FL-25 Democratic Primary Odds: Moskowitz Vs Larkin

Florida's redrawn 25th District votes Tuesday, August 18, and Kalshi has turned its Democratic primary into a market. Sitting Rep. Jared Moskowitz trades as a heavy favorite over Democratic Socialist Oliver Larkin; prices on this page refresh as the final week plays out.

Updated August 12, 2026 · 8 min read · By Sam Smith

Redistricting scrambled South Florida this cycle, and FL-25 is where the reshuffle is clearest. The seat's longtime incumbent, Debbie Wasserman Schultz, drew new lines that pushed her into a different district, and a sitting congressman from next door, Jared Moskowitz, moved in to run for it. That set up a two-candidate Democratic primary between Moskowitz, a well-funded moderate, and Oliver Larkin, a Democratic Socialist running a door-knocking insurgency. Kalshi lists a separate yes/no contract for each candidate, and we asked six AI models to score the whole field without letting any of them see a single price. What they said, and where it lines up with the market six days out, is below.

The Quick Answer

The FL-25 Democratic primary is Tuesday, August 18, 2026, and Jared Moskowitz is a heavy favorite. Kalshi prices his nomination near 91¢ and Oliver Larkin near 11¢. Our price-blind, six-model panel, run August 12 without seeing any prices, agrees almost exactly: it blends Moskowitz to about 89% and Larkin to about 10%. The public polling points the same way, a Beacon Insights survey on July 22 had Moskowitz up 55.6% to 19.9%, with an even wider lead among voters who had already returned mail ballots. Larkin's campaign has released an internal poll claiming a tie; no independent public poll shows the race close.

• The Favorite: Jared Moskowitz, sitting FL-23 congressman running in the new FL-25, near 91¢ on Kalshi
• The Challenger: Oliver Larkin, a Democratic Socialist with a large field program but a roughly 5-to-1 total-fundraising deficit (and a wider cash-on-hand gap late), near 11¢
• The Panel: Moskowitz ~89% blend, Larkin ~10%, a rare near-match with the market rather than a gap
• The One-Line Read: a moderate incumbent-tier candidate with a public polling lead and a banked mail-vote advantage is where both the crowd and the models land, and they land close together.

The full model-by-model board, the fundraising and polling picture behind it, where the one contrarian data point sits, and a plain-English guide to reading these prices are all below.

Free: The Weekly PM Market Brief — the 8-model panel's graded record, the week's biggest market-vs-model gaps, and what's spiking next. One email, Sundays. Sign up in the box at the end of this article.

New to Kalshi event markets? Start with our plain-English guide to how Kalshi contracts actually work →

The Market At A Glance

Venue	Kalshi, a CFTC-regulated event-contract exchange (18+; availability varies by state, as of August 2026)
The Event	"FL-25 Democratic nominee?" (Kalshi event KXFLPRIMARY-25D26), one yes/no contract per candidate
The Contract	Each market resolves YES only for the candidate who wins the Democratic nomination for FL-25 at the August 18, 2026 primary, per Kalshi's settlement rules
Settlement Quirk	Florida primaries are decided by plurality, no runoff, so the nominee is whoever gets the most votes. Kalshi's listed November 3 expiration is a placeholder; the market resolves once the nomination is decided on August 18
Availability	Wasserman Schultz has a listed contract but is not a FL-25 candidate (she is running in FL-20); it is expected to resolve NO
Live Prices Below As Of	August 12, 2026. The board trades thin, and prices on this page refresh through primary week

These are model estimates, not predictions of fact and not financial or voting advice. Kalshi event contracts trade on a CFTC-regulated exchange; you must be 18 or older and in an eligible state to participate, and this article does not advocate for or against any candidate.

