Best regular-season record is a different question from the championship, and Kalshi prices it on its own board, one contract for each of the 32 teams. Winning it is a 17-game race about a high floor and an easy path, not a January survival test. Two games of the season have been played, and the board has already flipped twice: once during a game, and once while most of the country slept. The Los Angeles Rams, the preseason favorite at 15 cents, lost 27-7 to San Francisco in Melbourne on Thursday night, fell to 4 cents bid, 7 asked by midnight, and were bought back to 5 bid, 8 asked by morning. The 49ers' contract went from 5 cents to 50 in a 35-minute stretch spanning the third and fourth quarters, the 50 on a trade of 3.51 contracts; at 5:02 a.m. ET a single seller hit every bid down to 5 cents, and at 5:39 a.m. the contract printed 4. Underneath that headline, the highest bid on the board at our 7:55 a.m. read belongs to two teams that have looked nothing alike this week, Seattle and Baltimore at 8 cents, and the deepest bid still belongs to the Ravens, 17,654 contracts at 7, and they have not kicked off yet.
This is a price refresh, not a re-score. The verdict from our price-blind panel of eight AI models, generated August 16 before a snap had been played, stands as written: Seattle first, the Rams trimmed, Jacksonville the board's biggest gap in August. What Week 1 did to each of those calls, and what the price history says about a 49-cent print that was a 4-cent print five hours later, is the story below.
New to event contracts? Start with our step-by-step guide to trading NFL markets on Kalshi.
What changed (Sep 14, 12:24 AM ET)Kalshi's lead price on this board moved 10¢ → 22¢ (+12¢) since our last read. The panel's verdict below was written before this move and stands as published until the refresh lands.
The Quick Answer
After the first two games of the 2026 NFL season, Kalshi's best-regular-season-record board has no clear favorite. At our 7:55 a.m. ET read on Friday, September 11, the Seattle Seahawks (8 bid, 10 ask) and the Baltimore Ravens (8 bid, 9 ask) share the highest bid, with Buffalo at 7 bid, 8 ask. The San Francisco 49ers, who printed 50 cents during their 27-7 win in Melbourne and 49 on a 262-contract sale 73 minutes after it ended, were sold down to 4 cents by 5:39 a.m. and show 6 bid, 12 ask on a book with 21 contracts on one side and 17 on the other. The Los Angeles Rams, 15 cents when the panel ran on August 16, fell to 4 bid, 7 ask after the loss and were bought back to 5 bid, 8 ask overnight. Our price-blind panel's August 16 verdict is unchanged for this refresh: Seattle 11.7%, the Rams 10.1%, Jacksonville 9.7% against a market that prices the Jaguars at 1 bid, 2 ask. Seattle won 13-10 and lost Sam Darnold on its opening possession. — The Model Verdict panel
- Highest bid on the board: Seattle, 8¢ bid / 10¢ ask, and Baltimore, 8¢ bid / 9¢ ask
- Deepest bid on the board: Baltimore, 17,654 contracts bid at 7¢
- Biggest round trip: San Francisco, 5¢ at kickoff, 50¢ in the fourth quarter, 4¢ at 5:39 a.m.; now 6¢ bid / 12¢ ask
- Biggest faller: Los Angeles Rams, 15¢ on August 16 to 5¢ bid / 8¢ ask, after a 27-7 loss
- Panel favorite (August 16, frozen): Seattle Seahawks, 11.7% blend
- Biggest gap (frozen August 16 blend vs. this morning's price): Jacksonville, 9.7% blend against a 1¢ bid / 2¢ ask
- Settles: on or about February 8, 2027, on regular-season records only
The full 32-team table is directly below, then a print-by-print account of what moved in Week 1 and overnight, the panel's frozen case for Seattle and Jacksonville, the tie rule that shapes every number on this board, and a plain-English guide to reading a book this thin.
Stay ahead of the markets.
Daily insights and expert picks on Kalshi, Polymarket, and what's moving markets.
Free forever. Unsubscribe anytime.
The Market at a Glance
| Market | Pro Football Best Regular Season Record: which team finishes the 2026-27 regular season with the best win-loss record |
| Venue | Kalshi |
| Ticker | KXRECORDNFLBEST-27 (32 separate yes/no contracts, one per NFL team) |
| Settles on | Regular-season records only. The contract terms explicitly exclude the playoffs |
| Tie rule | NFL tiebreakers are applied first; any team still tied is paid 1 divided by the number of tied teams |
| Highest bid | Seattle Seahawks, 8¢ bid / 10¢ ask, and Baltimore Ravens, 8¢ bid / 9¢ ask |
| Deepest bid | Baltimore Ravens, 17,654 contracts bid at 7¢ |
| Panel favorite | Seattle Seahawks, 11.7% blend (August 16 run, not re-scored) |
| Prices as of | Kalshi, read September 11, 2026, 7:55 a.m. ET; the 12:39 a.m. and 1:19 a.m. ET reads from earlier editions are named where the text compares them |
| Verdicts generated | August 16, 2026, price-blind; frozen for this refresh |
These are model estimates, not predictions of fact and not financial or betting advice. Kalshi event contracts trade on a CFTC-regulated exchange; you must be 18 or older and in an eligible state to participate. Prices on this page refresh as the season develops, and the "as of" date above carries the timestamp.
Hottest Prediction Markets Right Now
- 2028 Democratic presidential nominee · $181M traded
- FedEx St. Jude Championship Winner · $91M traded
- 2027 Pro Football Champion · $63M traded
- 2028 U.S. Presidential Election winner? · $60M traded
- 2028 Republican presidential nominee · $57M traded
- Pro Baseball Champion · $54M traded
Every market above links to our full AI model verdict; browse them all on the OddsShopper prediction markets hub, and see how every settled call actually scored on the full graded scoreboard.
