NFL Coach Hot Seat: Which Coaches Could Be Out Before The 2026 Season?
Kalshi turned the NFL's coaching carousel into 22 separate contracts, one for each coach it lists, each asking the same blunt question: will this man be out of his job before September 1, 2026, nine days before the Week 1 openers? We asked six AI models to score every one of them without letting a single one see the market prices. The answer they came back with is close to unanimous, and it lines up with what the market is already telling you at a glance: a wall of one- and two-cent prices. But the models did find one name worth stopping on, and one piece of market behavior worth explaining.
The Quick Answer
The panel says almost nobody leaves before the season. Of the 22 coaches on the board, 18 sit at the same 0.3% base-rate floor for a preseason departure, which is close to zero. Four drew a hair above it, but only Kyle Shanahan, at a 3.8% blend, is materially off the floor, and the reason is a July 14 car crash that left him with a severe concussion, not anything about his job security. Add up all 22 numbers and they come to about 0.1 expected departures, which works out to roughly a 1-in-10 chance that any of these coaches is gone before the season, with Shanahan accounting for more than a third of it. The full model-by-model board, the Shanahan case, and the one place the market and the models part ways are all below.
Get the grade when this settles. This board resolves on September 1, and we score every Model Verdict call in public once the market settles. Grab the re-scored board and our next verdict drop from the email box at the bottom of this page so you catch the settlement receipt.
New to Kalshi's event markets? Start with our step-by-step guide to how NFL event markets work on Kalshi before you read the board.
The Market
| Venue | Kalshi, a CFTC-regulated event-contract exchange (18+; availability varies by state, as of August 2026) |
| The Event | "Which Pro Football coaches will be out before Sep 1, 2026?" (ticker KXCOACHOUTNFL), 22 separate yes/no contracts, one per head coach |
| The Contract | Each market resolves YES if that team's head coach leaves, or announces he will leave, before September 1, 2026, per Kalshi's settlement rules. A firing during the regular season settles NO here. |
| Settles | September 1, 2026, nine days before the Week 1 openers on September 10 (markets close early if a coach departs first) |
| Prices Below As Of | the morning of August 12, 2026 (the panel itself ran August 3) |
These are thin books. Most contracts show a one- or two-cent YES ask with no bid underneath, so treat every price as a soft read of where a few traders sit, not a deep market. Prices on this page refresh during the news cycle; the figures here are stamped as of the morning of August 12, 2026.
The Setup: Why Coaches Almost Never Leave In August
The whole board rests on one fact about how the NFL works: the coaching-change calendar is tied to the end of the season, not the start of it. Firings cluster in-season and on "Black Monday," the day after Week 18. The 2026 hiring cycle was a busy one, with Pete Carroll's one-and-done exit in Las Vegas among more than half a dozen openings, and every last one of them resolved back in January and February. None happened in the spring or summer.
Search the modern record for a head coach fired for performance between the draft and Week 1 and you come up empty. The only mechanism that has ever pulled a coach out in the offseason is a sudden shock to his health or a scandal, and even those tend to land right after a season rather than in training camp. Gary Kubiak stepped down in Denver for health reasons, a complex-migraine condition after a prior mini-stroke, but that was January 2017, immediately post-season. John Fox needed aortic-valve surgery in November 2013 and returned to the sideline. The closest thing to a summer coaching casualty this year was the Commanders releasing a tight ends coach after a June DWI arrest, an assistant, not a head coach.
Put a number on it and a head coach leaving in the March-to-September window runs somewhere under 1% per coach in a given year, on the order of one league-wide event per decade. So for every coach on this board, "he stays" is the heavy structural favorite. Nothing should lift a coach off that floor short of a concrete, active health or legal situation. Exactly one coach has one.