How Redistricting Set This Race

The number that explains this primary is not a poll, it is a map. Four fetched facts define the board:

• The Old Incumbent Is Gone From FL-25. Debbie Wasserman Schultz represented FL-25, but Florida's 2026 GOP-led mid-decade redistricting redrew South Florida. Under the new lines, Wasserman Schultz opted to run in FL-20, a safe Democratic seat where she faces a seven-candidate field. She is not on the FL-25 ballot.
• A Sitting Congressman Moved In. Jared Moskowitz has represented FL-23 since 2023, but the redraw dismantled his district, so he announced he would run in the new FL-25, which stretches along the coast from Delray Beach to Miami Beach.
• The General-Election Math Changed Too. The redraw shifted FL-25 from a seat Kamala Harris carried by about 5 points in 2024 to one Donald Trump would have won by 9, and both the Cook Political Report and Sabato's Crystal Ball rate the November race a toss-up. That is the general; this market is the Democratic primary.
• The Primary Field Is Two Candidates. On the Democratic side it is Moskowitz versus Oliver Larkin, a first-time candidate and member of the Democratic Socialists of America. Both had earlier been running in the old FL-23 before the lines moved.

The Board

Here is the narrow, phone-friendly view: the live Kalshi price next to the panel's blend. The full model-by-model table is just below it.

Candidate	Context	Live market	AI blend
Jared Moskowitz	Sitting FL-23 rep; moderate; ~11-to-1 cash-on-hand lead	~91¢	89.3%
Oliver Larkin	DSA challenger; grassroots field program	~11¢	9.8%
Debbie Wasserman Schultz	Not on the FL-25 ballot (running FL-20)	~1¢	~0%

These are model estimates, not predictions of fact and not financial or voting advice. The live-market column is as of August 12, 2026 on a thin board; the AI blend is the simple mean of the six model numbers taken the same day, price-blind, after a revision round. The two live contracts (Moskowitz ~91¢, Larkin ~11¢) sum to about 102, with Wasserman Schultz pinned near the one-cent floor.

More live boards from the same panel: Florida Republican governor primary (Byron Donalds ~98¢, the same August 18 Florida ballot on the GOP side) · Wisconsin Democratic governor primary (David Crowley, now settled YES near 100¢ after an election-night reversal) · Minnesota Republican governor primary (Lisa Demuth, settled YES near 100¢). Prices fetched August 12, 2026.

Every Seat's Number (Price-Blind)

[Editor's note: the six model verdicts below were generated price-blind on August 12, 2026, before the August 18 primary. Each model saw a fetched, dated summary of the candidates, the redistricting timeline, the fundraising gap, and every public poll, but no market prices. The "Market" column is the live price the panel was blind to, shown here only for comparison. One point of fact the models were given and that the article confirms: Debbie Wasserman Schultz is not on the FL-25 ballot, so her contract cannot resolve YES regardless of its listed price.]

Candidate	Market	Fable	Opus	Sonnet	GLM	Kimi	DeepSeek	Blend	Gap (blend% − mkt¢)
Jared Moskowitz	91¢	90.0	89.0	91.0	90.0	86.0	90.0	89.3%	-2
Oliver Larkin	11¢	9.0	10.0	8.0	10.0	13.0	9.0	9.8%	-1
Debbie Wasserman Schultz	1¢	0.0	0.0	0.0	0.0	0.0	0.0	~0%	-1

These are model estimates, not predictions of fact and not financial or voting advice. The Blend column is the simple mean of the six model numbers in its row, taken after a revision round in which each model read the others' anonymized reasoning. The last column is the directional gap, the blend percentage minus the raw market price in cents; read it as direction and size, not a precise edge. The two live contracts already sum to about 102, so there is almost nothing to strip out, and the panel lands within a point or two of the market on every row, an unusually tight agreement.

A note on the panel. Our standard Model Verdict lineup runs eight seats. On this run two were unavailable (the ChatGPT and Gemini seats were not installed on the machine that produced these verdicts), so this board runs on six models: Claude Fable, Claude Opus, Claude Sonnet, GLM, Kimi, and DeepSeek. All six scored the field on August 12, 2026 without seeing any prices, working from a fetched, dated summary of the candidates, the withdrawal-and-redistricting timeline, the fundraising figures, and every public poll. Each model's number also lands in our back-tested Model Verdict ledger, where past price-blind calls are graded against settled results so the method is measured, not just asserted.