What Actually Separates The Field
Every model worked from the same fetched picture: each team's final 2025 record, win percentage and point differential; each team's actual 2026 opponent list, scored by how good those opponents were last season; 20 seasons of best-record outcomes; and the contract's settlement rules. The panel saw no market prices of any kind. Not this board's prices, and deliberately not sportsbook win totals either, so that its read is its own rather than a translation of someone else's line. It also saw no Week 1 results, because none existed; everything in this section is the August 16 card.
Four facts did most of the work.
- Point differential beat win-loss record, everywhere. Denver went 14-3 at only plus-90, a profile built on close games. The Rams went 12-5 at plus-172. Seattle went 14-3 at plus-191, the best mark in the league. Every seat regressed 2025 records toward point differential, which is why the panel has the Rams roughly level with the 14-win teams and Denver below them.
- Last year's best record is weak evidence. Across the last 20 seasons, a team that held the NFL's best record repeated it the following year three times in 19 chances, about 16%. That single number is why the panel does not simply re-rank the 2025 standings, and why Seattle leads on its differential rather than on its 14-3.
- The schedule is worth a win. Houston's 2026 opponents went .474 last season, the seventh-softest slate in the league, and Jacksonville's went .490, the 11th. Chicago drew the hardest at .550, then Miami at .541, with Green Bay and Arizona tied at .538. Across the league that span is worth roughly a win, in a race usually decided by one.
- The winning number is about 14, and it is often shared. Over the last 20 seasons the best record averaged 13.8 wins, never below 12. It was shared by two or more teams in nine of those 20 seasons, including a five-way tie at 12-4 in 2014 and a three-way tie at 14-3 last season, when Seattle, New England and Denver all finished there.
That last fact is the one this board's prices look least willing to pay for, and it changes the arithmetic of the whole race.
The Board
The narrow view first, built to read on a phone. "Kalshi mid" is the midpoint of the live bid and ask in cents, with the bid and ask beside it, because after Week 1 several of these books are wide enough that a single price would mislead. "Last" is the most recent trade. "De-vig" rescales the mid so the 32 rows sum to 100: at this read the 32 mids add to 94 cents, because the bids are thin and the asks wide (the 32 asks add to 115, the 32 bids to 73), so the de-vig column nudges each mid up rather than down. "AI blend" is the eight-model average from the August 16 revision round, frozen; it has not been re-scored for this refresh. Team rows carry the 2025 record and point differential, then the 2025 win percentage of each team's 2026 opponents.
| Team | 2025 record, point diff | 2026 opponents' 2025 win% | Kalshi mid (bid / ask) | Last | De-vig | AI blend (Aug 16) |
|---|---|---|---|---|---|---|
| Seahawks | 14-3, +191 | .514 | 9¢ (8 / 10) | 10¢ | 9.6% | 11.7% |
| 49ers | 12-5, +66 | .496 | 9¢ (6 / 12) | 4¢ | 9.6% | 4.4% |
| Ravens | 8-9, +26 | .479 | 8.5¢ (8 / 9) | 9¢ | 9.0% | 3.1% |
| Bills | 12-5, +116 | .528 | 7.5¢ (7 / 8) | 8¢ | 8.0% | 6.2% |
| Lions | 9-8, +68 | .467 | 6.5¢ (6 / 7) | 7¢ | 6.9% | 6.2% |
| Rams | 12-5, +172 | .516 | 6.5¢ (5 / 8) | 9¢ | 6.9% | 10.1% |
| Chargers | 11-6, +28 | .522 | 4.5¢ (4 / 5) | 5¢ | 4.8% | 1.8% |
| Bengals | 6-11, -78 | .450 | 4.5¢ (4 / 5) | 3¢ | 4.8% | 2.3% |
| Eagles | 11-6, +54 | .481 | 4.5¢ (4 / 5) | 4¢ | 4.8% | 5.2% |
| Texans | 12-5, +109 | .474 | 3.5¢ (3 / 4) | 5¢ | 3.7% | 8.8% |
| Broncos | 14-3, +90 | .512 | 3.5¢ (3 / 4) | 3¢ | 3.7% | 4.5% |
| Chiefs | 6-11, +34 | .536 | 3.5¢ (3 / 4) | 3¢ | 3.7% | 3.1% |
| Packers | 9-7-1, +31 | .538 | 3.5¢ (2 / 5) | 2¢ | 3.7% | 1.9% |
| Patriots | 14-3, +170 | .531 | 2.5¢ (2 / 3) | 2¢ | 2.7% | 7.8% |
| Vikings | 9-8, +11 | .519 | 2.5¢ (2 / 3) | 3¢ | 2.7% | 1.2% |
| Bears | 11-6, +26 | .550 | 2.5¢ (2 / 3) | 2¢ | 2.7% | 1.5% |
| Jaguars | 13-4, +138 | .490 | 1.5¢ (1 / 2) | 3¢ | 1.6% | 9.7% |
| Giants | 4-13, -58 | .498 | 1.5¢ (1 / 2) | 1¢ | 1.6% | 0.3% |
| Steelers | 10-7, +10 | .495 | 1.5¢ (1 / 2) | 1¢ | 1.6% | 1.6% |
| Cowboys | 7-9-1, -40 | .493 | 1.5¢ (1 / 2) | 3¢ | 1.6% | 0.7% |
| Colts | 8-9, +54 | .465 | <1¢ (0 / 1) | 2¢ | 0.5% | 3.6% |
| Buccaneers | 8-9, -31 | .491 | <1¢ (0 / 1) | 1¢ | 0.5% | 0.8% |
| Falcons | 8-9, -48 | .465 | <1¢ (0 / 1) | 1¢ | 0.5% | 0.8% |
| Saints | 6-11, -77 | .434 | <1¢ (0 / 1) | 1¢ | 0.5% | 0.5% |
| Panthers | 8-9, -69 | .521 | <1¢ (0 / 1) | 1¢ | 0.5% | 0.5% |
