Kickoff week is the first time a Heisman Trophy odds board stops being a preseason argument and starts being a scoreboard, and the Kalshi board arrives there in a specific shape. It carries 36 separate player contracts now, up from 29 in July. 397,134 contracts changed hands in the last 24 hours alone. Only two of the 36 have played a 2026 snap. Most of the rest open between Thursday night and Monday (three of the 36 already carry a status flag against them: one is not playing this season, one is hurt, one lost his starting job in camp, all covered below), which makes this the last honest moment to compare what the crowd believes to what a price-blind read of the same players says. Since our July 30 scoring, the market has walked toward the models on most of its top ten: it trimmed CJ Carr, trimmed Arch Manning, cut John Mateer by more than a third, and bid Dante Moore up by almost half again. Among the ten highest July prices, the one that sits exactly where our July 30 snapshot left it is the one the models care about most. Julian Sayin is 6.5¢ today, the same mid as July 30 after 443,418 contracts changed hands on his book in between, and he is the panel's re-scored favorite, with an edge over the market that is wider than it was in July. (Thirteen contracts further down the board also show the same mid as July, on books where a single cent is the whole spread; the top ten is where a cent means something.)
This page re-fetches every Kalshi price at draft time, re-runs a price-blind AI panel on a card that now carries the fall-camp status of every listed player and the two box scores that exist, and stacks the two side by side on all 36 names. The movers come first, then the full board, then the single pattern that explains most of the gap column, then the individual stories on the widest gaps. One of those stories, tied for the market's biggest rise since July, is the name the panel cut the hardest.
The Quick Answer
As of September 3, 2026, the market's favorite is Notre Dame's CJ Carr at 11.5¢ and its second choice is Texas' Arch Manning at 9.5¢. A four-seat, price-blind AI panel scores Ohio State's Julian Sayin first at a 12.0% seat median against a 6.5¢ price, and puts Carr at 5.5%, still the widest fade on the board. Since July 30 the market has moved toward the panel on Carr, Manning, Moore, Mateer and Hoover, and away from it on Jeremiah Smith and Gunner Stockton. The full board, every seat's number, the seven new listings and the three contracts the August status flags killed are all below.
- Market Pick: CJ Carr, 11.5¢ (about 10.8% once you re-scale a board whose prices sum to 106¢)
- Model Pick: Julian Sayin, 12.0% seat median (market 6.5¢)
- Biggest Model Fade: CJ Carr, 5.5% against 11.5¢, a 5.4-point gap
- Biggest Sleeper Gap: Gunner Stockton, 5.8% against 2.5¢, and the market moved the wrong way on him since July
- The Panel's Logic In One Line: it pays for completion percentage on a top-five program, with the AP poll, the schedule and the August status flags as the multipliers, and the market pays for the stage. Sayin's 77.0% at the consensus No. 1 sits at 6.5¢; Manning's seven ranked games at 61.4% sat at 11.5¢ in July. The three names the panel prices meaningfully above the market own the three best 2025 completion rates among the top ten's returning starters.
- Where The Edge Lives, In Cents: the panel's 12.0% on Sayin is worth about 12.7¢ on a board that sums to 106¢, and the panel's most conservative seat, Fable's 11.0, is worth about 11.7¢, so at 6.5¢ the market is at roughly half of either read. Carr's 5.5% is worth about 5.8¢, so at 11.5¢ the panel is not fading a bad quarterback, it is fading a price almost double its number.
Stay ahead of the markets.
Daily insights and expert picks on Kalshi, Polymarket, and what's moving markets.
Free forever. Unsubscribe anytime.
What Moved Since July 30
The panel was re-run this week, so both columns changed, and the honest way to read the table is to ask whether each market move closed the July gap or opened it. Prices are Kalshi bid-ask midpoints, July 30 against September 3 at 6:00 PM ET. The panel column is July's eight-seat number against this week's four-seat median.
| Player | Kalshi Jul 30 | Kalshi Sep 3 | Panel Jul 30 | Panel Sep 3 | What the market did |
|---|---|---|---|---|---|
| Dante Moore, Oregon | 5.5¢ | 8¢ | 8.0% | 10.2% | Came to the panel, and the panel moved higher |
| Jeremiah Smith, Ohio State | 4¢ | 6.5¢ | 3.6% | 2.1% | Went the other way; the gap opened to 4.0 |
| Arch Manning, Texas | 11.5¢ | 9.5¢ | 8.7% | 9.0% | Came all the way; the gap is now zero |
| John Mateer, Oklahoma | 5.5¢ | 3.5¢ | 1.2% | 2.1% | Came most of the way |
| CJ Carr, Notre Dame | 12.5¢ | 11.5¢ | 4.6% | 5.5% | Narrowed, still the widest fade |
| Darian Mensah, Miami | 9¢ | 8¢ | 7.5% | 7.0% | Both moved down; priced about right |
| Gunner Stockton, Georgia | 3.5¢ | 2.5¢ | 5.0% | 5.8% | Went the other way; now the biggest sleeper gap |
| Trinidad Chambliss, Ole Miss | 6.5¢ | 7.5¢ | 6.0% | 5.0% | Crossed from fair to a mild fade |
| Josh Hoover, Indiana | 5.5¢ | 4.5¢ | 3.9% | 4.0% | Came to the panel; now priced about right |
| Julian Sayin, Ohio State | 6.5¢ | 6.5¢ | 10.5% | 12.0% | Did not move; the gap opened to 5.8 |
Model estimates, not financial advice.