Jared Moskowitz: Why The Favorite Is Priced SO High

Every model made Moskowitz the favorite, and none put him below 86%. The case rests on three fetched facts that all point the same way. First, the public polling: Beacon Insights, a Colorado firm, has surveyed the race multiple times this cycle, and its most recent July 22 wave (n=532 likely Democratic primary voters, margin of error ±3.6) had Moskowitz at 55.6% to Larkin's 19.9%, with 24.5% undecided, up from a 51.2%-to-19% June benchmark. Second, the money: Moskowitz has raised roughly $2.5 million to Larkin's roughly $550,000 and carried an 11-to-1 cash lead into the final 10 days. Third, and heaviest in the models' reasoning, the banked vote: among the 21% of respondents who had already returned mail ballots, Moskowitz led 75.3% to 19.3%.

That last number is why the panel sits so high with a week still to run. As DeepSeek put it, the mail-vote lead "is the most concrete data point and creates a massive hurdle for Larkin." Sonnet made the same point about direction: the early-vote subsample is "actual cast votes, not stated preference," and it shows a wider margin than the topline, which "undercuts any 'quiet momentum' story for Larkin since committed early voters already favor Moskowitz more than the topline." Opus reduced it to arithmetic: with a fifth of the electorate already voting 75-19, "Larkin must win the remaining electorate by an implausible margin even if the 24.5% undecided break heavily his way."

Oliver Larkin: The Case For The Tail

The panel did not price Larkin to zero, and its reasoning for keeping him in double digits is worth stating plainly, because it is also where the market and the models most nearly disagree. Larkin runs the kind of campaign that has produced primary upsets before: a Democratic Socialist field operation built on more than 30,000 doors knocked, roughly 100,000 calls and texts, and 70-plus volunteers, running on Medicare for All, a $25 hourly minimum wage, and refusing PAC money. His campaign has publicized an internal poll claiming a virtual tie.

The models weighed that and discounted it, but not to nothing. Fable kept Moskowitz at 90 rather than 95 specifically because of Larkin's large 30,000-door field program, "the classic mechanism for polling misses in low-turnout primaries." Sonnet noted that "DSA-style grassroots campaigns occasionally overperform polls (AOC 2018, Bowman 2020)," while adding that "those upsets came against thin or absent polling, not a widening 35+ point lead confirmed by banked votes." Kimi, the high seat on Larkin at 13%, flagged the internal poll directly: "if true, massive polling miss, but no independent confirmation and campaign internals average +6 bias." That is the tail the 10% blend is paying for, an August primary with a large undecided share and a real ground game, weighed against a public polling lead every independent survey confirms.

Where The Panel Changed Its Mind

After round one, every model read the other seats' anonymized numbers and reasoning and could revise. The movement was small but pointed in one direction, toward the banked-vote evidence.

• GLM Moved Up, And Said Why. It lifted Moskowitz from 82% to 90% after reading the group, concluding that "the peer consensus reflects this reality, correctly weighting the early vote over the raw top-line poll." — GLM
• Fable Moved Down, Toward The Tail. It trimmed Moskowitz from 93% to 90%, giving Larkin's ground game more room: it discounted "for the 24.5% undecided share, a single-pollster text-methodology risk, and Larkin's ... large field program (30K doors) ... that keeps me at 90 rather than 95." — Fable
• Sonnet Held At 91%, On Purpose. It saw no reason to move because "independent polling didn't just hold but widened over two waves," and the mail-vote subsample "undercuts any 'quiet momentum' story for Larkin." — Sonnet

The net effect was convergence: the six seats started spread from 81% to 93% on Moskowitz and finished bunched from 86% to 91%, landing the blend at 89.3%, a hair under the market's 91¢. That the panel tightened toward the market rather than away from it is the finding here, this is a board where the crowd and the models mostly agree.