| Dolphins | 7-10, -77 | .541 | <1¢ (0 / 1) | 1¢ | 0.5% | 0.4% |
| Browns | 5-12, -100 | .429 | <1¢ (0 / 1) | 1¢ | 0.5% | 0.4% |
| Commanders | 5-12, -95 | .502 | <1¢ (0 / 1) | 1¢ | 0.5% | 0.4% |
| Cardinals | 3-14, -133 | .538 | <1¢ (0 / 1) | 1¢ | 0.5% | 0.2% |
| Raiders | 3-14, -191 | .529 | <1¢ (0 / 1) | 1¢ | 0.5% | 0.2% |
| Jets | 3-14, -203 | .517 | <1¢ (0 / 1) | 1¢ | 0.5% | 0.2% |
| Titans | 3-14, -194 | .476 | <1¢ (0 / 1) | 1¢ | 0.5% | 0.2% |
The row that needs reading twice is the second one. San Francisco's 9-cent mid is arithmetic on a book that has 21 contracts bid at 6 and 17 offered at 12. The first bid with real size is 5,190 contracts at 4, the level the contract held every day from late August through kickoff, and the first offer with real size is 9,490 contracts at 14, with another 10,000 at 16. Read the 49ers as "4 if you are selling, 14 if you are buying," and read the 9.6% de-vig as what the arithmetic says, not what the depth says; the last trade, 4 cents, is the number the depth agrees with. Seattle's row is the opposite kind of book: 63 contracts bid at 8 and 3,000 at 7 against 21,579 offered at 10, which is a two-sided market with a firm ceiling. The Rams sit at 5 bid, 8 ask with 157 contracts at 5, 406 at 4 and 1,000 at 3 on the bid side and 9,829 offered at 9, a real book at a price a third of the 15 cents this page carried before Week 1. Two rows further down argue with the panel's whole method. Cincinnati went 6-11 at minus-78, the worst differential among the 20 teams priced above a penny, and trades 4 bid, 5 ask on 46,642 contracts of open interest, a full cent above Houston, which went 12-5 at plus-109 and drew the seventh-softest slate. Kansas City went 6-11 at plus-34, the widest gap between record and differential on the board, and sits at 3 bid, 4 ask. Read through the lens the panel trusts, the Bengals are the board's most expensive contract for what they did last year and the Chiefs the cheapest; read through the market's, both are priced on their quarterbacks, which the panel's card never saw.
More live boards from the same panel: NFL win totals 2026, the same 32 teams scored one line at a time; Super Bowl 2027 champion; NFL MVP 2026; and the NFC West division board, refreshed on Thursday morning before the Melbourne game. Each page carries its own "as of" stamp.
What Moved In Week 1
Two games have been played: New England at Seattle on Wednesday, September 9, and San Francisco at Los Angeles on Thursday, September 10, at the Melbourne Cricket Ground, which the Associated Press recap on ESPN calls the first NFL regular-season game in Australia. Here is the board before, at midnight and at this morning's read, for the seven contracts that either played or moved. "August 16" is the price this page carried before Week 1; "12:39 a.m." is the first post-Melbourne edition of this page.
| Contract | August 16 | Sept 11, 12:39 a.m. (bid / ask) | Sept 11, 7:55 a.m. (bid / ask) | Last trade | Week 1 |
|---|---|---|---|---|---|
| San Francisco 49ers | 5¢ | 35 / 49 | 6 / 12 | 4¢ | Won 27-7 at the Rams, Melbourne |
| Los Angeles Rams | 15¢ | 4 / 7 | 5 / 8 | 9¢ | Lost 7-27 to the 49ers |
| Seattle Seahawks | 9¢ | 1 / 5 | 8 / 10 | 10¢ | Won 13-10 over New England; Darnold hurt on the first possession |
| New England Patriots | 4¢ | 2 / 3 | 2 / 3 | 2¢ | Lost 10-13 at Seattle |
| Baltimore Ravens | 7¢ | 7 / 9 | 8 / 9 | 9¢ | At Indianapolis, Sunday 1 p.m. ET |
| Houston Texans | 3¢ | 2 / 5 | 3 / 4 | 5¢ | Hosts Buffalo, Sunday 1 p.m. ET |
| Jacksonville Jaguars | 2¢ | 1 / 3 | 1 / 2 | 3¢ | Hosts Cleveland, Sunday 1 p.m. ET |
The 49ers, print by print. The San Francisco contract showed a 4-cent bid and a 5-cent ask on every daily candle from August 25 through Thursday's kickoff. The first move came at 10:37 p.m. ET, with the 49ers leading 17-7 late in the third quarter: 2,400 contracts bought at 5 cents in a single second, followed four seconds later by 18 contracts at 13. Brock Purdy's third touchdown pass, to Deebo Samuel, made it 24-7 at 10:44 p.m., and 50 more contracts went at 13 two minutes after it. Eddy Pineiro's 56-yard field goal made it 27-7 at 10:57 p.m., and at 11:00 p.m. a run of 236 contracts printed at 13, 15 and 26. At 11:12 p.m., with the game in its last five minutes, 8 contracts traded at 26 and 3.51 contracts at 50, the high print of the contract's life. The final whistle came at 11:23 p.m.; the 49ers had won 27-7, with Brock Purdy 25-of-34 for 205 yards and three touchdowns and Matthew Stafford 15-of-25 for 155, each throwing one interception, per ESPN's box score. In all, 2,716 contracts changed hands between 10:37 and 11:12 p.m., against 16,694 contracts of open interest.