The row worth staring at is the last one. Every other name among the market's ten highest July prices moved at least a cent, and most of those moves closed distance to the models. Sayin's mid came back to exactly where the July snapshot had it, through five weeks of camp reporting, an AP preseason poll that made his team the consensus No. 1 with 40 first-place votes, and a panel that raised him. That is the forward promise of this page: by the end, the question is not whether the models like Sayin, it is why the crowd, which agreed with them on nearly everything else, has not paid him a cent more.
The Market At A Glance
| Venue | Kalshi, a CFTC-regulated event-contract exchange (18+; availability varies by state, as of September 2026) |
| The Event | "Heisman Trophy Winner" for the 2026-2027 college football season (Kalshi event KXHEISMAN-27), 36 separate yes/no player contracts |
| The Contract | Verbatim: "If [player] wins the College Football Heisman Trophy in the 2026-2027 college football season, then the market resolves to Yes" |
| Settlement | Per Kalshi's sport-award contract terms, resolution follows the award's official announcement; the finalists are named December 7 and the trophy is presented December 12, 2026 |
| The Catch | There is still no "Field" contract: if anyone not listed wins, all 36 contracts resolve NO. There is also no delisting clause, which matters for one name below |
| Timing | Contracts carry an expected expiration of January 1, 2027 and can close early once the winner is announced |
| Volume | The exchange's counter reads 11,647,995 contracts traded across the event, up from 3,987,073 on July 30, with 397,134 of them in the last 24 hours |
| Prices Below As Of | September 3, 2026, 6:00 PM ET (bid-ask midpoints; 36 independent order books, so the prices sum to about 106¢ rather than 100) |
| Source | Kalshi's public market API, event KXHEISMAN-27, fetched September 3, 2026 at 6:00 PM ET (22:00 UTC): bid, ask and last trade for all 36 contracts, volume and open interest as contract counts. The July 30 baseline is the same fetch from that date, kept in our ledger |
These are model estimates, not predictions of fact and not financial advice. Kalshi event contracts trade on a CFTC-regulated exchange; you must be 18 or older and in an eligible state to participate.
Every number in this piece gets graded in public once the market settles: the full graded scoreboard keeps the panel's complete record, wins and misses alike.
What Actually Wins A Heisman
The award's structure still decides most of this board, and the panel worked from the same winner ledger since 2000 it used in July. Two iron laws:
- It Is A Quarterback Award, And The Seat Table Shows The Panel Pricing It That Way. 21 of the last 26 winners played quarterback. The five exceptions, Reggie Bush (2005), Mark Ingram II (2009), Derrick Henry (2015), DeVonta Smith (2020) and Travis Hunter (2024), each won on a historic individual season: four of them for teams that reached the national title game, and Hunter on a two-way workload nobody had logged in decades, at a 9-4 Colorado. That became the panel's test for every non-quarterback: never "can it happen," always "is THAT kind of season plausibly on the table for him." You can watch the test being applied in the per-seat table below. Jeremiah Smith had to clear two bars, a record receiving season and out-voting his own quarterback, and he is the only non-quarterback the panel keeps above 1%. Bo Jackson, Kewan Lacy, Jadan Baugh and Malachi Toney, four skill players with real 2025 production, all land between 0.4% and 0.6%, and Colin Simmons, the only edge rusher listed, gets 0.2%.
- It Tracks Team Success, And The Panel Now Has A Live Number For It. Whoever wins, his team was in the national conversation that year. The AP preseason poll, which the July card did not have, puts Ohio State first, Oregon second, Georgia third, Notre Dame fourth, Texas fifth, Indiana sixth and Miami seventh. Eleven of the 36 listed players start at quarterback for an AP top-11 program, and those 11 medians add up to about 67 of the panel's 100 points. The other 25 contracts and the unlisted field share the rest, which is the ledger's shape, not a stylistic choice.
One structural fact carries over from July and got sharper: the most recent winner, Indiana's Fernando Mendoza, took the 2025 trophy in a 16-0 national-title season, and the quarterback who now runs that Indiana offense is on this board at 4.5¢.