What Would Change The Panel's Mind

The vote is six days out and mostly one-directional, but the triggers that would move the number are concrete. Each pushes a specific way:

• A Credible Independent Public Poll Showing Single Digits. Every model named this as the one event that would force a downward revision on Moskowitz. A non-campaign survey confirming Larkin's internal "tie" would push the number toward Larkin; another Beacon-style 30-point result pushes it toward Moskowitz.
• Election Day Turnout Composition. If Larkin's 30,000-door field program manufactures an outsized Election Day electorate that the mail-vote skew missed, that pushes toward Larkin. A low, mail-heavy turnout that mirrors the banked vote pushes toward Moskowitz.
• A Late Scandal Or Gaffe. Several seats named a disqualifying, credible late revelation about Moskowitz as the classic upset mechanism; absent one, it pushes toward Moskowitz by default.
• The Result Itself, On August 18. A called race or concession resolves the board to the winner and pushes the price to 100¢/0¢.

Dark Horses The Panel Won't Dismiss

This is a two-candidate ballot, so the "dark horse" is not a third name, it is the scenario the favorite's price leaves on the table.

• Larkin's Field-Driven Upset (Blend ~10%). The models' entire Larkin number is the tail where an underfunded insurgent's ground game turns out irregular voters the polls undercount. Kimi held it highest at 13%; every seat kept it in the double digits rather than folding to the market. It is a real, if secondary, path, and it is the only one on the board.
• A Polling-Industry Miss (Embedded In The Tail). Fable and Kimi both flagged single-pollster, text-message methodology risk. If Beacon systematically missed a mobilized young electorate, the topline could be softer than 55-20. No independent poll suggests it, which is why the tail is 10% and not 30%.

Debbie Wasserman Schultz, listed at about 1 cent, is not a dark horse at all, she is not on this ballot, covered in the traps below.

The Traps: Rows That Look Like Signal But Are Not

Transparency matters more than a clean story, so here is where the table would mislead you:

• The Wasserman Schultz Contract Cannot Win. She is the former FL-25 incumbent, but redistricting moved her, and she is running in FL-20, not FL-25. Her contract sits near the one-cent floor and is expected to resolve NO. Do not read a listed price as candidacy.
• The Internal "Tie" Poll Is Not Independent. Larkin's campaign released a poll claiming a close race. Campaign internals are strategic releases and, as one model noted, average a multi-point house bias. Weigh it against the public Beacon waves, not alongside them.
• A Thin Board Makes The Price Softer Than It Looks. This market trades on low liquidity. The ~91¢ is a real read, but getting in or out at any single number is not always possible, and small trades can move the quote.

Settlement Timeline

Primary / Decision	Tuesday, August 18, 2026 (Florida primary; plurality, no runoff). Vote-by-mail already underway
How It Resolves	Each Kalshi contract resolves YES only for the FL-25 Democratic nomination winner; withdrawn or non-candidate names cannot win regardless of listed price
Kalshi Settlement	Contracts resolve once the nomination is decided; Kalshi's listed November 3, 2026 expiration is a placeholder, not the decision date
Next Catalyst	The general election is November 3, 2026, in a district both Cook and Sabato rate a toss-up under the new lines
Re-Score Policy	We re-score this board whenever the story moves, and post a settlement retrospective on this same URL once the result is final. Live prices above are as of August 12, 2026; the panel verdicts are as of August 12, 2026.

How To Read These Prices

The models can tell you where they parted from the crowd, but the number in front of you still has to be understood in context. Here is how to read a market like this one, plus the honest friction of a thin, event-driven board.

Work an example. Say Moskowitz's contract shows a 91-cent YES price. Taken at face value, that implies about a 91% chance he wins the nomination. To sanity-check it you would normally strip the board's overround by adding up every candidate's YES price to see how far past 100 it sums, but here the two live contracts (Moskowitz ~91¢, Larkin ~11¢) already sum to about 102, so there is almost nothing to strip and 91¢ maps to roughly 89% to 91%. That is unusually clean: on a crowded field the favorite's number is always a little softer than it looks, but a two-horse board like this one is close to honest. The habit still matters, checking what a fair, spread-free version of a price looks like is the whole idea behind converting Kalshi's cents into real odds before comparing anything.

