2028 Presidential Election Odds: Two Different Favourites

Published 2 October 2026 · Prices frozen 2 October 2026, 21:30 UTC

Search 2028 presidential election odds today and the top results will tell you Vance is on 30.4% and Ocasio-Cortez on 7.5%. Neither number has been true for months, and more to the point neither says which of the six live boards it came from. There are six, across two exchanges, holding $2.71bn and 370 million contracts between them, and they do not agree with each other — not on who the Democratic favourite is, and not on whether the Democrats are a 57% shot or a 65% one. We read both venues in the same pass, pulled the full lifetime daily history on every rung that matters, swept 12,000 open Kalshi events to find the ones nobody quotes, and summed each board by hand. Two of the four disagreements turn out to be real. The other two are arithmetic nobody checked.

Where the 2028 presidential election odds stand

Start with the six contracts a reader is most likely to have seen quoted. Polymarket runs three candidate boards and a party board; Kalshi runs the same four plus a trifecta board and an electoral-college margin board. The live versions of all four Polymarket boards sit on our 2028 party odds page and the three candidate pages beside it. On the headline question — who becomes president — the two venues are within two-thirds of a point of each other. On almost everything else they are not.

Read that table one row at a time and the shape of the disagreement is already visible. The two venues agree almost exactly on Vance becoming president, split 6.25 points on his getting nominated, and split in opposite directions on the two Democrats at the top of the field. A venue that was simply pricing the Democrats higher across the board would be a fee or a funding story. One that prefers Ocasio-Cortez while the other prefers Ossoff is a disagreement about the primary itself.

Polymarket says Ocasio-Cortez, Kalshi says Ossoff

This is the finding. Two of the largest political prediction markets in the world, reading the same field on the same afternoon, do not have the same front-runner for the Democratic nomination — and the disagreement is not confined to the top of the board.

Polymarket, whose Democratic nominee board carries 128 rungs to Kalshi’s 48, is dearer on the three Californian-and-New-York names at the top and cheaper on almost everyone below them. Kalshi pays more for governors: Beshear is 6.30¢ there against 3.10¢ on Polymarket, which is the widest proportional gap on the board at more than double. Ossoff, the only name the two venues effectively agree on, is also the name with a general election to win first — more on that below.

None of these gaps is tradeable in the way a glance suggests, because the two venues have different fees, different funding rails and no common settlement. But a 3.75-point gap on a rung quoted 17.7/17.8¢ on one venue and 13/14¢ on the other is not noise either. Both books are tight. They simply disagree.

The rulebooks agree, and the prices still do not

Every cross-exchange split this site has taken apart in the last month has turned out to be a wording difference. Netanyahu’s exit contract was 1.5 points apart because Kalshi accepted an announced intention to leave and Polymarket did not. Greenland moved on whether a base lease counted. The Venezuela boards priced opposite tails because one of them named a jail. So the first thing to check here was the resolution text, and for once it is a dead end.

That leaves order flow, and the two boards are built for different traders. Polymarket’s Democratic board carries $1.287bn of lifetime volume across 128 rungs, which makes it the largest single market on the platform — larger than anything on the 2026 midterms, larger than any war or leader-exit contract. Kalshi’s has 48 rungs and 155.2m contracts of open interest, which is a serious book by any standard but a differently shaped one: fewer long-shot rungs, more weight in the middle of the field.

The replacement clause is worth keeping in mind anyway, because it is the one place the two could still come apart. A nominee who accepts in August 2028 and is replaced in September pays Yes on Polymarket by explicit instruction. Kalshi’s silence on the point is the kind of gap that gets resolved by a committee rather than by a sentence, and 2024 is recent enough that nobody should treat the scenario as hypothetical.

Three routes to a 2028 election odds party price

“Democrats 62%” or “Democrats 64%” is the line that has made it into most coverage, and it comes from a single contract: Polymarket’s party board, which asks only which party’s candidate is elected. It is also, by a distance, the thinnest instrument in this entire cluster. There are four other ways to get at the same number from the same two exchanges, and three of them land somewhere else.