How This Board Was Built
Each model received the same fetched data card and no market prices. The card carries every listed player's roster-verified 2026 school, position and class (Darian Mensah is a Miami Hurricane, not a Duke Blue Devil, and the card says so), his full 2025 stat line, his program's complete 2025 record, the 2026 AP and Coaches preseason polls, each school's full 2026 schedule with its ranked opponents, the award's 2000-2025 winner ledger, and a dated status section: Alabama naming Keelon Russell its starter over Austin Mack on August 22, Tennessee naming true freshman Faizon Brandon on August 24, Missouri ruling Ahmad Hardy out of its opener with no timetable, and the reporting that Texas Tech will not play Brendan Sorsby in 2026. It also carries the only two 2026 box scores that exist, Jayden Maiava's and CJ Bailey's openers, with an instruction to weigh one game as one game.
Four of the panel's eight seats priced this cycle: Claude Fable, GPT, Gemini and GLM. Each priced the entire 37-outcome partition, the 36 contracts plus the field, in a single pass, with the numbers required to sum to 100. After a blind first round, each seat read the other three's anonymized numbers and reasoning and could revise once. The blend is the seat median of the revised round, not an average, and the first-round numbers are preserved in our ledger so the revisions stay honest. Every one of these calls was written to the public grading ledger the moment it was made. The July board was an eight-seat blend; this one is a four-seat median, so a move in the panel column is partly the season's new information and partly a different set of judges, and the per-seat table below lets you separate the two.
The Board: All 36 Names
Sorted by the panel's seat median. The gap column is the median minus the re-scaled market share (each price divided by the 106¢ the board sums to), so it is the panel's disagreement with the crowd in probability points.
| Player | School (Pos) | Kalshi Now | Kalshi Jul 30 | AI Blend | Gap |
|---|---|---|---|---|---|
| Julian Sayin | Ohio State (QB) | 6.5¢ | 6.5¢ | 12.0% | +5.8 |
| Dante Moore | Oregon (QB) | 8¢ | 5.5¢ | 10.2% | +2.7 |
| Arch Manning | Texas (QB) | 9.5¢ | 11.5¢ | 9.0% | 0.0 |
| Darian Mensah | Miami (QB) | 8¢ | 9¢ | 7.0% | −0.6 |
| Gunner Stockton | Georgia (QB) | 2.5¢ | 3.5¢ | 5.8% | +3.4 |
| CJ Carr | Notre Dame (QB) | 11.5¢ | 12.5¢ | 5.5% | −5.4 |
| Trinidad Chambliss | Ole Miss (QB) | 7.5¢ | 6.5¢ | 5.0% | −2.1 |
| Josh Hoover | Indiana (QB) | 4.5¢ | 5.5¢ | 4.0% | −0.2 |
| Marcel Reed | Texas A&M (QB) | 3.5¢ | 4.5¢ | 3.5% | +0.2 |
| Keelon Russell | Alabama (QB) | 3.5¢ | 2.5¢ | 2.7% | −0.6 |
| Sam Leavitt | LSU (QB) | 3.5¢ | 3¢ | 2.7% | −0.6 |
| Jayden Maiava | USC (QB) | 3.5¢ | 3.5¢ | 2.2% | −1.1 |
| Jeremiah Smith | Ohio State (WR) | 6.5¢ | 4¢ | 2.1% | −4.0 |
| John Mateer | Oklahoma (QB) | 3.5¢ | 5.5¢ | 2.1% | −1.2 |
| Bear Bachmeier | BYU (QB) | 1.5¢ | new listing | 2.0% | +0.6 |
| Demond Williams Jr. | Washington (QB) | 1.5¢ | 1.5¢ | 1.4% | 0.0 |
| Bryce Underwood | Michigan (QB) | 1.5¢ | 1.5¢ | 1.3% | −0.1 |
| Faizon Brandon | Tennessee (QB) | 1.5¢ | new listing | 1.3% | −0.1 |
| Will Hammond | Texas Tech (QB) | 1.5¢ | 2.5¢ | 1.2% | −0.2 |
| Devon Dampier | Utah (QB) | 0.5¢ | new listing | 1.1% | +0.6 |
| Rocco Becht | Penn State (QB) | 1.5¢ | 1.5¢ | 1.1% | −0.3 |
| Kevin Jennings | SMU (QB) | 1.5¢ | 1.5¢ | 1.1% | −0.3 |
| Kewan Lacy | Ole Miss (RB) | 0.5¢ | 0.5¢ | 0.6% | +0.2 |
| Byrum Brown | Auburn (QB) | 1.5¢ | 1.5¢ | 0.6% | −0.8 |
| Bo Jackson | Ohio State (RB) | 0.5¢ | 0.5¢ | 0.6% | +0.1 |
| Malachi Toney | Miami (WR) | 1.5¢ | 2.5¢ | 0.5% | −0.9 |
| LaNorris Sellers | South Carolina (QB) | 2.5¢ | 1.5¢ | 0.5% | −1.9 |
| Jadan Baugh | Florida (RB) | 0.5¢ | new listing | 0.4% | −0.1 |
| Jaron-Keawe Sagapolutele | California (QB) | 1.5¢ | 1.5¢ | 0.3% | −1.1 |
| Drew Mestemaker | Oklahoma State (QB) | 1.5¢ | 1.5¢ | 0.3% | −1.1 |
| Kamario Taylor | Mississippi State (QB) | 0.5¢ | new listing | 0.2% | −0.3 |
| Colin Simmons | Texas (EDGE) | 0.5¢ | new listing | 0.2% | −0.3 |
| CJ Bailey | NC State (QB) | 0.5¢ | new listing | 0.1% | −0.3 |
| Ahmad Hardy | Missouri (RB) | 0.5¢ | 0.5¢ | 0.1% | −0.3 |
| Austin Mack | Alabama (QB) | 0.5¢ | 0.5¢ | 0.1% | −0.4 |
| Brendan Sorsby | listed under Texas Tech; not playing in 2026 (QB) | 0.5¢ | 0.5¢ | 0.0% | −0.5 |
| The Field (Any Unlisted Player) | — | no contract | no contract | 10.0% | — |
Model estimates, not financial advice. The gap column is computed from unrounded seat medians, so two rows that display the same median can differ by a tenth. A 0.5¢ entry is a 0-bid, 1¢-ask book: the exchange's minimum tick shapes the bottom of the board, so read the longshot prices as "under a cent," not as precise probabilities.
