The Quick Answer
Byron Donalds is the favorite to be Florida's next governor, and the market just took a step toward David Jolly. Kalshi prices the Republican side at 79.5 cents against 21.5 for Jolly's Democratic side, as of early afternoon on September 19. Our eight-model AI panel, run price-blind on primary night, reads the race at 78.5% Republican; the gap between market and model is down to a point. The first nonpartisan post-primary poll published on the evening of September 17, St. Pete Polls for Florida Politics, and it has Donalds 43.5, Jolly 43.4. Ten hours before it published, a third quarter-million-contract sell order hit the Republican book. What the poll does to this page, and why the order came first, is below.
Stay ahead of the markets.
Daily insights and expert picks on Kalshi, Polymarket, and what's moving markets.
Free forever. Unsubscribe anytime.
The Board: Market Vs. Model
| Outcome | Nominee | Kalshi | AI Blend (after revision) |
|---|---|---|---|
| Republican | Byron Donalds | 79.5¢ | 78.5% |
| Democratic | David Jolly | 21.5¢ | 20.9% |
| Neither Party | independent or minor-party win | not listed | 1.0% |
The contracts settle on party, not on a person, so nothing on this board is a bet on Donalds or Jolly individually. A win by any independent or minor-party candidate on Florida's November ballot would resolve both contracts NO, a path the panel prices at about 1% and the market does not quote at all.
The Republican side traded at 81.5 cents in primary week, closed August 29 on an 83-84 quote, and opened August 30 at 84 and printed 85 before the first of three large sell orders took it to a 79 low and an 81 close. It closed between 80 and 82 every session from September 3 through September 16, then closed at 78 on September 17, its lowest close in the daily record we keep on this ticker, which starts August 26, and 79 on September 18. It sits at 79.5 this afternoon, the midpoint of a 79-80 quote, a point below our September 11 fetch and a point above the panel's frozen number. Jolly's side was quoted at 20.5 on September 11, printed as high as 25 on September 12 on a few hundred contracts, has closed between 20 and 22 every session since, and is 21.5 now, the midpoint of 21-22. Since the September 11 fetch: Republican 80.5 to 79.5, Democratic 20.5 to 21.5. Our own graded record says a gap this size is not an edge.
About 1.68 million contracts (face value) have matched across the pair, up from about 1.12 million on September 11, and the spread is one cent on both sides. Roughly 266,000 of the new contracts came in one order on September 17, described in the next section. Polymarket runs the same question on about $84,000 of volume, 79.5 cents Republican, level with Kalshi.
More live boards from the same panel: The 2026 Senate hub, where Florida's Senate special (Moody) sits at 89.5¢ Republican · Georgia governor · Democratic side (Bottoms) 54.5¢ · Michigan governor · Democratic side (Benson) 90.4¢ last trade. Prices fetched September 19, 2026.
The panel's verdicts were generated the evening of August 18, 2026, price-blind and frozen since. Prices on this page were last fetched from the exchanges on September 19, 2026, about 1:35 p.m. ET.
What Moved Since September 11
Three things, in an order that matters. The order came first, on the morning of Thursday, September 17. The poll came that evening. The finance reports came Saturday, September 19.
At 9:44:11 a.m. ET on September 17, 32 fills in a single microsecond put 266,058 contracts through the Republican book, sweeping the bids from 81 down to 78 and finishing with two blocks of 125,000 at 78. It is the third quarter-million-contract sell on this ticker in 18 days, after 278,196 on August 30 and 253,654 on September 10, and the first the book did not absorb: the earlier two closed their sessions back at 81 and 82, this one closed at 78. Open interest rose by about 271,000 that session, so the order opened new positions rather than closing old ones. The book has rebuilt a level lower. The wall that bid 80 on September 3 and was gone by September 11 now bids 78, about 266,000 contracts deep, and the roughly 207,000 contracts offered at 82 on September 11 have grown to about 254,000. The Republican side traded 286,000 contracts on September 17 against fewer than 22,000 on every other day since September 11. Nothing public that morning explains it: St. Pete Polls was still in the field, and the day's campaign news was both candidates working Orlando on property taxes and data centers.
