Prices as of August 25, 2026, 6:11 p.m. ET. Record compiled from 26 dated sources; see the sourced record file for every citation.
This is the steelman case for Scott Brown. The companion piece argues the other side: Why John E. Sununu (R) Will Win the NH Senate Primary. The Democratic half of the September 8 ballot runs through Why Chris Pappas Will Win.
On February 1, 2026, Donald Trump gave his "Complete and Total Endorsement" in New Hampshire's Republican Senate primary to John E. Sununu, the man who had published an op-ed in the New Hampshire Union Leader on the eve of the January 2024 presidential primary under the headline "Donald Trump is a loser." He did it over Scott Brown, the ambassador to New Zealand and Samoa he had appointed in 2017, according to Fox News. Brown's statement that night ended with "Let's keep working!" Seven months later he is still working, and his whole case now fits in four words he gave the Boston Globe's Alex Beam on August 21: "New Hampshire breaks late."
The market has not broken at all. On Kalshi, where a price in cents is a probability, Brown's contract to win the Republican nomination asks 3.4 cents, about +2840 at a sportsbook. Sununu asks 97.8 cents, roughly -4445. The freshest poll, Saint Anselm College's August 17-18 survey of likely Republican primary voters, has Sununu at 54 percent and Brown at 24.
And then, on August 25, the front-runner declined the WMUR-TV debate. The Sununu campaign told the Union Leader that "John has already done two debates and 20 joint appearances with Scott Brown," and that with days to go it is "focused on the town-to-town, person-to-person campaign." Brown's answer: "He doesn't want to show up. He feels privileged that he's somehow entitled to a debate-free campaign. I feel bad for the people of New Hampshire." Then, for the cameras: "I think he's a little bit scared." That is the whole argument in one news cycle: a challenger who wants the stage, a favorite who says he is done sharing it, and two weeks of campaign left.
Stay ahead of the markets.
Daily insights and expert picks on Kalshi, Polymarket, and what's moving markets.
Free forever. Unsubscribe anytime.
The Board
As of August 25, 2026, 6:11 p.m. ET (Kalshi, cents = probability; the bid is what a buyer will pay, the ask is what a seller wants):
| Outcome | Yes price (buyers bid / sellers ask) | Reads as |
|---|---|---|
| John E. Sununu | 97.5/97.8¢ | about 98% (≈ -4445 in sportsbook terms) |
| Scott Brown | 2.0/3.4¢ | about 3% (≈ +2840) |
| Chris Sununu | 0/1.8¢ | about 2%, no bid |
| Dan Innis | 0/1.9¢ | about 2%, no bid, 37 contracts traded in the market's lifetime |
The board says the primary is a formality: the favorite at 98, the 2014 nominee at 3, the favorite's brother and a fourth name at a penny or two, neither with a bid. Neither of those two filed: the filing period closed June 12, 2026, and Citizens Count's September 8 ballot list carries eight Republicans led by Sununu and Brown, so the Chris Sununu and Dan Innis contracts are dead paper, not live probability. Polymarket, checked the same evening, had Brown at 3.05 cents and Sununu at 96.25, so the two exchanges agree on Brown to within half a cent and on Sununu to about a cent and a half. Brown's market is the older one, opened December 4, 2024, ten months before Sununu entered, and it still carries more lifetime volume than the front-runner's, about 21,600 contracts to John E. Sununu's 12,700 (the dormant Chris Sununu contract holds the most, about 22,700, with nothing traded in the last 24 hours). It is quiet now: 12 Brown contracts and 8 Sununu contracts traded in the last 24 hours, and Brown's last print was 2.3 cents, below the prior 3.5-cent mark on the tape. The contract settles on the nomination itself, which New Hampshire decides in a single round on September 8, 2026: whoever has the most votes that night is the nominee.
What the Models Think
We also asked a panel of AI models the same question without showing them the price, only the sourced record this article is built on. Six of the panel's eight seats ran this board. Their blend, the midpoint of the six answers, put Brown at 3.5 percent, a hair above the 3.4-cent ask, and the widest gap between any two models was 4 points, from 2 to 6. The reason they gave most often was the same three-part stack: a 30-point deficit in the freshest poll, the endorsement, and nearly four times the cash on the other side. One model put the late-break problem plainly: comebacks of this size "typically require a late scandal or external shock, none of which is evident here."
For once the panel and the market agree almost to the cent, and that is its own kind of finding. There is no mispriced number for Brown's case to hide behind; the argument below has to beat the record on its merits. These are model estimates, not predictions: the market has real money behind it, and the panel has only the record.