Then, at 12:36 a.m. ET, 73 minutes after the game ended, 262 contracts of San Francisco were sold at 49 cents, in seven fills over three seconds. The same pattern, 262.06 contracts sold at the bid, hit 20 of the 32 team contracts on this board inside the same three seconds: Baltimore at 7 to 9, Buffalo at 6, the Rams at 4 to 6, Seattle at 1 and 2, Jacksonville at 1 and 2, and on down to the Giants and Steelers at 1. Whoever sold, sold every team they could find a bid on, at the bid. On the 49ers that bid was 49 cents, though a snapshot taken 34 seconds earlier had shown the book at 10 bid, 50 ask. The price history does not say who or why, and this page will not guess. What it does say is that the only trade ever printed above 26 cents on the 49ers is that 262-contract sale, plus the 3.51 contracts that set the high. The 35-cent bid that survived it was 3.51 contracts deep.
That bid did not last the hour. A second read at 1:19 a.m. ET showed San Francisco at 10 bid, 14 ask, last trade 14 cents, with 27.5 contracts on the bid. The 35 and the 49 were gone, and the book had settled where the depth said it would.
By morning, the 49ers were 4 cents. Between 1:32 and 4:33 a.m. ET, buyers took about 315 contracts at 12, 13 and 14 cents, the largest single fill 106 contracts at 13 at 1:37 a.m. Then, at 5:02:57 a.m., one seller hit every bid on the book down to 5 cents in a single second: 567 contracts in seven fills at 10, 9, 6 and 5 cents, the largest of them the 448 contracts at 9 that this page had named at midnight as the first bid with real size. At 5:39:14 a.m. a second sale of 145 contracts walked the price from 9 to 4, with 97 contracts printing at 4 cents, the lowest print since Thursday's kickoff and the contract's last trade at our 7:55 a.m. read. The book behind that print has 21 contracts bid at 6 and 17 offered at 12; the size sits at 4 on the bid, 5,190 contracts, and at 14 on the offer, 9,490. In the 24 hours to the read the 49ers' contract traded 4,535 contracts, and its open interest is 17,453. The price history gives no reason for the 5:02 a.m. sale and this page will not add one; what the book does explain is the size of the drop, because the bids above 9 were never more than a few dozen contracts deep.
The sourced reason for the 49ers' move is the game itself: a 20-point win over the market's favorite at a neutral site where the Rams were the designated home team, in the first of the six NFC West games this page flagged in August as the thing most likely to separate the top of the board. The size of the move was another matter. A team that is 1-0 with 16 to play is not a 49% proposition for the best record in a 32-team league, and by 5:03 a.m. the 448 contracts that had been waiting at 9 cents were the price. The mid is what it is; the depth is the price, and on this contract the depth was right within five hours.
The Rams, before and after. Los Angeles closed between 13 and 16 cents on every session from August 30 through Wednesday, and at 6:44 p.m. ET on Thursday, 2,951 contracts were bought at 16. At 7:55:48 p.m., 41 minutes before the first snap, 2,212 contracts were sold in a single second, walking the price from 15 down to 5, the largest fill 1,043 contracts at 11. Buyers stepped back in, and by 8:54 p.m., early in the first quarter, 148 contracts had printed at 16 again. This page has no sourced reason for the 7:55 p.m. sale and will not invent one.
The game supplied the rest. From 10:45 p.m., with the 49ers up 24-7 and then 27-7, the Rams were sold from 7 down to 1 cent on small size, 10 contracts printing at a penny at 10:47 p.m. Between 11:10 and 11:12 p.m., 722 contracts were bought back from 5 up to 9. The 12:36 a.m. sweep took 262 contracts out at 4 to 6, and the book stood at 4 bid, 7 ask at midnight, with 10,264 contracts offered at 9. Overnight the buyers came back: between 1:27 and 5:12 a.m. ET, 682 contracts of the Rams were bought at 7, 8 and 9 cents, 272 of them in one fill at 9 at 2:02 a.m., and the last trade at our 7:55 a.m. read is 9 cents against a book of 5 bid, 8 ask, with 9,829 contracts still offered at 9. The Rams traded 8,419 contracts in the 24 hours to the read, the most of any contract on the board, and their open interest has grown from under 49,000 on August 26 to 82,161, also the largest on the board, so this is the one contract where the price is backed by real positions on both sides.
Seattle won, got cheaper, then got dearer. The Seahawks beat New England 13-10 on Wednesday night, Drake Maye throwing three interceptions and Drew Lock, in relief of Darnold, going 16-of-22 for 187 yards and a 45-yard touchdown to Jaxon Smith-Njigba. The contract had closed at 6 bid, 7 ask on every session from August 26 through Tuesday. The game ended at 11:24 p.m. ET Wednesday; within three minutes 1,940 contracts were bought at 5 to 7, and at 12:16 a.m. Thursday 1,153 contracts printed at 3 cents, the low; and between 12:43 and 12:45 a.m. Thursday, about 3,930 contracts were bought at 7, 8 and 9. Around midnight on Thursday night, the start of Friday, it printed 7 to 9 again on buying that lifted offers rather than resting bids, and when the 12:36 a.m. Friday sweep arrived the bid side held only 1 and 2 cents, so its 262 contracts sold there.