The poll is the one this page said would re-score it. St. Pete Polls, surveying for Florida Politics, reached 913 likely general-election voters by automated call and text from September 15 to 17 and found Donalds 43.5, Jolly 43.4 before rounding, with 4% for someone else and 9% undecided, a margin of error of 3.2 points, published at 7:48 p.m. ET on September 17. It is the first post-primary head-to-head from a pollster with no party sponsor, and it lands inside the margin, where the sponsored Change Research poll said the race was. The detail underneath is the one the panel argued about in August: Jolly leads 48 to 36 among voters outside the two major parties and takes 13% of Republicans to Donalds' 8% of Democrats, while Donalds holds 74% of his own party against Jolly's 81% of his, the seven-point consolidation gap a 48% primary predicted and the one the Claude Fable seat shaved a point for on August 18. The market's answer through Saturday is the Democratic side quoted 21 to 22 against 19 to 20 the two sessions before it, on one-directional buying: about 62,000 of the 66,000 Democratic contracts traded on September 17 and 18 were buyers lifting the offer. By the rule this page set on September 11, a nonpartisan read inside three points pushes the panel's number down several points. The panel has not been re-run for this edition; the seats below are the August 18 verdicts, frozen, and the re-score is now the pending item on this page rather than a hypothetical.
The finance reports, filed with the Division of Elections and reported by Florida Politics on September 19, say the money has not followed the poll. Jolly raised more than $590,000 from September 5 to 11 and held nearly $3.3 million in his campaign account plus a little more than $3 million in his Florida 2026 committee. Donalds raised nearly $3.2 million across his campaign account and his Friends of Byron Donalds committee over the same dates, including $1.5 million from the Seminole Tribe of Florida on September 9, and retains nearly $44 million of the almost $120 million the two accounts have raised. Jolly has started spending, with a multi-million-dollar statewide ad buy this week. The handicappers' current reads, per the same report: Inside Elections rates the race Lean Republican in a note dated September 17, whose timing against the 7:48 p.m. poll we cannot tell, while Cook Political Report, at Likely Republican since August 27, and Sabato's Crystal Ball hold there.
Neither Jay Collins nor James Fishback has endorsed, on anything we could find through September 19; the silence is now a month old. Debates are the next test and there is no date for one: on September 18 Donalds' campaign said it had accepted three, with Telemundo 51 and NBC 6, Spectrum News, and News4Jax, while Jolly's camp said he had accepted two others, Tampa Bay Tiger Bay and CBS Miami, and accused Donalds of ducking those. At its August 30 open the market sat five and a half points above the panel, on September 11 two, and this afternoon one, and the largest order of the month printed ten hours before the poll that would have explained it. The board above is the market's number; what follows is the panel's, dated August 18.
Donalds Vs. Jolly: What The Record Says
Donalds has represented Florida's 19th district in Congress since 2021 and has carried President Trump's endorsement since February 2025, per the race's public record. He raised $13.79 million through August 13 in his campaign account alone, per Florida Division of Elections filings, roughly seven and a half times what Collins raised, and he has never run statewide. The number the panel kept returning to is his 48% primary share: normal in an eleven-candidate field, but it means he starts the general needing to consolidate the more than 880,000 Republicans who voted for someone else.
He also arrives with an attack already answered on the record. In the primary's closing weeks, Fishback publicly alleged that a federal investigation into Rep. Cory Mills had expanded to include Donalds. On August 7, 2026, a Department of Justice spokesperson told Newsweek that "a law enforcement official confirms this is absolutely false"; Newsweek reported the same day that the documents Fishback posted had not been independently authenticated, and quoted Donalds spokesperson Gates McGavick calling the claim a "baseless and defamatory sideshow." Donalds' nomination price never traded below 96.5 cents on either exchange after August 5.
Jolly represented Florida's 13th district as a Republican from 2014 to 2017, published a 2018 NBC opinion piece arguing that "you can't be never-Trump and be a Republican" (Ballotpedia's record of his career), and registered as a Democrat in April 2025. He took 61% of a six-candidate Democratic primary, the cleanest mandate either nominee carries into November. His pitch is that a former Republican can pull disaffected Republicans across; the counterweight is his own file. CBS News Miami reported in August 2025 that Jolly's sharpest criticism was coming from fellow Democrats, chiefly on abortion: as a Republican congressman he said he believed life begins at conception, and he now says he supports abortion rights. The panel weighed that as a base-turnout question rather than a persuasion one.
