
We asked six AI models to price this game on Wednesday morning without showing them a single market price, only a card of the facts above: rosters, injuries, the Super Bowl pressure numbers, Vrabel's depth comment, the play-caller change. Each had to price the winner contract and the 3.5-point contract from one consistent view of the final margin. Two of the eight seats we normally run, ChatGPT and Gemini, were not available for this run, so this is a six-seat read and should be weighed as one.
| Seat | Seattle wins | Seattle by more than 3.5 |
|---|---|---|
| Claude Fable | 59.5% | 45.5% |
| Claude Opus | 66.0% | 54.0% |
| Claude Sonnet | 63.0% | 49.0% |
| GLM 5.2 | 64.0% | 54.0% |
| Kimi K3 | 62.5% | 49.5% |
| DeepSeek V4 | 68.5% | 44.0% |
| Six-Seat Blend (Simple Average) | 63.9% | 49.3% |
| Kalshi, Midpoint Of Bid And Ask, Sept. 2 | 62.5¢ | 50.5¢ |
Generated September 2, 2026, with no market price in the prompt. Model estimates, not predictions of fact and not financial advice.
On the winner the six seats land within a point and a half of the exchange, the same place our August 16 read put it against a 64-cent market, and well under the eight seats that scored the full Week 1 slate in a read last updated August 24, all of which had the Patriots between 42% and 45% against a 37-cent price; this run's blend has them at about 36%. On the 3.5 rung the blend is a coin flip, one point under Tuesday's buyers and one point over Pinnacle, and the blend hides a real split. Opus and GLM see Seattle by about 5 to 6 and put the cover at 54%; Fable and DeepSeek see a margin centered closer to 3 and put it in the mid-40s. Every seat named the same two facts as the drivers, Seattle's pass rush against Maye's protection on one side and Seattle's gutted cornerback room against A.J. Brown on the other, and they simply weighted them differently. Fable's note is the one that explains the whole page: the gap between "wins" and "wins by 4" is wide precisely because so much of Seattle's winning margin lands on exactly 3. DeepSeek's pair is the odd one out. A 68.5% chance to win against a 44% chance to cover puts about a quarter of all outcomes on Seattle winning by exactly 1, 2 or 3, roughly double what any other seat or the market assigns to those three margins, and it is the number pulling the blend down. Without that seat the other five average 50.4% on the cover, which is the exchange's own midpoint.
The models, in other words, reproduced the sportsbook board and the exchange's midpoint. They did not find a reason the number should be 4.
More live boards from the same panel: New England vs Seattle, who wins the opener (Seattle wins, 63¢ to buy) · Kalshi NFL picks, every Week 1 game seat by seat (New England wins, 39¢) · the Kalshi NFL Week 1 board (all 16 games priced). Prices as of September 2, 2026, 8:40 a.m. ET.
- Patriots vs Seahawks Picks: The Champs Aren't The Same Team
- 49ers vs Rams Picks: The Home Team Flies In Last
- College Football Week 2 Predictions: The Machines Won't Take Texas
- 2026 Governor Races: Every Kalshi Board We Track
- 2026 Senate Races: Every Kalshi Board We Track
The Bottom Line
The Seahawks vs Patriots spread did not move because the market learned something. It moved because a few large orders on Kalshi decided, on Tuesday evening, that 3.5 was a point too cheap for the champions, and nobody who prices this game for a living agreed with them by Wednesday morning: the winner market stayed at 63, DraftKings stayed at -110 both ways, Pinnacle kept its lean toward the Patriots, and six models with no price in front of them called the cover a coin flip. The other side of the same question had been bought four hours before the sweep, at 33 cents, by someone who never got the headline either.
What settles the argument is not the tape but the injury reports. The first official ones before kickoff will name the player who moves this number, Christian Gonzalez, whose full participation is the difference between a Patriots secondary with a No. 1 corner and one without, and they will show how thin Seattle's own cornerback room is once Arnold and Clark are accounted for. A Gonzalez extension before kickoff cuts the same direction as his health: it is the headline most likely to pin this number at 3.5 or take it to 3, not to 4. Until one of those arrives, the sharpest read of this game is still the one the board has given all summer: Seattle wins about 62% of the time, Seattle wins by 4 or more about half the time, and the half point in between is the most expensive one in football.
If you price this game anywhere, the live odds screen shows every book's number on this spread side by side, so you can see which sportsbook is shaded toward the side you want before you pay the juice. Our analysts post free expert picks every day, the opener included, and the full toolkit behind them, the odds screen, the de-vig math and the positive expected value board, starts with a free week of OddsShopper Pro. Code FIELDGOAL20 takes 20% off your first month if you stay past the week.
Every Kalshi price above was fetched from the exchange on September 2, 2026, and the trade-by-trade history is Kalshi's own tape, matched to the exchange's 24-hour volume figure to within 20 contracts. Sportsbook prices are from the OddsShopper live odds screen at the same hour. Roster and injury facts are cited to ESPN, Boston.com and the teams as linked or named. Model figures are estimates, not predictions of fact and not financial advice. Kalshi is a CFTC-regulated event-contract exchange; 18+, eligible states only. If you or someone you know has a gambling problem, call 1-800-GAMBLER.