More live boards from the same panel: Jeremiah Smith's Heisman odds, tied for the market's biggest rise (6.5¢) · Trinidad Chambliss' Heisman odds (7.5¢) · Jayden Maiava's Heisman odds after his opener (3.5¢) · the Heisman dark-horse case for Hoover (4.5¢) and Stockton (2.5¢) · the college football championship board.
The top row is the whole page in miniature. Sayin's 6.5¢ re-scales to about 6.1% of the board, and the four seats put him at 11.0, 11.5, 12.4 and 13.0, which means the panel's lowest number on him is still nearly five points above the market's. Only one other name has all four seats that far from the price, and it is Carr, in the opposite direction.
Every Seat's Number
The revised-round board by seat, so you can see where the four disagreed rather than just the median. Each seat's column is its own board scaled to sum to exactly 100 (a seat's raw numbers can land a few tenths over or under), so a value here can sit a tenth away from the raw figure in the ledger; the medians and gaps are computed from these scaled numbers.
| Player | Kalshi | Fable | Gemini | GLM | GPT | Median |
|---|---|---|---|---|---|---|
| Julian Sayin | 6.5¢ | 11.0 | 12.4 | 13.0 | 11.5 | 12.0 |
| Dante Moore | 8¢ | 9.0 | 10.9 | 11.0 | 9.5 | 10.2 |
| Arch Manning | 9.5¢ | 9.0 | 9.0 | 10.0 | 8.5 | 9.0 |
| Darian Mensah | 8¢ | 7.0 | 7.0 | 7.0 | 7.0 | 7.0 |
| Gunner Stockton | 2.5¢ | 5.0 | 7.0 | 6.0 | 5.5 | 5.8 |
| CJ Carr | 11.5¢ | 5.5 | 5.5 | 6.0 | 5.5 | 5.5 |
| Trinidad Chambliss | 7.5¢ | 5.5 | 5.0 | 4.0 | 5.0 | 5.0 |
| Josh Hoover | 4.5¢ | 4.0 | 4.5 | 4.0 | 4.0 | 4.0 |
| Marcel Reed | 3.5¢ | 3.5 | 3.5 | 2.0 | 3.8 | 3.5 |
| Keelon Russell | 3.5¢ | 2.5 | 3.5 | 0.5 | 3.0 | 2.7 |
| Sam Leavitt | 3.5¢ | 3.0 | 2.5 | 0.5 | 3.2 | 2.7 |
| Jayden Maiava | 3.5¢ | 2.0 | 2.5 | 0.5 | 3.0 | 2.2 |
| Jeremiah Smith | 6.5¢ | 3.0 | 1.5 | 1.5 | 2.8 | 2.1 |
| John Mateer | 3.5¢ | 3.0 | 2.5 | 0.5 | 1.8 | 2.1 |
| Bear Bachmeier | 1.5¢ | 2.5 | 1.5 | 1.5 | 2.5 | 2.0 |
| Demond Williams Jr. | 1.5¢ | 1.8 | 1.0 | 0.5 | 2.2 | 1.4 |
| Bryce Underwood | 1.5¢ | 2.0 | 1.5 | 0.5 | 1.2 | 1.3 |
| Faizon Brandon | 1.5¢ | 1.5 | 1.5 | 0.5 | 1.1 | 1.3 |
| Will Hammond | 1.5¢ | 1.0 | 1.5 | 0.5 | 1.5 | 1.2 |
| Devon Dampier | 0.5¢ | 1.2 | 1.0 | 0.5 | 2.4 | 1.1 |
| Rocco Becht | 1.5¢ | 1.2 | 1.0 | 0.5 | 1.4 | 1.1 |
| Kevin Jennings | 1.5¢ | 1.2 | 1.0 | 0.5 | 1.2 | 1.1 |
| Kewan Lacy | 0.5¢ | 0.8 | 0.5 | 0.5 | 0.8 | 0.6 |
| Byrum Brown | 1.5¢ | 0.7 | 0.5 | 0.5 | 1.0 | 0.6 |
| Bo Jackson | 0.5¢ | 0.7 | 0.1 | 0.5 | 0.6 | 0.6 |
| Malachi Toney | 1.5¢ | 0.6 | 0.1 | 0.5 | 0.5 | 0.5 |
| LaNorris Sellers | 2.5¢ | 0.6 | 0.1 | 0.5 | 0.4 | 0.5 |
| Jadan Baugh | 0.5¢ | 0.5 | 0.1 | 0.5 | 0.3 | 0.4 |
| Jaron-Keawe Sagapolutele | 1.5¢ | 0.4 | 0.1 | 0.5 | 0.3 | 0.3 |
| Drew Mestemaker | 1.5¢ | 0.4 | 0.1 | 0.5 | 0.2 | 0.3 |
| Kamario Taylor | 0.5¢ | 0.3 | 0.1 | 0.5 | 0.1 | 0.2 |
| Colin Simmons | 0.5¢ | 0.3 | 0.1 | 0.5 | 0.1 | 0.2 |
| CJ Bailey | 0.5¢ | 0.2 | 0.1 | 0.5 | 0.1 | 0.1 |
| Ahmad Hardy | 0.5¢ | 0.2 | 0.0 | 0.5 | 0.1 | 0.1 |
