The House
Sikt: Democrats 90%
Chance Democrats win control:
Sikt 90%Kalshi 91%Polymarket 92%
Sikt Midterm Bench preview, Oct 27, 2026. Graded after the results.
2026 midterm predictions: AI forecast and live odds
Americans vote on Tuesday, November 3, 2026. Live prediction market odds for control of Congress and the Senate races that decide it, from Kalshi and Polymarket, refreshed several times a day.
7 days to election day
Who controls Congress?
The Senate
Sikt: Democrats 60%
Chance Democrats win control:
Sikt 60%Kalshi 64%Polymarket 66%
Sikt Midterm Bench preview, Oct 27, 2026. Graded after the results.
Senate races to watch
Closest races first, by the average of Kalshi and Polymarket. Sikt’s latest Midterm Bench forecast is shown first, and in green on each chart. Each card links to live odds, Sikt’s read, history and the market rules.
Iowa Toss-upHinson vs. TurekAshley Hinson (R) winsSikt 51%Kalshi 58%Polymarket 58%Maine Collins vs. JacksonTroy Jackson (D) winsSikt 54%Kalshi 60%Polymarket 62%Texas Paxton vs. TalaricoJames Talarico (D) winsSikt 60%Kalshi 66%Polymarket 68%Ohio Husted vs. BrownSherrod Brown (D) winsSikt 65%Kalshi 68%Polymarket 70%Michigan Rogers vs. El-SayedAbdul El-Sayed (D) winsSikt 76%Kalshi 68%Polymarket 69%Nebraska Ricketts vs. OsbornPete Ricketts (R) winsSikt 56%Kalshi 72%Polymarket 72%Alaska Sullivan vs. PeltolaMary Peltola winsSikt 53%Kalshi 74%Polymarket 74%New Hampshire Sununu vs. PappasChris Pappas (D) winsSikt 89%Kalshi 90%Polymarket 92%Georgia Ossoff vs. CollinsJon Ossoff (D) winsSikt 93%Kalshi 95%Polymarket 96%North Carolina Cooper vs. WhatleyRoy Cooper (D) winsSikt 95%Kalshi 96%Polymarket 97%
Sikt Midterm Bench
Sikt is forecasting these races.
Sikt Intelligence is running a live benchmark on the 2026 midterms. Our AI superforecaster puts its own probability on control of Congress and the key Senate races. Forecasts are up for control of the House and Senate and for every key Senate race below, updated each round, next to Kalshi and Polymarket, and every forecast will be scored against the results.
| Race | Outcome | Sikt | Market | Gap (pts) |
|---|---|---|---|---|
| Iowa | Ashley Hinson | 51% | 58% | -7 |
| Maine | Troy Jackson | 54% | 62% | -8 |
| Texas | James Talarico | 60% | 66% | -6 |
| Ohio | Sherrod Brown | 65% | 69% | -4 |
| Michigan | Abdul El-Sayed | 76% | 69% | +7 |
| Nebraska | Pete Ricketts | 56% | 72% | -16 |
| Alaska | Mary Peltola | 53% | 74% | -21 |
| New Hampshire | Chris Pappas | 89% | 90% | -1 |
| Georgia | Jon Ossoff | 93% | 95% | -2 |
| North Carolina | Roy Cooper | 95% | 97% | -2 |
SiktKalshiPolymarket
Updated 2026-10-27 15:35 UTC. Market is the average of Kalshi and Polymarket. Every forecast will be scored against the results.
How much will each race be won by?
Sikt’s middle estimate of the winner’s lead, in points, its 80% range (the result should land inside it 8 times in 10) and Kalshi’s margin markets at the start of round 9, Oct 27, 2026. Closest races first. In the chart, Democrats (and in Nebraska, the independent Osborn) are to the right of the middle line. How to read the margins →
Sikt: middle estimate, bar = 80% rangeKalshi market, same marks
| Race | Sikt | Sikt’s 80% range | Market | R ← → D |
|---|---|---|---|---|
| Maine | Jackson +0.7 | Collins +6.1 to Jackson +6.8 | Jackson +0.9 | |
| Alaska | Peltola +1.0 | Sullivan +5.2 to Peltola +7.0 | Peltola +2.7 | |
| Iowa | Hinson +1.1 | Hinson +6.9 to Turek +4.8 | Hinson +0.7 | |
| Texas | Talarico +1.4 | Paxton +4.2 to Talarico +6.9 | Talarico +1.4 | |
| Nebraska | Ricketts +2.0 | Ricketts +8.5 to Osborn +4.4 | Ricketts +3.4 | |
| Ohio | Brown +2.2 | Husted +3.6 to Brown +8.0 | Brown +2.2 | |
| Michigan | El-Sayed +4.0 | Rogers +1.5 to El-Sayed +9.6 | El-Sayed +1.9 | |
| New Hampshire | Pappas +7.6 | Pappas +1.5 to Pappas +13.8 | – | |
| North Carolina | Cooper +10.4 | Cooper +4.5 to Cooper +16.2 | Cooper +13.2 |
Seats, Sikt’s middle estimate and 80% range: Senate: Democrats win 51 seats (48 to 54), market 52; House: Democrats win 232 seats (219 to 244), market 239; House popular vote D +8.2, market D +9.4.
Just for fun
The long shots.
Politics has a long tail. Some very unlikely events, with illustrative odds: not live markets, and not Sikt Midterm Bench forecasts.
Long shotGeopoliticsWILD-001
The Statue of Liberty is sent back to France before 2035?
Sikt says
1.2%
Range 0.6–2.2%
Market says
2.7%
Sikt sees it 1.5 pts less likely than the market
Long shotGeopoliticsWILD-002
Trump and Kim Jong Un play golf together before 2029?
Sikt’s read: They met three times in 2018–19 and Trump says he’d do it again. Golf is the hard part: it needs a course and security both sides accept. Leaning no.
Sikt says
1.1%
Range 0.6–2%
Market says
3.4%
Sikt sees it 2.3 pts less likely than the market
How to read midterm odds
- The president’s party usually loses ground. It has lost House seats in every midterm since World War II except 1998 and 2002, a strong base rate.
- The Senate is about the map. Only 35 of 100 seats are up, so which states vote matters as much as the national mood.
- Markets and polls answer different questions. Polls measure support today; markets price who will win. See prediction markets vs. polls.
- Long shots are overpriced. A candidate at 5% usually wins less often than that; see the favorite-longshot bias.
FAQ
When are the 2026 midterm elections?
Tuesday, November 3, 2026. Voters elect all 435 members of the House and 35 of the 100 senators, including special elections in Ohio and Florida.
Who is favored to win the House?
As of Oct 27, 2026, the prediction markets price a Democratic House at: Kalshi at 91% and Polymarket at 92%. These are market prices, not guarantees, and they can move quickly in the final weeks.
Who is favored to win the Senate?
As of Oct 27, 2026, the prediction markets price a Democratic Senate at: Kalshi at 64% and Polymarket at 66%. These are market prices, not guarantees, and they can move quickly in the final weeks.
What is the Sikt Midterm Bench?
Sikt Intelligence's live test on the 2026 midterms: our AI superforecaster forecasts the key Senate races ahead of election day, and its forecasts are published next to the market odds and scored against the results.
Are prediction markets accurate on elections?
Often, and they have beaten polls in many studies, but not always: they can be confidently wrong, and long shots tend to be overpriced.
Market data from the public Kalshi and Polymarket APIs, shown for information only. Prices are midpoints or last trades and can be several hours old. Sikt Intelligence is not affiliated with Kalshi or Polymarket. Not financial or investment advice.