The sourced reason is Sam Darnold. He was hurt on Seattle's opening possession, and on Thursday, per ESPN's Adam Schefter and NFL Network's Ian Rapoport, an MRI showed the hip is intact and the injury appears muscular. Head coach Mike Macdonald called that "really, really good news." Sources told ESPN Darnold is expected to miss the Seahawks' next game, at Arizona on September 20, and asked whether he would play, Macdonald said: "I don't know yet. In my mind I'm kind of thinking Drew is going to play." Seattle is 1-0, its quarterback is expected back soon, and at midnight its best-record contract was a 1-bid, 5-ask book with 20,000 contracts offered at 10: the panel's favorite going cheaper for a reason the panel never saw. This page said at the time it was the row most likely to look different by Sunday night. It did not take until Sunday. At the 1:19 a.m. ET follow-up read Seattle showed 7 bid, 8 ask; at 1:38 a.m. about 261 contracts were bought at 8 and 9; at 6:06 a.m. another 21 printed at 10; and at our 7:55 a.m. read the Seahawks are 8 bid, 10 ask, last trade 10 cents, with 63 contracts at 8 and 3,000 at 7 on the bid and 21,579 offered at 10. That is the highest bid on the board, shared with Baltimore, and two cents above the 6-bid, 7-ask book Seattle had closed at every session of the week before the injury.
Baltimore, quietly, holds the deepest bid. The Ravens have not played, and their contract barely moved: 7 cents on August 16, 7 bid and 9 ask at midnight, 8 bid and 9 ask at 7:55 a.m. after 52 contracts printed at 9 at 7:38 a.m. What changed is what stands behind the number. At 6:02 p.m. ET on Wednesday, September 9, before the season's first snap, 29,728 contracts of Baltimore were bought at 9 and 10 cents in a single second; four minutes later, 31,214 contracts of Minnesota were bought at 3 cents, the largest single fill on this board in our window. The Ravens' bid carried 19,217 contracts at 7 at midnight and 17,654 at 7 this morning, with 76 more at 8; either figure is more than the entire open interest of the 49ers' contract, 17,453. Baltimore opens at Indianapolis on Sunday at 1 p.m. ET. The 14 game lines left in Week 1, Sunday and Monday, the next 14 results this board has to absorb, are on the live odds screen, where you can shop the number across every major book.
Across the whole board, 36,098 contracts traded in the 24 hours to the read, open interest stands at 811,594 contracts, and the sum of the 32 mids is 94 cents, with the asks summing to 115 and the bids to 73.
Every Seat's Number
The full per-model board for the top 12, exactly as generated on August 16 and not re-scored since. What stands out is how little the seats disagree at the top and how much they collectively disagree with the market: all eight put Seattle first or second, all eight put Jacksonville at or above 9.0%, and all eight put Baltimore below 4%. None of them saw Week 1.
| Team | Fable | Opus | Sonnet | ChatGPT | Gemini | GLM | Kimi | DeepSeek | Blend |
|---|---|---|---|---|---|---|---|---|---|
| Seattle Seahawks | 11.5% | 12.7% | 12.5% | 11.2% | 11.5% | 11.5% | 11.5% | 10.8% | 11.7% |
| Los Angeles Rams | 10.0% | 10.9% | 9.5% | 10.5% | 11.5% | 9.5% | 9.5% | 9.7% | 10.1% |
| Jacksonville Jaguars | 9.5% | 9.8% | 10.0% | 9.5% | 10.0% | 9.1% | 9.0% | 10.7% | 9.7% |
| Houston Texans | 8.5% | 8.7% | 8.5% | 8.7% | 9.5% | 8.7% | 8.3% | 9.7% | 8.8% |
| New England Patriots | 6.5% | 8.8% | 9.5% | 7.7% | 7.5% | 7.0% | 7.5% | 7.7% | 7.8% |
| Buffalo Bills | 5.5% | 5.6% | 7.0% | 6.5% | 6.0% | 6.0% | 6.5% | 6.7% | 6.2% |
| Detroit Lions | 6.5% | 5.6% | 6.0% | 5.8% | 7.0% | 6.6% | 5.0% | 7.2% | 6.2% |
| Philadelphia Eagles | 5.0% | 4.4% | 5.0% | 5.2% | 5.5% | 5.5% | 5.0% | 5.7% | 5.2% |
| Denver Broncos | 4.5% | 4.4% | 7.0% | 4.2% | 4.0% | 4.5% | 5.0% | 2.5% | 4.5% |
| San Francisco 49ers | 4.0% | 4.0% | 5.0% | 4.7% | 5.0% | 4.6% | 4.5% | 3.5% | 4.4% |
| Indianapolis Colts | 3.8% | 4.2% | 2.3% | 3.8% | 3.0% | 3.6% | 3.5% | 4.7% | 3.6% |
| Baltimore Ravens | 3.5% | 3.0% | 2.0% | 3.3% | 3.5% | 3.0% | 3.0% | 3.5% | 3.1% |
Every seat on this panel is graded against real market settlements — records to date: Gemini 86% on 4,292 graded calls · GLM 81% on 2,367 graded calls · Kimi 83% on 2,345 graded calls · DeepSeek 80% on 2,392 graded calls. Recomputed daily; the full scoreboard is public.
These are model estimates, not predictions of fact and not financial or betting advice. Every number in this piece gets graded in public once the market settles, and you can see the running record on the full graded scoreboard.