The pattern is not ideological, it is liquidity. The two biggest books in the table — Polymarket’s $712m candidate board and Kalshi’s 70.8m-contract one — land on 56.9% and 57.9%, a single point apart. Kalshi’s party board, with 583,000 contracts, agrees with them at 58.0%. The three numbers above 65% come from the three thinnest instruments on the list, and the thinnest of all, ECMOV’s 7,022 contracts, swings between 65.0% and 74.7% depending on whether you read it at the midpoint or at the last trade. Three of its fourteen rungs have traded exactly once.

Which leaves Polymarket’s party board as the odd one out: $2.46m of lifetime volume, roughly 0.2% of the candidate board sitting beside it on the same exchange, carrying the number that gets quoted. It is not wrong, and it has been drifting up for a year — 52.50¢ at its low on 5 November 2025, a 66.50¢ high on 14 September 2026, 59.50¢ a month ago. But if you want the 2028 party price that the most money actually stands behind, add up the candidates and you get 57%, not 65%.

Why one board adds up to 134.50¢

A mutually exclusive board of candidates should sum to about 100¢. If it sums well under, the market is leaving money on the table; well over, and somebody could in theory sell the whole field. Polymarket’s Republican nominee board sums to 134.50¢, which is not a 34-point arbitrage. It is one rung that has never traded.

The rung is Robert F. Kennedy Jr., shown at 49.00¢. Look at what is behind that number: no bid at all, a 98¢ ask, $15 of resting liquidity, zero lifetime volume, and a CLOB price history that comes back as an empty array — there is no series to chart because there has never been a trade. The 49¢ is the arithmetic midpoint of nothing and 98¢. Every other rung on that board has a two-sided market; this one has a formatting convention.

It matters beyond tidiness, because that 49¢ is the second-highest number on Polymarket’s Republican board. A screenshot of the top of that board shows Vance on 50¢ and Kennedy on 49¢, which reads as a two-horse race and is not one. Kalshi’s equivalent board sums to 94.90¢ with no artefact of this kind; its long-shot rungs are honest 1¢ lines on names like Jeff Bezos and Candace Owens, each with real if small volume behind them. The lesson generalises: before you believe the second row of any board, check whether it has ever traded.

What a nomination is actually worth

Two boards on the same exchange let you back out a third number nobody quotes directly: the probability a candidate wins the presidency given that they are nominated. Divide the winner price by the nominee price and you have it. This is where the party disagreement stops being abstract.

Vance converts at 41.1% on Polymarket and 45.5% on Kalshi. The two leading Democrats convert at 70% and above on both venues. That is the same claim as the party number, said a different way: these markets think the 2028 general election leans Democratic, and they think the Republican nomination is the easier of the two to win. Vance is better than even money to be nominated and worse than even money to be elected if he is.

The one row that breaks the pattern is Newsom, who converts at 48.3% on Polymarket and 64.6% on Kalshi — a 16-point split, the widest derived disagreement in the cluster. Polymarket traders are paying for Newsom to be nominated and declining to pay for him to win; Kalshi traders treat him like any other Democrat. Whichever read is right, it is the clearest single place where the two venues are pricing different theories rather than different spreads.

Newsom has lost 23 points since November 2025

None of these prices arrived where they are. Polymarket’s CLOB keeps a daily series back to the day each rung opened in July 2025, and the shape of the cycle so far is one long decline for the early front-runner against two steady climbs behind him.

Newsom peaked at 39.50¢ on 6 November 2025, the day after that year’s off-cycle elections, with Ocasio-Cortez on 12.50¢ behind him. He has given up 23.25 points since, bottoming at 12.50¢ on 14 September 2026 before this month’s small recovery. Ocasio-Cortez has gone the other way: 7.50¢ at her low in September 2025, a 22.65¢ high on 24 August 2026, 17.75¢ now. The first daily close on which she led the board was 13 August 2026, at 20.80¢ against Newsom’s 17.35¢ — Newsweek picked the crossover up intraday the evening before.