How the Ambassador Lost the Endorsement and Kept Running
Two lines of Brown's résumé price this contract. The first is 2014, when he came within three points of Jeanne Shaheen in this very race, 51 percent to 48: statewide name recognition he did not have to buy. The second is the ambassadorship, confirmed 94-4 in June 2017 and held until December 2020, the loyalty credential that was supposed to make Trump's endorsement his. He got into the race in June 2025 as the first major Republican in, and by October 14, 2025, Fox News reported he had raised $1.2 million in a quarter with about $900,000 on hand. The 2010 Massachusetts special election, 51.86 percent to 47.10 over Martha Coakley, is the upset he keeps pointing to.
Eight days later, on October 22, 2025, John E. Sununu launched his comeback bid, and the race stopped being Brown's. Sununu held this seat from 2003 to 2009 before Shaheen took it from him. The Globe's James Pindell counted "19 Sununu ballots since 1972" and wrote that rejecting the family while defying Trump is "close to prohibitive." Brown tried anyway. On WMUR, he said of Sununu: "What did he have to do to sell his soul? Because when the president's gonna call on him, he's going to say, 'You know John, I want this. You gotta jump.'"
The general election explains why the party closed ranks so fast. The same Kalshi fetch priced the November general at 11 to 13 cents for the Republican side and 88 to 91 for the Democratic side; the New Hampshire Senate model-verdict board and the 2026 Senate election odds hub carry the race. Saint Anselm's August poll has Chris Pappas beating Sununu 48 to 41 and Brown 49 to 37, with Trump's state approval at 37 percent. A party reading those numbers picks the smaller deficit, and the balance-of-power board shows why one seat matters.
The Late-Break Case, and the Record It Has to Beat
Brown's argument has four moving parts, and the honest version names each one's price.
The first is the ballot itself. New Hampshire runs semi-closed primaries: an undeclared voter can walk in on September 8 and ask for a Republican ballot, then return to undeclared before leaving the polling place, per the Secretary of State's own guidance. Brown read the endorsement as an opening on exactly that flank. "It opened up a new lane for me to get support from the independents who really control the voting here in New Hampshire," he told the Washington Examiner in May. The March Emerson poll that had him down 48 to 19 also had a third of Republican primary voters undecided, and the August Saint Anselm number still leaves 22 points unaccounted for. That is the entire mathematical basis for "breaks late." One honest line on the precedent: the famous New Hampshire comebacks are presidential-primary stories, and even the best of them, Hillary Clinton in January 2008, closed an average eight-point deficit to win by three, per CBS News. I could not find a New Hampshire state primary that moved 30 points in two weeks. If one exists, it is the best argument Brown has, and nobody has made it.
The second is Trump himself, and here the primary electorate cuts differently from the state. Sununu, asked by NHPR in July whether he would have launched Trump's Iran war, said "no, no." Brown said "I love what he's doing with Iran. He needs to finish the job." In a state where two-thirds of voters opposed the Iran decision in UNH's late-July poll, that is a general-election problem. In a Republican primary, it is the loyalty record Brown is running on, against a rival whose endorsement arrived despite the "loser" headline, not because of it. At the June 24 debate in Hampton, Brown drew the line himself: "I support 95% of what he does, but respectfully, he's not right all the time," and added that he is the one who can say "no, sir" to the White House "because I didn't get his endorsement."
The third is the stage. Sununu skipping the WMUR debate leaves Brown asking for a stage his rival has vacated, and the Sununu campaign's own language, "with just days to go," frames the calendar as the only thing Brown has left to work with. Brown's precedent is 2010, which he describes as "being down 41 points" before he won by under five; the deficit is his characterization, not a fetched number, and the models named the flaw at once: that was a general election in a nationalized moment, not a September primary against a Trump-endorsed opponent with the family name.
The whole case in one line."If I thought I would lose, I wouldn't be in the race," Brown told the Globe on August 21. The market's reply is 3.4 cents; the poll's reply is 30 points; the calendar's reply is two weeks.
The fourth is money, and it is where the case is thinnest. At the end of June, per FEC filings reported by the Globe, Sununu had nearly $2.7 million on hand to Brown's $736,000, with the NRSC also running ads for Sununu. Brown's response, to the Washington Examiner in May, is discipline rather than scale: "We have a budget. We're sticking to that budget. We have a plan. We're executing the plan." And, in the Globe on August 21: "He's wealthy and they like self-funders. Just follow the money." Sununu's answer at the Hampton debate was flat: "Nobody's contribution makes any difference. For someone to stand up and say, 'well, yes, it does,' that makes me question where they're coming from."
Now the record the case has to beat, because it is the same record the models read. On March 25, 2026, NH Journal reported that Strong As Granite, a super PAC backing Brown, spent about $130,000 on mailers tying Sununu to Jeffrey Epstein, the primary's loudest controversy so far. The mailer pointed to a 2010 email in the Epstein files that said "john sununu, has good stories," and asked, "Why would a Sununu show up in the files…?" It never said which Sununu, and the email could refer to the former senator or his father; the same report found no evidence that either man met or communicated with Epstein. The mailers landed the day after Saint Anselm had Sununu ahead 49 to 28. Asked about the mailer, Brown told NH Journal he had nothing to do with it or the PAC's strategy: "I'm sure there are going to be pro-Scott and pro-John, etc. ads throughout the entire campaign." The Sununu campaign answered: "Scott Brown is losing the primary by over 20 points and continues to lose endorsements. Perhaps lies and smears are how they do politics in Massachusetts, but in New Hampshire people prefer the truth." Republican consultant Matthew Bartlett called it "desperate and crazy." Whatever the mailer did to the race, the gap in Saint Anselm's own numbers went from 21 points in March to 30 in August.