The seat that looks worst after one game is the one that had the 49ers lowest, DeepSeek at 3.5%, and the seats that look best are Sonnet and Gemini at 5.0%. Neither deserves the grade yet. One road win in September is a data point, not a season, and the panel's own case rested on differential and schedule over each team's 17-game season rather than on any single result. The market's own reaction is the more interesting one to watch.
Seattle Seahawks: The Panel's Favorite, On A Number The Market Doesn't Rate
Seattle is the team whose 2025 season agrees with itself most completely. It won 14 games, it outscored opponents by 191 points, the best figure in the league by 19 points over the Rams, and it drew a middling 2026 schedule at .514. The panel's case, in the August 16 run, is that a differential that large is the closest thing football has to a repeatable strength signal, while a 14-3 record on its own is not.
The market had Seattle at 9 cents when the panel ran on August 16, second on the board behind only the Rams. The panel blended it to 11.7%, and every one of the eight seats placed it first or second. Sonnet put the gap plainly, describing Seattle as combining "the single best 2025 differential (+191, clearly ahead of #2 at +172) with a merely average 2026 schedule, giving it both the highest floor and a real ceiling."
The counterweight was the same 16% repeat rate that hurts New England and Denver. Seattle is defending the best record in football, and history says teams in that position usually give it back. Week 1 added a second counterweight the panel could not have known about: the differential it trusted was built with Darnold at quarterback, and Seattle's next game, at Arizona, is expected to be played without him. The 13-10 win over New England counts the same as any other in the standings; what it says about a 17-game ceiling with Drew Lock starting is the open question the market priced at 1 bid, 5 ask at midnight and 8 bid, 10 ask by morning.
Jacksonville Jaguars: The Gap Week 1 Didn't Touch
This is the disagreement worth arguing about, and Week 1 did not touch it. Jacksonville went 13-4 last season with a plus-138 point differential, fourth-best in the league. Its 2026 opponents went .490 last season, the 11th-softest slate in the NFL. On the panel's two main inputs, high floor and easy path, the Jaguars grade out as a top-four team.
Kalshi priced them at 2 cents in August and prices them at 1 bid, 2 ask now, about 1.6% on the de-vig mid, which is a rounding error away from treating a 13-win division winner as a longshot. The panel blends them at 9.7%, and no seat went below 9.0%. That is a six-fold disagreement, the largest ratio gap on any contract with a two-sided book, and with Seattle's book repaired overnight it is again the largest gap in points, 8.1. Someone bought before dawn: at 3:17 a.m. ET, 904 contracts of Jacksonville were bought at 2 and 3 cents, 533 of them at 3 in one fill, and 170 more went at 2 and 3 at 5:11 a.m.; 1,073 contracts in all, the first real buying on the contract in our window, against 36,376 contracts still offered at 3. The Jaguars' first test is Cleveland at home on Sunday at 1 p.m. ET.
Houston is the same argument at a slightly lower volume: 12-5, plus-109, the seventh-softest 2026 schedule at .474, priced at 3 cents in August and 3 bid, 4 ask now, against an 8.8% blend. Opus grouped them together, describing Jacksonville specifically as a team that "pairs the fourth-best 2025 differential (+138) with a .490 opponent slate," and concluding that "that combination of a high floor and a soft slate is exactly what a best-record run needs." Houston hosts Buffalo on Sunday, a game between two teams the panel had inside its top six.
The honest caveat: the panel's card carried last season's results and this season's opponents, not this season's rosters. If the market's price on Jacksonville reflects an offseason the models underweighted, the market is right and the gap is an artifact. That is precisely the kind of call this page exists to settle in public.
The Los Angeles Rams: Where The Panel Faded The Favorite
The Rams were the one place the panel was less enthusiastic than the market. Los Angeles went 12-5 with a plus-172 differential, which the panel treats as elite and better than its record. It also drew a .516 schedule, just harder than the league median, and it was the one team the market had clearly singled out, at 15 cents.
The panel blended the Rams to 10.1%, a clear second on its own board but well under the 12.8% the market implied after de-vigging in August. The disagreement was never about whether the Rams are good. It was about what a single team's share of this particular race can be worth: the winning number has been shared in nine of the last 20 seasons, and the team holding the best record has given it back the following year in 16 of 19 chances.
One game later the market has gone past the panel. At 5 bid, 8 ask the Rams' de-vig mid is 6.9%, well under the panel's 10.1%, and the roles have swapped: the panel that faded the favorite in August now has Los Angeles on the cheap side of its line too, a 3.2-point gap, behind Jacksonville, Houston and New England. That is what a 27-7 loss does to a contract whose whole case was a high floor. It is also one game, the plus-172 differential the panel leaned on was built over 17 of them, and 682 contracts were bought back overnight at 7 to 9 cents.
Green Bay was the sharper version of the same fade, at 6 cents on August 16 against a 1.9% blend, on a plus-31 differential and a schedule tied with Arizona for the third-hardest in the league; the Packers now sit at 2 bid, 5 ask ahead of Sunday's game at Minnesota. Baltimore, at 8 bid and 9 ask against 3.1%, and the Chargers, at 4 bid and 5 ask against 1.8%, are the other two places the panel thinks the board is paying for reputation over the two numbers it trusts. Baltimore is now the board's deepest bid, so that disagreement has real money on the other side of it.
The Tie Rule That Caps This Whole Board
Here is the structural fact that shapes every number above, and it is the reason no model on this panel put any team over 12.7%.