That same discipline of comparing a number to its fair value is the method behind our other boards. The Florida Republican governor primary verdict runs the identical playbook on the same August 18 Florida ballot, and every contest we track this way, Senate seats and governors' races alike, is collected on our 2026 Senate election odds hub.

Every number in this piece gets graded in public once the market settles; you can see the method's full record on the graded scoreboard.

This board lives on Kalshi and our panel does not trade it. But if you came for the method rather than this one race, OddsShopper's free expert picks today apply the same shop-the-number, find-the-fair-value discipline to the live sports slate.

A few honest points before reading too much into any number in the table above:

• Eligibility. Kalshi is a CFTC-regulated event-contract exchange, not a sportsbook. You must be 18 or older and located in an eligible state, and availability varies by state and can change.
• Liquidity And Spread. This board trades thin, so the "price" is softer than one number suggests, and getting in or out at your number is not always possible.
• A Moving Story. Prices on this page refresh through primary week, and the panel estimates reflect what was known on August 12, 2026.
• Not Advice. Everything here is analysis for information, not a recommendation to trade or to vote. These are model estimates, not predictions of fact and not financial or voting advice.

FAQ

Who is favored in the FL-25 Democratic primary? As of August 12, 2026, sitting U.S. Rep. Jared Moskowitz is a heavy favorite. Kalshi prices his nomination near 91 cents, with Democratic Socialist challenger Oliver Larkin near 11 cents. Our price-blind six-model panel agrees, blending Moskowitz to about 89 percent and Larkin to about 10 percent. Public polling from Beacon Insights had Moskowitz up 55.6 percent to 19.9 percent on July 22, with an even larger lead among voters who had already returned mail ballots.

When is the FL-25 primary and how does the Kalshi market settle? The Florida primary is Tuesday, August 18, 2026, and vote-by-mail is already underway. Kalshi lists a separate yes/no contract for each candidate under the event KXFLPRIMARY-25D26, and each resolves YES only for whoever wins the Democratic nomination for FL-25. Florida primaries are decided by plurality with no runoff. Kalshi's listed November 3, 2026 expiration is a placeholder, but the market resolves once the nomination winner is determined at the August 18 primary.

Is Debbie Wasserman Schultz running in FL-25? No. Wasserman Schultz was the FL-25 incumbent, but Florida's 2026 redistricting redrew the lines and she is now running in FL-20, a safe Democratic seat with a seven-candidate field. She is not on the FL-25 Democratic ballot. Her Kalshi contract remains listed at about 1 cent, but she cannot win the FL-25 nomination and it is expected to resolve NO.

What did the AI models say? Our six-model panel scored the field price-blind on August 12, 2026, seeing a fetched, dated summary of the candidates, the redistricting timeline, the fundraising gap, and every public poll, but no market prices. After a revision round in which each model read the others' anonymized reasoning, the panel blended Moskowitz to about 89 percent, Larkin to about 10 percent, and Wasserman Schultz to roughly zero. Each number lands in our public graded ledger and will be scored against the result once Kalshi settles.

The Bottom Line

This is the rare board where the crowd and the models mostly nod at each other. Redistricting cleared the FL-25 incumbent out, a well-funded moderate congressman moved in, and a Democratic Socialist challenger is running the kind of grassroots campaign that has sprung primary upsets before, just not against a 35-point public polling lead already confirmed by a 75-19 edge among banked mail ballots. Kalshi prices Jared Moskowitz near 91 cents and Oliver Larkin near 11; our price-blind six-model panel lands at about 89 and 10, a hair softer on the favorite and paying for exactly one thing, the tail where a big field program beats the polls. We will grade that call in public once the August 18 vote settles. If this method is your thing, our guide to Kalshi versus Polymarket is the natural next read.

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.