The Board
Sorted by the AI blend. The market column is the YES ask as of August 12; an asterisk flags a thin book where the last trade printed higher than the current ask.
| Coach | Team | Market (YES) | AI Blend |
|---|---|---|---|
| Kyle Shanahan | San Francisco | 2¢* | 3.8% |
| Jim Harbaugh | LA Chargers | 1¢ | 0.5% |
| Sean McVay | LA Rams | 1¢* | 0.4% |
| Dan Quinn | Washington | 2¢ | 0.4% |
| Andy Reid | Kansas City | 1¢ | 0.3% |
| Brian Schottenheimer | Dallas | 1¢ | 0.3% |
| Nick Sirianni | Philadelphia | 3¢ | 0.3% |
| Dan Campbell | Detroit | 1¢ | 0.3% |
| Matt LaFleur | Green Bay | 1¢ | 0.3% |
| Kevin O'Connell | Minnesota | 1¢ | 0.3% |
| Dave Canales | Carolina | 1¢ | 0.3% |
| Kellen Moore | New Orleans | 1¢ | 0.3% |
| Todd Bowles | Tampa Bay | 2¢ | 0.3% |
| Mike Macdonald | Seattle | 1¢ | 0.3% |
| Mike Vrabel | New England | 3¢ | 0.3% |
| Aaron Glenn | NY Jets | 2¢ | 0.3% |
| Zac Taylor | Cincinnati | 2¢ | 0.3% |
| DeMeco Ryans | Houston | 1¢ | 0.3% |
| Shane Steichen | Indianapolis | 2¢ | 0.3% |
| Liam Coen | Jacksonville | 1¢ | 0.3% |
| Sean Payton | Denver | 1¢ | 0.3% |
| Ben Johnson | Chicago | 1¢ | 0.3% |
Every Seat's Number
The full price-blind panel, six models across four AI labs.
| Coach | Market | Fable | Opus | Sonnet | GLM | Kimi | DeepSeek | Blend |
|---|---|---|---|---|---|---|---|---|
| Kyle Shanahan | 2¢ | 4.0 | 4.5 | 4.0 | 4.0 | 4.5 | 2.0 | 3.8% |
| Jim Harbaugh | 1¢ | 0.5 | 0.6 | 0.5 | 0.4 | 0.7 | 0.3 | 0.5% |
| Sean McVay | 1¢ | 0.4 | 0.5 | 0.4 | 0.3 | 0.5 | 0.3 | 0.4% |
| Dan Quinn | 2¢ | 0.4 | 0.4 | 0.4 | 0.3 | 0.4 | 0.3 | 0.4% |
| Andy Reid | 1¢ | 0.3 | 0.4 | 0.3 | 0.3 | 0.3 | 0.3 | 0.3% |
| Brian Schottenheimer | 1¢ | 0.3 | 0.3 | 0.3 | 0.3 | 0.3 | 0.3 | 0.3% |
| Nick Sirianni | 3¢ | 0.3 | 0.3 | 0.3 | 0.3 | 0.3 | 0.3 | 0.3% |
| Dan Campbell | 1¢ | 0.3 | 0.3 | 0.3 | 0.3 | 0.3 | 0.3 | 0.3% |
| Matt LaFleur | 1¢ | 0.3 | 0.3 | 0.3 | 0.3 | 0.3 | 0.3 | 0.3% |
| Kevin O'Connell | 1¢ | 0.3 | 0.3 | 0.3 | 0.3 | 0.3 | 0.3 | 0.3% |
| Dave Canales | 1¢ | 0.3 | 0.3 | 0.3 | 0.3 | 0.3 | 0.3 | 0.3% |
| Kellen Moore | 1¢ | 0.3 | 0.3 | 0.3 | 0.3 | 0.3 | 0.3 | 0.3% |
| Todd Bowles | 2¢ | 0.3 | 0.3 | 0.3 | 0.3 | 0.3 | 0.3 | 0.3% |
| Mike Macdonald | 1¢ | 0.3 | 0.3 | 0.3 | 0.3 | 0.3 | 0.3 | 0.3% |
| Mike Vrabel | 3¢ | 0.3 | 0.3 | 0.3 | 0.3 | 0.3 | 0.3 | 0.3% |
| Aaron Glenn | 2¢ | 0.3 | 0.3 | 0.3 | 0.3 | 0.3 | 0.3 | 0.3% |
| Zac Taylor | 2¢ | 0.3 | 0.3 | 0.3 | 0.3 | 0.3 | 0.3 | 0.3% |
| DeMeco Ryans | 1¢ | 0.3 | 0.3 | 0.3 | 0.3 | 0.3 | 0.3 | 0.3% |
| Shane Steichen | 2¢ | 0.3 | 0.3 | 0.3 | 0.3 | 0.3 | 0.3 | 0.3% |
| Liam Coen | 1¢ | 0.3 | 0.3 | 0.3 | 0.3 | 0.3 | 0.3 | 0.3% |
| Sean Payton | 1¢ | 0.3 | 0.3 | 0.3 | 0.3 | 0.3 | 0.3 | 0.3% |
| Ben Johnson | 1¢ | 0.3 | 0.3 | 0.3 | 0.3 | 0.3 | 0.3 | 0.3% |
Model estimates generated August 3, 2026, price-blind, from a six-model panel after a revision round where each model read the others' reasoning. These are model estimates, not predictions of fact and not financial or trading advice. Models are frequently wrong; the market price reflects real traders' money. A note on the panel: six of the eight seats in our standard Model Verdict lineup ran on this board.