The head-to-head polling file the panel scored was entirely pre-primary. The public Donalds-versus-Jolly polls compiled on the race's polling page split by sponsor: the nonpartisan surveys show Donalds leads clustered between five and seven points, and the Democratic-sponsored ones had Jolly ahead by one to four. The freshest of those the panel saw was Hart Research, sponsored by the Democratic Governors Association, with Jolly 46, Donalds 45 on August 10 to 13. Every seat on the panel discounted it for sponsorship and none dismissed it, which is the right handling of a partisan poll that agrees with a plausible story. The September 7 to 9 Change Research poll (Jolly 47, Donalds 44) and the September 15 to 17 St. Pete Polls survey (Donalds 43.5, Jolly 43.4) both arrived after the panel froze; they are covered above and have not re-scored any seat.
Underneath the polling, the structure is heavy and one-directional. No Democrat has been elected governor of Florida since 1994 or carried the state for president since 2012, and Republicans hold every statewide office and legislative supermajorities. DeSantis won by 19.4 points in 2022 and Trump carried the state by 13 in 2024.
Against that, the one Democratic near-miss is the most instructive fact on the card. The last open-seat Florida governor race, in 2018, was decided by four tenths of a point, powered by a blue-wave midterm, an open seat and an untested statewide Republican nominee. The first two conditions are present again (the generic ballot averaged Democrats +6.6 across six public aggregators as of August 17, with presidential approval at 37.6%), and the third is arguable, since Donalds has never won statewide. What has changed since 2018 is the baseline itself: Florida's registration advantage flipped Republican, and the presidential margin moved nearly ten points right. That is the whole argument inside the roughly 20 cents both the market and the panel give Jolly: how much of 2018 can repeat in a state that has moved ten points right.
Every Seat's Number
Both rounds shown, Republican-side probability. The headline blend is the median of the revised column. On a phone this table scrolls sideways.
| Seat | R1: Republican | R2: Republican | R2: Democratic | R2: Neither |
|---|---|---|---|---|
| Claude Opus | 88.5 | 85.0 | 14.3 | 0.7 |
| Claude Fable | 83.0 | 82.0 | 17.0 | 1.0 |
| Gemini 3.1 Pro | 82.0 | 80.0 | 19.5 | 0.5 |
| Claude Sonnet | 79.0 | 79.0 | 20.0 | 1.0 |
| DeepSeek V4 | 78.0 | 78.0 | 21.8 | 0.2 |
| ChatGPT (GPT-5.5) | 75.0 | 77.0 | 22.0 | 1.0 |
| GLM 5.2 | 70.0 | 75.0 | 24.0 | 1.0 |
| Kimi K3 | 70.0 | 75.0 | 24.0 | 1.0 |
Every seat on this panel is graded against real market settlements — records to date: GPT 86% on 9,653 graded calls · Claude Fable 89% on 2,856 graded calls · Claude Opus 89% on 2,995 graded calls · Claude Sonnet 88% on 2,971 graded calls · Gemini 87% on 8,676 graded calls · Kimi 87% on 4,322 graded calls · GLM 84% on 3,969 graded calls · DeepSeek 84% on 4,090 graded calls. Recomputed daily; the full scoreboard is public.
Model estimates generated August 18, 2026, price-blind from live fetched data. Models are frequently wrong; the market price reflects real traders' money.
Read down the revised column and the entire disagreement is Claude Opus at 85 against GLM and Kimi at 75, a ten-point argument about how much Florida's rightward baseline protects a nominee who has never run statewide. Every number in this piece landed in our backtest ledger the evening it was generated and gets graded against the certified result on the public model verdict scoreboard.
Where The Panel Changed Its Mind
Round one ran from 70 to 88.5; the revision round closed that spread to ten points, and six of eight seats moved. The top came down on a double-counting argument:
"I accepted the point that I was double-counting structure. My 88.5 implied roughly a ten-point win with tiny variance, but the actual polling median across the head-to-heads is four to six points, and Trump-era midterm polls were nearly unbiased. A five-point median lead with 11 weeks left, an open seat, a Democrats +6.6 generic ballot and 37.6% approval is 'clear favorite,' not 'near-lock.'" (Claude Opus, 88.5 to 85.0)
The bottom of the range, GLM 5.2 and Kimi K3, came up on the mirror image: a five-point August deficit is historically a 20-to-25% comeback proposition for the trailing side, and Florida's 19-point Republican governor win in 2022 shows the state's rightward drift resists national waves.
The most specific revision came from the fact nobody else had weighed:
"Shaved one point for the consolidation observation: Donalds won his primary with only 48%, an untested-statewide nominee needing to unify Collins and Fishback voters. A Trump endorsement usually heals that quickly, but 'usually' isn't 'certainly' for a first-time statewide candidate." (Claude Fable, 83.0 to 82.0)
Claude Sonnet and DeepSeek V4 held, Sonnet on the grounds that no peer introduced new evidence and Florida's baseline has shifted materially rightward since 2018.