Ossoff’s path is the quietest and the steepest in proportional terms: 3.55¢ in August 2025, still 3.45¢ in February 2026, then a climb through the spring and summer to 17.25¢ on 22 September. A rung that was a 3¢ long shot a year ago is now the Kalshi favourite. Vance, meanwhile, is 9.25 points off his own post-election peak and down 2.80 in the last 30 days, with Rubio down roughly 10 points since mid-July. Neither Republican has an obvious floor in the series yet.

The midterms are 32 days away, at the same number

There is an election before this one. Polling day for the 2026 midterms is Tuesday 3 November, 32 days from this snapshot, and Polymarket prices a Democratic sweep of both chambers at 64.50¢ — the same number, to the cent, that its party board gives a Democratic president in 2028. The coincidence is worth stating precisely because it is a coincidence: these are different questions with different resolution dates, and nothing links them mechanically.

The last two rows are the ones to sit with. Ossoff is the Kalshi favourite for the 2028 Democratic nomination at 16¢ while his own re-election, 32 days out, prices at 97.25¢ against Mike Collins’s 2.65¢ — so the nomination case is not hostage to the seat. And Kalshi’s 2028 trifecta board already prices a full Democratic sweep of presidency, House and Senate at 53¢, two years and one midterm before it could happen. The live 2026 balance-of-power board moves daily from here to polling day, and our Senate-flip piece walks the seat ladder behind it.

What to watch before the first contest

The calendar is the next thing to move these boards, and Kalshi lists it where Polymarket does not. Its board on which state votes first in the 2028 Democratic primary has New Hampshire at 83¢ against South Carolina’s 5¢ and Iowa’s 7¢ — a reversal of the order the party used in 2024, and a question the Democratic National Committee settles rather than the voters. Its convention-host board has Denver at 43¢ and Philadelphia at 37¢.

What we would actually watch is whether the two nominee boards converge after 3 November. A midterm result is the first hard information either venue has had since these rungs opened, and the 3.75-point gap on Ocasio-Cortez is exactly the kind of spread that closes when real news arrives. If it does not close — if Polymarket keeps paying more for the New York and California names and Kalshi keeps paying more for governors — then the two books have genuinely different populations in them, and the right answer to “what are the 2028 odds” is to quote both.

One number to keep an eye on either way: Kalshi has already opened a 2032 party board, KXPRESPARTY-2032, with the Democrats at 46¢ and the Republicans at 47¢ on 316 contracts of volume. That is what a market with no information looks like. Everything above is what two years of information looks like, which is the only reason any of it is worth reading.

Related analysis

• Democratic Presidential Nominee 2028 — the live 128-rung board, chart, volume and full resolution criteria
• Republican Presidential Nominee 2028 — live odds, and the untraded rung that inflates the sum
• Presidential Election Winner 2028 — the candidate board the party price should be checked against
• Which party wins the 2028 US presidential election — the party contract, live
• Georgia Senate 2026 — Ossoff's own re-election, the race behind his 2028 price
• 2026 midterms odds: the Senate flipped in 30 days — the seat ladder, both exchanges

Sources: the Polymarket Gamma and CLOB public APIs (events 30829, 31875, 31552, 33228, 32228 and 57642) and the Kalshi public events, series and single-market APIs (KXPRESNOMD-28, KXPRESNOMR-28, KXPRESPERSON-28, KXPRESPARTY-2028, KXPRESPARTY-2032, POWER-28, ECMOV-28NOV07, KXPRESMATCHUP-28NOV07, KXDEMFIRSTCONTEST-28 and KXDNC28HOST-28), both read at 2 October 2026, 21:30 UTC; resolution-criteria quotations are from the market rules as published on Gamma and Kalshi. The 12 August 2026 crossover was reported by Newsweek. The election date of Tuesday 7 November 2028 follows 3 U.S.C. §21, which defines election day as the Tuesday next after the first Monday in November, and the candidate field and primary calendar are as recorded in the 2028 United States presidential election record. Prices captured 2 October 2026, 21:30 UTC and not updated since. Nothing here is investment advice — see our disclaimer.