Kyle Shanahan, San Francisco: The One Real Risk
Shanahan is the only coach on the board the panel lifted meaningfully off the floor, and the reason has nothing to do with the 49ers' win total. On July 14, 2026, he was in a car crash in Palo Alto that left him with a broken nose, broken ribs, a broken left hand, stitches above his right eye, and a severe concussion, with a brief hospital stay. He was limited at training camp with lingering concussion symptoms, and Chris Foerster took over day-to-day duties while he recovered.
That is why every model put him above the field, in near-lockstep. Opus called it "a health tail, though he's still favored to return." DeepSeek, the most conservative seat, held lowest at 2% and stated its whole case in one line:
Only Kyle Shanahan carries real risk (2%) due to a July car crash and lingering concussion symptoms. All other coaches lack any concrete pre-Sep-1 departure signal, so they sit at the ~0.3% base rate. — DeepSeek
The honest read cuts both ways. General manager John Lynch has been plain about what is actually keeping him sidelined ("broke some ribs, that's painful, but that's not what's keeping him away. It's the concussion"), the team expects him back for the September 10 opener, and there is no talk of a firing or a resignation. The only event that pays this market is an actual departure, so what the models are pricing is the small chance that a serious head injury forces Shanahan to step away before September 1, not a firing and not a likely outcome. Even at the high end of the panel that chance sits under 5%.
How To Read Shanahan's Contract
Here is the one contract on the board worth walking through, because it is the only one where a real price and the panel meet. Shanahan's book, as of August 12, shows a 2-cent YES ask over a 1-cent bid, with the last trade printed at 3 cents. On a market that settles at either 100 cents (he leaves before September 1) or 0 cents (he stays), that cluster of prices means the market puts his departure somewhere around 2-3%. The panel's blend, generated price-blind on August 3, landed at 3.8%. Those two reads, one from real traders' money and one from six models that never saw a price, sit within a cent or two of each other, and the small gap runs in the direction you would expect as the July 14 crash recedes and Shanahan keeps progressing. The agreement is the signal here: both the crowd and the machines read the concussion as a small, real tail and nothing more.
Two honesty notes before anyone reads too much into a couple of cents. First, this is a thin book: stray prints ran as high as 19 cents in late July and 8 cents on August 2 before settling into 3-and-4-cent trades, so the "price" is a handful of orders, not a deep consensus, and the spread you would pay is real. Second, the contract closes early if Shanahan departs, so a YES holder is paid the moment an exit is announced rather than waiting for September 1. On the 14 coaches sitting at a 1-cent ask, the same math turns that price into a ~1% implied read, which is simply the market agreeing with the models that nothing is there.
The One Place The Market And The Models Split: Mike Vrabel
Look at the board again and you will notice the busiest market on it is not Shanahan's. It is Mike Vrabel's. His contract carries by far the most open interest and trading volume of any coach here, many times the rest of the field, and the market holds his YES at a 3¢ ask over a 2¢ bid, the strongest bid on the board, while Aaron Glenn's sits at 2¢. Most coaches without a story are still parked at a 1-cent ask, and the stray higher asks elsewhere are quotes rather than conviction: Nick Sirianni's 3¢, the tallest of them, sits over no bid at all with a last trade at a penny.
The models do not follow the money here. All six put Vrabel at the 0.3% base rate, and Glenn right beside him. There is no departure storyline for either one. Vrabel is a second-year coach in New England holding routine late-July camp pressers about roster items. Glenn is a second-year coach in New York whose general manager publicly backed him going into the season, citing "a better comfort level going into Year 2." The panel's read is that the extra pennies on Vrabel and Glenn are name recognition and retail volume rather than any signal, so it left both at the floor.