| Austin Mack | 0.5¢ | 0.1 | 0.0 | 0.5 | 0.1 | 0.1 |
| Brendan Sorsby | 0.5¢ | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 |
| The Field | — | 9.0 | 11.0 | 22.0 | 7.9 | 10.0 |
Model estimates, not financial advice.
The most interesting cell is not at the top. Darian Mensah is the only live player all four seats priced identically, 7.0 across the row (Brendan Sorsby's four zeroes are a different kind of agreement), which is as close to a settled opinion as a price-blind panel produces; the market has him at 8¢, so his is the rare line where the crowd and the machines simply agree. The widest disagreement inside the panel is the field itself: GLM keeps 22% for a winner Kalshi has not listed, GPT keeps 7.9%, and the median of 10.0% is what the board carries. Read that spread as the honest uncertainty of scoring a race in which 34 of 36 candidates have not played yet.
One Pattern Before The Stories
Read the gap column across the top ten and one thing explains most of it: the panel is paying for completion percentage, and the market is paying for the stage. The three quarterbacks the panel prices meaningfully above the market are Sayin, Moore and Stockton (Marcel Reed's +0.2 is inside the rounding noise), and they own the three best 2025 completion rates among the top ten's returning starters, 77.0%, 71.8% and 69.7% (Alabama's Keelon Russell completed 73.3% in 2025, but on 143 passing yards as a backup, which is a sample, not a season). The two quarterbacks it prices furthest below the market, Carr and Chambliss, completed 66.6% and 66.1%. Manning, whose 61.4% is the lowest in the top ten, is the exception that proves the rule in the other direction: the market cut him two cents since July, and at 9.5¢ his re-scaled share now lands on the panel's 9.0% exactly, the crowd having priced out the brand premium the July panel objected to. Efficiency is the models' proxy for a season voters will remember; ranked games in national television windows are the crowd's. Both matter, and the schedule column on the card gives the crowd its due: Manning has seven ranked opponents, the most on the board, and Sayin six. But the pattern tells you what kind of evidence will move this board next. A 30-of-38 afternoon from Sayin in a marquee game closes the widest gap on the board from the market's side; a 20-of-35 win for Carr in one does not open it.
Julian Sayin, Ohio State: The Favorite Whose Price Is Where July Left It
The case did not change; the margin did. Sayin completed 77.0% of his passes in 2025 for 3,610 yards, 32 touchdowns and 8 interceptions on a 12-2 Ohio State team, and that team is now the consensus No. 1 in both polls, with 40 first-place AP votes and 38 from the coaches. Fable's read is the panel's: "Sayin leads — 77% completion and 32 TDs returning on the consensus No. 1 with six ranked games is the archetype." Every seat has him first after the revision round; the lowest number on him, Fable's 11.0, is nearly five points above his re-scaled 6.1% market share.