Brown carries his own entry, too. In October 2017, PBS NewsHour and ABC News reported that the State Department's Office of Inspector General had reviewed complaints about his comments at a July Peace Corps event in Apia, Samoa, where he told a food server she "could make hundreds of dollars as a waitress in the U.S." and told guests "they looked beautiful," and that senior State Department leadership counseled him on standards of conduct. His response then: "You're not Scott Brown from Rye, New Hampshire, anymore, you're an ambassador and you have to be culturally aware." Nothing in the 2026 record shows it being used against him in this primary, but it is in the file, and a late-breaking race surfaces files.
The number that matters most is favorability, the line the Globe pulled from UNH's survey of likely Republican voters: Sununu 54 percent favorable against about 17 unfavorable; Brown 32 favorable, 35 unfavorable. A late-break theory needs persuadable voters, and Brown first has to repair his standing with the electorate already eligible to choose him. Everything above depends on that number changing, and the only favorability reading in the record is August's; it is the wrong shape for a late break.
A Worked Example of the Trade
One Yes contract on Brown at the 3.4-cent ask pays $1.00 if he is the nominee: risk 3.4 cents to win 96.6, the +2840 structure the board translates to. The honest version of this contract is about whether anything between now and September 8, 2026 moves the number, because a long shot's value lives in the reprice, not the settlement. A public poll inside 20 points would be the first real evidence that the market needs to reprice Brown; no public poll this cycle has produced one. The other side of the same book is the favorite at 97.8 cents: risk 97.8 to win 2.2, about -4445, a 2.2 percent return in two weeks if the favorite holds. That certainty, not Brown's case, is what the market is really pricing.
The Bottom Line
Two weeks out, Scott Brown is a 3-cent candidate on a quiet board, the AI panel lands within a point of the market, and in Saint Anselm's own numbers the gap has widened from 21 points in March to 30 in August. Here is the number that would make this desk wrong: a public poll with Sununu inside 20 points, or a Brown bid above the 2 cents where it sits tonight, before September 1, 2026. No public poll this cycle has had Sununu inside 20. If one lands, Brown would finally have evidence for the undeclared-voter lane his case depends on; if nothing moves by September 1, the late break he is promising has a week left to exist, and the favorability line says it will not.
More on this: 2026 Senate Races: Every Kalshi Board We Track · Why John E. Sununu (R) Will Win the NH Senate Primary · Why Karishma Manzur (D) Will Win the NH Senate Primary · Why Chris Pappas (D) Will Win the NH Senate Primary · Will There Be A 2027 MLB Season? Some Owners Expect None
What Moves The Price From Here
- Any primary poll before September 8, 2026. Saint Anselm, UNH and Emerson have all published this year. A Sununu lead inside 20 points would put a bid above the 2 cents where it sits tonight; another 30-point reading would take it to the floor and leave the 3.4-cent ask as the only live quote.
- The WMUR-TV debate slot Sununu declined. If WMUR gives Brown airtime anyway, and a news cycle with it, the 24 percent he polled in August is the number to watch. The Sununu campaign's bet is that "town-to-town, person-to-person" beats television with days to go.
- Undeclared turnout on primary day. The June 2, 2026 deadline for registered voters to change affiliation has passed, but undeclared voters can still pick a Republican ballot at the polls. Brown needs that lane to show up for him, and the Democratic side, where Chris Pappas asked 97.4 cents on Kalshi's nominee board the same evening, gives an undeclared voter little reason to take the other ballot.
- The general, once the nomination is called. The Republican side of the November 3, 2026 general trades at 11 to 13 cents on the same exchange; the daily Kalshi picks sheet carries the New Hampshire line every morning. The primary market this article covers settles first, on the September 8, 2026 vote.
Hero illustration: OddsShopper, in the house collage style. Scott Brown photo by Online Guide to House Members and Senators, licensed Public domain as an official Congressional portrait; photos cropped, toned, and composited.
Prices quoted are live Kalshi market prices as of August 25, 2026, 6:11 p.m. ET, and will keep moving until this market settles after the September 8, 2026 vote.). OddsShopper covers prediction markets as an independent analyst: any model or panel figures referenced are model estimates, not predictions of fact and not financial advice. This article does not advocate for or against any candidate or party. Trading involves risk; contracts can go to zero. The panel's graded record lives on the public Model Verdict Scoreboard.