Kalshi's own rules for these contracts read: "If two or more teams are tied for the best regular-season record, official NFL tiebreaker procedures will be applied in order until a single team remains. If multiple teams remain tied after all applicable tiebreakers, dead-heat rules apply and each tied team's market will resolve at 1 / N, where N is the number of tied teams."
Ties at the top are not exotic. The best record was shared in nine of the last 20 seasons. In the 17-game era specifically, which began in 2021, it has been shared in four of five: Green Bay and Tampa Bay at 13-4 in 2021, Kansas City and Philadelphia at 14-3 in 2022, Detroit and Kansas City at 15-2 in 2024, and last season's three-way tie at 14-3. Only Baltimore in 2023 finished alone at the top.
The word doing the quiet work in Kalshi's rule is "applicable." The NFL publishes tie-breaking procedures for division ties and for wild-card ties inside a conference, plus a separate set for draft order at the selection meeting. It does not publish a procedure for ranking an AFC team against an NFC team in the standings. So two teams tied at the top of the league from the same conference will usually be separated by head-to-head, conference record or strength of victory, while a tie that spans the two conferences has no seeding tiebreaker to apply and heads for the 1/N split.
Last season is the worked example. Seattle finished 14-3 in the NFC; New England and Denver finished 14-3 in the AFC. The two AFC teams could be separated by conference procedures. Separating the survivor from Seattle is the case the rulebook does not cover.
The panel found this on its own and then split on how much it matters, which is the most interesting disagreement of the run. Most seats put somewhere between a quarter and 40% of a leading team's value into shared outcomes. Opus argued for less, and named the exact mechanism: "Ties occur in 45% of seasons, but the contract applies NFL seeding tiebreakers until one team remains, and those reliably separate same-conference clubs. Only cross-conference standoffs (JAC/NE/HOU vs SEA/LAR) actually dead-heat at 1/2 or 1/3."
One more thing worth knowing before acting on any of this: Kalshi has never settled this contract. The 2026-27 season is the only event the exchange lists in the series as of this read, so the dead-heat clause has never been tested here in practice. It is also the rule that makes a 49-cent price on any single team so hard to justify: even the favorite has to land in the 55% of seasons where the best record is not shared.
Where The Panel Changed Its Mind
The blend above comes from a second scoring round on August 16. Each model priced the board alone, then read the other seven boards and their reasoning with the names stripped out, and could revise. All eight seats completed both rounds. Three moves are worth showing.
DeepSeek made the largest revision on the board. In round one it led with Houston at 10.0% and had Seattle down at 6.5% and New England at 3.5%, with Cincinnati carried far above the field. After reading the rest of the panel's differential-first case, it lifted Seattle to 10.8% and New England to 7.7%, cut Cincinnati by six points, and settled on a summary of "point differential and schedule strength drive the board."
Gemini moved hardest on the Rams, raising them from 8.0% to 11.5%, the highest number any seat gave Los Angeles. It had been the seat most aggressive about fading 2025 records generally, and it accepted from the others that plus-172 on a 12-5 season is the profile that fade is supposed to reward, not punish.
The seat that refused to move was Opus, on ties. Every other model held that the 45% shared-record rate should haircut the leaders by roughly a third. Opus revised its numbers, lifting Seattle from 9.2% to 12.7%, but held its dissent, arguing that "the tie rule flattens less than the panel thinks because NFL tiebreakers resolve intra-conference ties first." It ended the run with the highest single number on the board and the narrowest view of the dead-heat discount, and it is the only seat whose position on ties survived contact with seven others.
Dark Horses The Panel Won't Dismiss
Below the top five, several teams earned a named reason in the August 16 run to stay off the floor.
- Indianapolis Colts (3.6% blend against a 0-bid, 1-ask book). An 8-9 record hiding a plus-54 point differential, paired with the fourth-softest 2026 schedule at .465. DeepSeek carried the Colts highest at 4.7%. This is the panel's clearest example of its two favorite signals pointing the same way at a team nobody is discussing, and at 5:11 p.m. ET on Thursday someone bought 4,679 contracts of it at 2 cents. The Colts host Baltimore on Sunday.
- Detroit Lions (6.2% against 6 bid, 7 ask). A 9-8 season worth more than that on differential at plus-68, behind the sixth-softest slate at .467. Detroit shared the best record as recently as 2024, so this is not a hypothetical ceiling.
- Philadelphia Eagles (5.2% against 4 bid, 5 ask). The Eagles carry 11,054 contracts bid at 4 cents, one of the deepest bids on the board, and the 12:36 a.m. sweep sold 262 into it without moving the price. The panel's August case was plus-54 behind the 10th-softest schedule at .481; no seat put the Eagles in its top four, and no seat put them outside its top nine either.
- Pittsburgh Steelers (1.6% against 1 bid, 2 ask). Another deep bid at a low price: 11,666 contracts at a penny, also swept at 12:36 a.m. The panel will not price a 10-7 team at the very bottom of the board on a plus-10 differential and an average slate, even though the market effectively does.
What Would Change The Panel's Mind
In August the seats named the specific, checkable developments that would move this board. Here is where each stands after Week 1.
- A quarterback injury to any of the top five. Happened, first. Sam Darnold was hurt on Seattle's opening possession of the season and is expected to miss Week 2 at Arizona, with an MRI showing the hip intact and the injury muscular. The market took Seattle's book from 6/7 to 1/5 at midnight; buyers had it back to 8/10 by 7:55 a.m. Of the other four named quarterbacks, Drake Maye and Matthew Stafford played their openers and neither appears on his team's post-game injury report, which listed Davis Allen and Kam Curl for the Rams and, for Seattle's opponent, nobody at quarterback; Jacksonville and Houston play on Sunday.