"Sayin is first: No. 1 Ohio State, 77% completion, 32 TDs, and repeated ranked-stage opportunities." — GPT
So why has the price not moved? The board's own numbers suggest an answer, and it lives two rows down in the Ohio State column. Since July the market has bid Jeremiah Smith from 4¢ to 6.5¢, the same price as his quarterback. The crowd is treating the Ohio State candidacy as a coin with two faces and pricing both; the panel treats it as a single ballot and gives it almost entirely to the man throwing the passes. Both views agree Ohio State produces a Heisman finalist. They disagree about which one, and that disagreement is worth 5.8 points on the top line and 4.0 on Smith's. The first data point that matters arrives September 12 at Texas, the game three seats named outright as the one that most reshapes this board.
CJ Carr, Notre Dame: Still The Market's Favorite, Still The Panel's Hardest Fade
The market trimmed Carr a cent, from 12.5¢ to 11.5¢, and the panel raised him nearly a point, from 4.6% to 5.5%. That narrowed the board's widest gap from about 7.5 points to 5.4, and it is still the widest. Nobody disputes the resume: 2,741 yards, 24 touchdowns, 6 interceptions at 66.6%, on a 10-2 team that opens the polls fourth. The panel's objection is comparative, and it hardened with the poll data: on this board Carr's efficiency season is the fourth or fifth best quarterback resume, his program starts behind Ohio State, Oregon and Georgia in the rankings, and his schedule carries three ranked opponents against Sayin's six and Manning's seven. Gemini summarized the whole top tier in one sentence that leaves him out: it "aggressively favor[s] quarterbacks leading AP Top 10 programs, heavily weighing Julian Sayin and Dante Moore, whose elite 2025 completion rates (77.0% and 71.8%) and preseason rankings (No. 1 and No. 2) provide the strongest platforms." Three seats at 5.5 and one at 6.0: the fade is not one model's opinion, and after Mensah's four identical 7.0s it is the panel's tightest agreement in the top ten, with no seat within four points of the market. Notre Dame opens against Wisconsin on a neutral field on Sunday, September 6, the first chance for the crowd's favorite to make the models' number look wrong.
Jeremiah Smith, Ohio State: One Of The Market's Two Biggest Risers Is The Panel's Biggest Cut
Smith is the line where the two sides moved in opposite directions and both moved a lot. The market took him from 4¢ to 6.5¢, tied for the board's largest rise since July; the panel took him from 3.6% to 2.1%, and now prices him 4.0 points under his re-scaled share, the second-widest fade on the board. The resume is real: 87 catches, 1,243 yards and 12 touchdowns as a sophomore, and Ohio State confirmed this week that he plays in Saturday's opener. The models' problem is structural, not personal. The DeVonta Smith template requires a record-setting receiving season on an unbeaten team, and it also requires the receiver to out-vote his own quarterback, who happens to be the panel's favorite. Gemini gives him "1.5% as Ohio State's WR1" and treats every other non-quarterback as a near-zero; Fable calls him "the only live non-QB (DeVonta Smith template), but Sayin splits his ballot." That is the callback to the top of the board: the same Ohio State ballot the crowd is pricing twice, the panel is pricing once. If you hold the Smith side of that argument, you are also holding the Sayin fade, whether you meant to or not.
Gunner Stockton, Georgia: The Market Walked Away From The Panel's Sleeper
Stockton is the one top-ten name where the July gap opened from the market's side. His price fell from 3.5¢ to 2.5¢ while the panel raised him from 5.0% to 5.8%, and the result is a 3.4-point gap that trails only Sayin's among the names the panel likes more than the crowd does. The profile is the one the pattern section predicts: a fifth-year starter who completed 69.7% of his passes with a 24-to-5 touchdown line, ran for 462 yards and 10 scores, and starts at quarterback for the AP's No. 3 team. Gemini's 7.0 is the panel's high number, and the revision round is where the median was made: one seat entered it treating Stockton as a rounding error and left it with him tied for fifth, the largest single revision on any named player, detailed in the section on where the panel changed its mind. Georgia opens against Tennessee State on Saturday and Western Kentucky the week after, which is the market's honest counterargument: a quarterback on a run-heavy team with no early stage has to win voters on volume he has not shown yet.
Dante Moore, Oregon: The Market Came To The Panel And The Panel Kept Going
Moore was July's cleanest positive gap, 8.0% against 5.5¢, and the market closed most of it, bidding him to 8¢. The panel then re-scored him at 10.2%, so the gap is 2.7 points and he is the panel's clear second choice. The case is Sayin's case one tier down: 71.8% completion for 3,565 yards and 30 touchdowns on a 13-2 team that opens the polls second, with four ranked opponents including Ohio State and USC. Fable moved him from 6.5 to 9.0 in the revision round, the seat's largest upward move, and Gemini and GLM both hold him near 11. Oregon opens with Boise State on Saturday and travels to Oklahoma State in Week 2, where Drew Mestemaker, the board's gaudiest 2025 stat line at 4,379 yards, now plays for a program that went 1-11; the panel keeps him at 0.3% for that reason and no other.