The one useful disagreement on an otherwise quiet board: where real traders' attention has pooled, the models see nothing that should move the number off the floor.
Dark Horses The Panel Won't Dismiss
Below Shanahan, three coaches drew a whisker above the base rate, each for a stated reason, and none higher than 0.5%.
- Jim Harbaugh, Chargers (0.5%). He had a heavy offseason medically, a cardiac ablation after an in-game atrial-flutter episode in October 2025, plus a hip replacement. The models nudged him up for the health history, then trimmed him back in the revision round because the record is clear he is cleared and running the offseason, with an "A-grade" from his doctors. A resolved risk, not an active one.
- Sean McVay, Rams (0.4%). There is real chatter about McVay eventually walking away, but every report frames it as a post-2026-season decision tied to an aging roster core, not a summer exit. Sonnet flagged exactly that: the talk is "explicitly post-2026-season, not pre-Sep-1."
- Dan Quinn, Washington (0.4%). The one coach here with staff churn around him, with Washington having shuffled its offensive coordinator group this offseason and the panel citing front-office turmoil. Opus gave him the nod for it. Even so, hot-seat friction is a January risk, not an August one.
The Convergence Round
After the first pass, each model saw the others' anonymized numbers and reasoning and got the chance to revise. On a board this lopsided there was not much to move, but the panel tightened rather than scattered. Opus pulled Shanahan from 5.0 down to 4.5, Fable moved the other way from 3.5 up to 4.0 after peers stressed the delegated camp duties, and Kimi trimmed Harbaugh from 1.0 to 0.7 once it re-read that his health issues are resolved. DeepSeek held firm and said exactly why:
Held firm; no new evidence in peers' reasoning—only Shanahan has a concrete health signal, and 2% adequately reflects the tail risk of an unlikely pre-Sep-1 health-related departure given the team's expectation of his return. — DeepSeek
The result is a panel that agrees on the shape of the board more tightly after talking than before: one name near 4%, everyone else at the structural floor.
What To Take From This Board
This is the rare Model Verdict where the AI and the market mostly shake hands. The wall of one- and two-cent prices is correct; preseason coaching departures essentially do not happen outside a health or legal shock. Between now and settlement, the one input that can actually move this board is the 49ers' practice reports: if Shanahan is still limited in the last week of August, his tail fattens, and the moment he takes back full control of camp, the last live contract on the board effectively dies. The bigger question most fans mean when they say "hot seat," which coach gets fired first once games start, is a different market that lives in the season and settles in January; this board is only about who is gone before Week 1.
Every number in this piece gets graded in public once the board settles on September 1. You can see how the lane's past calls have held up on the full graded scoreboard. To be explicit: these are model estimates for information, not financial advice and not a recommendation to trade any contract, and the models are wrong on a regular basis. Kalshi's event contracts are CFTC-regulated and are for adults 18 and older where legal.
For more of the panel's NFL work, see our verdict on the 2026 NFL win totals, our Kalshi NFL Week 1 board, and the panel's 2027 Pro Football champion field board. And if you would rather act on the games than the sideline, our free expert picks, NFL included, are updated all season, wherever you are.
FAQ
Can an NFL head coach really be fired before the season starts?
It is close to unheard of. Coaching changes are tied to the end of the season, and the modern record has no head coach fired for performance between the draft and Week 1. Every offseason exit on record came from a sudden health event or a scandal, and even those are rare. That is why almost every contract on this board trades at a penny or two.
Why is Kyle Shanahan priced higher than everyone else?
Because of a July 14, 2026 car crash that left him with a severe concussion and had him limited at training camp while an assistant ran day-to-day duties. The 49ers expect him back for the September 10 opener, so the elevated number reflects the small chance a serious injury forces him to step away before September 1, not any risk of a performance firing.
When does this board settle?
September 1, 2026. Each individual coach's market can also close early if that coach departs before then. We publish a graded scorecard of every call once it resolves.
Hottest Prediction Markets Right Now
- 2028 Democratic presidential nominee · $166M traded
- Rocket Classic Winner · $124M traded
- 2027 Pro Football Champion · $51M traded
- Pro Baseball Champion · $40M traded
- Bitcoin price at the end of 2026 · $30M traded
- 2027 Pro Basketball Champion · $19M 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.