- Jacksonville and Houston starting fast. Pending. Both play at 1 p.m. ET on Sunday, the Jaguars at home to Cleveland and the Texans at home to Buffalo. If the panel's two big buys open 4-1 or better against those soft slates, the 1-cent and 3-cent bids should move sharply toward the panel. A 2-3 start pushes the argument back toward the market.
- The NFC West separating. One of six played. Seattle, the Rams and San Francisco play six games against each other, and the first of them, in Melbourne, went to the 49ers by 20. Rams-Seahawks in Week 16 and Week 18 still bookend the race. The market's answer so far is the 49ers' round trip described above, 5 cents to 50 and back to 4; the panel's answer is unchanged until it is re-scored.
- The November 2026 trade deadline. Unchanged. The last roster-shaping catalyst before records are locked. A contender adding at the deadline, or a 4-5 team selling, moves this board more than anything else on the calendar between now and January.
Settlement Timeline
| Regular season kicked off | Wednesday, September 9, 2026 (New England at Seattle); Week 1 concludes Monday, September 14 |
| Trade deadline | Early November 2026, the next roster-shaping catalyst |
| Best record decided | End of the 2026-27 regular season, early January 2027, at 17 games per team |
| Contract settles | Kalshi lists an expected expiration of February 8, 2027, once records are final; the playoffs in between do not count |
| Tie handling | NFL tiebreakers first; any team still tied is paid 1 divided by the number of tied teams |
The prices on this page carry an "as of September 11, 2026, 7:55 a.m. ET" stamp, and the blend carries an "as of August 16, 2026" stamp; the panel is re-scored when the story moves, not on every price move. Every settled call is graded in public.
How To Read These Prices
A few things worth knowing before you read any board like this one, and especially this one after Week 1.
- A wide book is a missing order, not a view. San Francisco at 35 bid, 49 ask at midnight meant one side of the market had gone missing, not that the market thought the 49ers were a 42% proposition; by 5:03 a.m. the 448-contract bid at 9 had been hit and the price was 5. At 6 bid, 12 ask this morning the same rule applies in both directions: 21 contracts on the bid and 17 on the offer, with the size at 4 and 14. On a book like that, the depth is the price and the mid is a formality.
- This board's overround lives in the asks. The 32 asks add up to 115 cents and the 32 bids to 73, so anyone buying the whole board pays 15 cents over a dollar and anyone selling it collects 27 under. The mids add to 94, which is why the de-vig column above nudges each mid up rather than down; at midnight, when the 49ers' ask was 49, the same 32 mids added to 121. Every comparison in this piece uses the de-vig column.
- On this board this morning, the tradable price on San Francisco is 4 to 14, not 9. The mid is 9 because the book is 6 bid, 12 ask on a few dozen contracts; the size sits at 4 on the bid and 14 on the offer. At midnight this page said a seller of 449 contracts would take the price to 5 in one print; at 5:02 a.m. a seller of 567 did exactly that. Where the panel and the market disagree, as they do on Jacksonville and now on the Rams, that gap is a question to investigate, not an answer; where the book and the mid disagree, as they do here, the book wins.
- Thin markets move, and Week 1 showed how. A 262-contract sale printed at 49 cents on the 49ers and at a penny on Seattle within the same three seconds, and five hours later the 49ers printed 4. Kalshi charges trading fees on these contracts and sets a position accountability level of $25,000 per strike per member, both of which matter more on a 2-cent contract than a 49-cent one.
- The panel is working with last season's results, not this season's roster or results. That is a deliberate design choice, and it is also the honest limit of what these numbers can tell you. The market has now seen two games and one quarterback injury that the panel has not.
The panel scored the whole-season championship version of this question in the Super Bowl 2027 champion model verdict, and the team-by-team win lines behind these numbers live in the 2026 NFL win totals board. The two teams at the top of the August board also have to get past each other and the 49ers first, and the NFC West division board prices that race. For picks rather than prices, the free expert picks page carries today's board.
The Bottom Line
Two games in, the market and the panel have swapped sides on the Rams and agree on almost nothing else, and the one thing the 49ers' round trip settled is how to read this board: on a contract where the bids above 9 cents were never more than a few dozen contracts deep, the depth was the price, and the mid was a formality that lasted five hours. The deepest real money on the board is still on Baltimore, a team every seat put under 4%, and the panel's own favorite, Seattle, won its opener, lost its quarterback for at least a week, and by morning shared the highest bid at 8 cents. Underneath all of it sits the shape of the race itself. The best record has been shared by two or more teams in four of the last five seasons, and Kalshi applies the NFL's tiebreakers first, paying each still-tied team 1/N only if the tie survives them. That ceiling is one reason no seat on this panel put any team over 12.7%, and it is the same reason a 49-cent print on a 1-0 team read as a thin book rather than a verdict, five hours before the book proved it. Fourteen more Week 1 games arrive Sunday and Monday. Every call on this page gets graded once the season is done.
Prices on this page are Kalshi's book. If you also trade on Polymarket, code OS4 gets new users a $50 trading bonus on a $10 deposit — affiliate link; terms as stated by Polymarket; 18+, availability varies by state.
More on this: NFL 2026 on Kalshi: 49 Live Boards, 10 Days To Kickoff · NFL Worst Record Odds 2026: The AI Panel Vs. Kalshi · NFL Last Undefeated Team Odds: The Rams Drew The Wrong Path · NFL Week 1 Kalshi Player Props: Six Contracts That Stick Out · NFL Week 1 Player Props: Six Numbers That Stick Out