Arch Manning And Darian Mensah: Two Prices The Crowd Got Right
Manning is the board's most complete convergence. In July the panel called his 11.5¢ a brand premium and priced him at 8.7%; the market has since cut him to 9.5¢, a re-scaled 9.0%, and the panel's median is 9.0%. Zero gap. The models still note the accuracy: Gemini writes that "his tougher schedule and lower 2025 efficiency (61.4%) introduce more failure paths," and Fable that "his 61.4% completion caps him below the hype, but seven ranked games is the most vote-manufacturing slate on the board." Both sentences are true at once, which is what a fairly priced contract looks like. Mensah is the other. The market shaded him from 9¢ to 8¢, the panel from 7.5% to 7.0%, and as the seat table shows, all four seats landed on exactly 7.0. GPT thinks that is still light: "Mensah is the main candidate I think the consensus may still underweight: 3,973 yards and 34 TDs transferring into No. 7 Miami is a clean Heisman profile if Miami wins early." He gets his first chance Friday night at Stanford, the earliest opener among the panel's top ten.
Trinidad Chambliss And LaNorris Sellers: Where The Crowd Bid Up And The Panel Did Not Follow
Chambliss crossed from fair to a mild fade. His price rose from 6.5¢ to 7.5¢ while the panel trimmed him from 6.0% to 5.0%, a 2.1-point gap. The numbers are strong, 3,937 passing yards with only 3 interceptions plus 527 on the ground for a 13-2 team, and Gemini opened the first round with him at 10.0 before cutting to 5.0 after reading the other seats, who weighed Ole Miss' AP No. 9 ranking and a six-ranked-opponent gauntlet as more risk than stage. Fable went the other way, calling him "the underrated one" (crediting another seat's first-round read) and nudging him up to 5.5. Ole Miss opens against No. 24 Louisville at a neutral site on Sunday, September 6, the earliest test against an AP-ranked opponent on the board (LSU draws Clemson a day earlier, but Clemson is ranked only in the coaches' poll). Sellers is the sharper version of the same disagreement: the market bid South Carolina's quarterback from 1.5¢ to 2.5¢, the panel cut him from 0.8% to 0.5%, and the card explains why: a 4-8 program unranked in both polls, a 2025 line of 13 touchdowns and 8 interceptions at 60.8%, and two of his receivers already out for the season.
The Seven New Listings, And The Three Contracts The Status Flags Killed
Kalshi added seven contracts since July, and the panel's read on them is compact. Bear Bachmeier, BYU's sophomore who ran for 11 touchdowns on a 12-2 team, is the only new listing the panel puts above 1.5%, at 2.0% against a 1.5¢ price. Faizon Brandon, the true freshman Tennessee named its starter on August 24, gets 1.3% at 1.5¢, and Gemini's caution covers him: it is "lower on unproven freshmen like Faizon Brandon, as SEC volume is notoriously hard to scale immediately." Devon Dampier, Utah's dual-threat senior who kicks off Thursday night against Idaho, gets 1.1% at half a cent, a small positive gap. Florida's Jadan Baugh, Mississippi State's Kamario Taylor and Texas edge rusher Colin Simmons all price under half a point, and CJ Bailey, whose NC State opener produced 84 passing yards, an interception and a lost fumble in a 34-8 loss, gets 0.1%.
Three older contracts matter more than any new one, because the card now tells the panel what the July card could not. Brendan Sorsby, whom July's panel carried at 1.9% against a 0.5¢ price, is not playing in 2026: the NCAA declared him permanently ineligible this spring over sports wagers, and although a Texas judge granted an injunction, his own court filing states that Texas Tech informed him on June 15 it would not permit him to play this season, after which he dropped the suit. He is still listed on the exchange under Texas Tech at 0.5¢, and every seat now prices him at 0.0. The contract has no delisting clause, so it simply resolves NO unless he wins, and the panel's number says what the rules cannot. Ahmad Hardy, the board's leading 2025 rusher at 1,649 yards, is out of Thursday night's Missouri opener with no return timetable after suffering a gunshot wound to the upper leg in May; the panel prices him at 0.1%. Austin Mack, who lost Alabama's quarterback competition to Keelon Russell on August 22, went from 1.1% in July to 0.1% now, while Russell, the new starter, rose from 1.8% to 2.7%. The market moved Russell too, from 2.5¢ to 3.5¢, and left Mack at half a cent: on the Alabama room, at least, the crowd read the same August 22 headline the models did.