One tension is worth naming in our own voice: the panel anchors on the same public polling that just missed this primary. The defense is that the polls got the frontrunner within half a point and only missed where an eleven-way undecided bloc would go, and a two-way general is the question those polls are good at. The first nonpartisan post-primary poll now exists, and it puts the two-way race inside a point; the re-score that tests the assumption is pending.
What Moves This Number Next
- The Second Nonpartisan Post-Primary Poll. The first is covered above. By the rule this page set, a nonpartisan read inside three points pushes the panel's number down several points at the re-score; a second one repeating a race inside a point locks that in, and Republican +8 or better pushes it back up. The seats above are the August 18 verdicts until the panel is re-run.
- Whether Collins And Fishback Endorse, And How Fast. In the days after the primary Collins said publicly he was not yet ready to endorse, though the two have talked. Consolidation inside two or three weeks reads as normal primary healing; a public holdout past Labor Day is the event that makes 48% look like a ceiling, and as of September 19 there is no endorsement from either.
- Jolly's First General-Election Finance Filings. Florida candidates file with the Division of Elections on a rolling schedule through the fall. If Jolly's early filings close the fundraising gap materially, the panel's Democratic number moves. The filings through September 11 did not: $590,000 raised against Donalds' $3.2 million over the same week.
- A Three-Point Move In The National Numbers. The Democratic case leans on the generic ballot and presidential approval as of August 17, neither re-measured for this update; public aggregators refresh them weekly, which makes them the cheapest thing on this list to check.
Election day is November 3, 2026. The contracts list Kalshi's expected expiration of January 8, 2027, with early settlement once a consensus of media organizations projects the winner. This page re-scores when the story moves; prices above are as of September 19, 2026.
What The Primary Taught About This Market
The August 18 primary was the market's first graded test on this race, and the pattern matters for November. Florida's Elections Canvassing Commission certified the count on August 27, 2026; vote figures here are the official results posted on Florida Election Watch.
- The Nominee Contracts Were Never In Doubt. Donalds closed at 99.9 cents on Kalshi, Jolly at 99.7, and both have since been finalized YES.
- The Margin Board Missed Badly. Polymarket's margin-of-victory favorite was "Donalds wins by 40 points or more" at 47.5 cents. The certified margin was 22.6 points, a band that traded at three cents.
- The Vote-Share Board Was A Coin Flip. "James Fishback under 10%" closed at 49.5 cents, the top price on the board; Fishback took 10.5% of the certified Republican ballots.
- The Polls Were The Cause. Per 270toWin's aggregate, Donalds finished half a point from his polling number and Collins finished nearly 12 points above his. The market inherited the polls' error on the undecided bloc wholesale.
Our own calls sit in the same ledger. Our Florida governor primary odds page named Collins the 77-cent second-place favorite, and Collins finished second. Our primary model verdict blended Donalds to about 82% against a 99-cent market price; he won, the crowd was better calibrated than our panel, and we take the loss on the scoreboard. Our Fishback page reported the even-money line on his vote share sitting just under 10% on the evening of the vote, and he landed just above it.
The Bottom Line
Florida is a Republican hold on both the market and the panel, and the gap between them is down to a point. Three quarter-million-contract sell orders in 18 days, the first taking the price from 84 to a 79 low before the book closed it back at 81, the second knocking it two cents for an afternoon, the third sweeping the bids from 81 to 78 and closing there, with only a cent and a half recovered since, say the book that used to absorb them has stopped absorbing. The nonpartisan poll this page was waiting for has landed at a tenth of a point, and until the panel is re-run its 78.5 is the number on this board with a date problem, not the market's. The number above is judgment with a date on it, and Georgia's board, with its favorite at 54.5 cents, shows what the same panel looks like when it sees a real contest.
Hero illustration: OddsShopper, in the house collage style. Old Florida State Capitol photo by DXR, licensed CC BY-SA 4.0; photos cropped, toned, and composited.
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: 2026 Governor Races: Every Kalshi Board We Track · Georgia Governor Odds 2026: Kalshi Board and Model Verdict · Michigan Governor Odds 2026: Kalshi Board and Model Verdict · Florida Senate Primary Odds 2026: Kalshi's Nixon Pileup · Minnesota Governor Primary Results: Demuth Beats Lindell By 10.9