Where The Panel Changed Its Mind
The revision round is where this board took its shape, and four arcs stand out (numbers here are each seat's board scaled to 100, as in the table above). GLM's Stockton move, 0.4 to 6.0, is the biggest single revision on any named player (only Gemini's field number, 4.5 to 11.1, travelled further), and the same seat raised Sayin from 9.1 to 13.0 and cut Manning from 12.1 to 10.0 after entering the round as the only seat with Manning first. Gemini made the opposite kind of concession, halving Chambliss from 10.0 to 5.0 and cutting Maiava from 5.0 to 2.5, and it raised its field number from 4.5 to 11.1 after reading the others, conceding that a race with 34 unplayed candidates deserves a bigger unlisted bucket. Fable moved the other way on the field, from 14.0 to 9.0, and explained why it stopped there: "I resisted B's 4.5% field: Mendoza won from off preseason boards last year, and Clemson, Louisville, Houston and every unlisted breakout live there, so 9% is my floor." GPT's revision was the quietest and the most specific, raising Mensah from 5.2 to 7.0 and Manning from 7.0 to 8.5 while trimming Stockton from 7.5 to 5.5, so the seat that opened with the board's most Georgia-friendly number ended near the median. Consensus formed on the top three in every seat's revised board, Sayin first, Moore second, Manning third (Fable has Moore and Manning level at 9.0), and the seats kept disagreeing exactly where the evidence is thinnest: the field, the freshmen, and how much a run-heavy team's quarterback can win on volume.
What Moves This Board Next
Three of the four seats named the same game by name, and the fourth pointed at Manning's early ranked games without naming one. Ohio State at Texas on Saturday, September 12, stacks the panel's first and third names against each other and puts Jeremiah Smith, tied for the market's biggest rise, on the same field, so one afternoon reprices the top of the board from every angle at once. Before that, the opening weekend carries the early tests in order: Mensah at Stanford on Friday night, Sayin and Smith against Ball State and Manning against Texas State on Saturday, Sam Leavitt's LSU debut against Clemson on Saturday night, Chambliss against Louisville and Carr against Wisconsin on Sunday, and Kevin Jennings' SMU at Florida State on Monday. Two more Week 2 games matter to the gap column: Oklahoma at Michigan, which puts Mateer and Bryce Underwood on the same field, and Oregon at Oklahoma State, Moore's first road test. This page re-scores after each batch of results lands on the card.
How To Read These Prices
Work one example. Carr's 11.5¢ is not a clean "11.5% chance": the 36 contracts are quoted in 36 independent order books, and their midpoints sum to about 106¢ rather than 100, so his re-scaled share is closer to 10.8%. The panel's 5.5% against that 10.8% is the real size of the fade. Three honest frictions apply to everything above. First, eligibility: Kalshi is a CFTC-regulated event-contract exchange, 18+ and state-dependent, and whether its sports contracts count as sports bets is an open legal question: an August 28 Ninth Circuit ruling held that Nevada can regulate them as sports betting, which splits with the Third Circuit's April ruling on New Jersey, so defer to the platform's own eligibility screen for your state. Second, liquidity: most of this board trades under 3¢ on thin books with a full cent between bid and ask, so the longshot "prices" are soft, and three of them, Sorsby, Hardy and Mack, are 0-bid, 1¢-ask books on players the status flags have already ruled out, which is 1.5¢ of the 106¢ sum sitting on contracts nobody is buying, and converting cents to odds matters more than usual before comparing anything to a sportsbook number. Third, time: this market does not settle until the award is presented on December 12, so a position holds capital for three months while the race gets decided by the schedules in the section above. Award futures are Kalshi's lane; for the Saturday games that decide this board, our college football live odds screen shops the number across every major book, the NFL live odds screen does the same for Sunday, and the free expert picks today page does the price-context work on the games themselves.
More on this: Darian Mensah Heisman Odds: A Favorite Built On 30 Throws · Malachi Toney Heisman Odds: One Record Night, 800K Contracts · AFC East Odds: Drake Maye Made The Super Bowl, Buffalo's Favored · Kansas Senate Race 2026 Odds: The Bet That Moved Nothing · GTA 6 Release Date: Traders Are Betting November Holds
The Bottom Line
The same price-blind protocol runs on the 2026 Emmys board, where the market's near-lock favorite gets the same skeptical treatment CJ Carr gets here.
Five weeks ago this page said the market had the wrong favorite, and five weeks of price movement have mostly agreed with it: Carr trimmed, Manning priced down to the panel's number, Moore bid up to it, Mateer and Hoover walked toward it. The crowd and the machines now disagree about the top of the Heisman Trophy odds board in one specific way. The market has spread the Ohio State candidacy across two contracts at 6.5¢ apiece; the panel has concentrated it in one, and that is why Julian Sayin, the only top-ten price that did not move since July, is the widest positive gap on the board while his own receiver is the second-widest negative one. The same panel scores the team version of this question on the college football championship board and the award's NFL twin in the NFL MVP model verdict. The number to carry into Saturday is still the one with no contract attached: a 10% field means the panel thinks one December in ten ends with a winner Kalshi has not listed, and after Mendoza, nobody on the panel treats that as a rounding error.
Verdicts generated September 3, 2026, price-blind, with a revision round. Prices as of September 3, 2026, 6:00 PM ET. Panel this cycle: Claude Fable, GPT, Gemini and GLM (four of the panel's eight seats). Model estimates with a public grading record, not financial advice.
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.
