Will Democrats win control of the U.S. House in the 2026 midterm elections?
As of September 26, 2026, we put the chance that Democrats win the House at 88 percent. They need a net gain of only three seats, lead the generic ballot by about eight points, and face a president near 33 percent approval. New Republican-drawn maps are the main reason this is not higher.
Why it matters: The House majority decides which party sets the investigative agenda, writes the spending bills that avert or trigger shutdowns, and can block or pass the president's priorities for the final two years of his term.
TPG forecast
88%
Very likely
Democratic House
Version 1: 88% (range 80% to 94%) · Kalshi 92%
Evidence cutoff Sep 26, 2026 · published Sep 26, 2026 · Very likely Democratic House
This forecast is on its first version. Each update adds a point; old versions are never edited.
The numbers behind the call
3
net seats Democrats need to win the House majority[1]
D+8.1
Silver Bulletin generic ballot average, late September[3]
33%
Trump job approval in the September Verasight poll, a new low[5]
18
of the 20 midterms since 1946 cost the president's party House seats[4]
$224M vs $174M
Democratic vs Republican House super PAC ad reservations[2]
5.6 pts
average shift toward Democrats in state legislative special elections since 2025[8]
What it means for you
Taxpayers and families
A Democratic House would make further party-line tax and spending packages through reconciliation very unlikely before 2029, freezing current tax law in place unless both parties agree. A Republican hold keeps that path open.
People who rely on health coverage
A Democratic majority would control the House Appropriations and Energy and Commerce agendas, making cuts to Medicaid or marketplace subsidies much harder to pass. A Republican hold leaves those programs open to further changes.
Federal workers and contractors
Divided government raises the odds of hard budget standoffs, because every funding bill would need a Democratic House and a Republican Senate and White House to agree. Shutdown risk rises in fiscal 2028 talks.
The administration
A Democratic House gains subpoena power over executive agencies, the Iran war, tariffs and spending, and could open impeachment inquiries. A Republican hold keeps oversight in friendly hands.
Businesses and markets
Gridlock usually means fewer large legislative surprises; investors would price in slower policy change and more fights over the debt limit and appropriations.
The read
Three lenses, always in this order. How the method works.
01
Psychology
What do the decision makers need?
Midterms are referendums on the party in the White House. Voters who are unhappy with prices, gas costs tied to the Iran war and the direction of the country have one lever in November: the House ballot. That thermostatic reflex shows up in the enthusiasm gap. In the Strength In Numbers/Verasight poll, Democrats lead by 9 among registered voters but by 12 among those who say they will definitely vote, which is the signature of a motivated out party.
Behavior on the public record points the same way. A record number of House members are leaving, and NPR counts close to 100 lawmakers from this Congress who will not return. Members of a majority that expects to keep power rarely walk away in such numbers. Republicans in tight races have also started breaking with the president on the Iran war and gas prices, which is what candidates do when they believe the national brand is a liability rather than an asset.
02
History
What happened the last times this came up?
The reference class is every midterm since 1946. The president's party lost House seats in 18 of those 20 elections. The two exceptions, 1998 and 2002, came with approval ratings in the 60s. In every midterm since 1950 where the president's late August Gallup approval was under 50 percent, his party lost House seats, and the median loss was about 29 seats, according to the American Presidency Project table.
The closest analogue is 2018, Trump's first midterm, when Democrats gained 41 seats with a generic ballot lead of roughly the size seen today. The analogy breaks on the map. Mid-decade redistricting in Texas, Missouri, North Carolina, Ohio, Florida and other states has added Republican-leaning seats, and Wikipedia's tally of the new maps nets about 12 Republican seats against 1 Democratic seat. That raises the bar for Democrats, though analyst Kyle Kondik judged the overall seat bias similar to 2018's.
03
Statistics
Base rate, adjustments, the number.
Start from the base rate: the president's party lost House seats in 90 percent of postwar midterms. Democrats need a net gain of three, so nearly any loss of that size flips the chamber. Adjust up for conditions: approval in the 30s matches the weakest readings in the postwar record, 1946 and 1950, and the Silver Bulletin generic ballot average is D+8.1, which alone implies a Democratic popular vote win of historic size. That pushes the raw number to about 95 percent.
Adjust down for structure. New Republican-drawn maps may cost Democrats several seats relative to 2024 lines, and Cook still rates 23 races as Toss Ups, 17 of them Republican held, with Democrats favored in 208 and Republicans in 205. Subtracting about 7 points for redistricting and polling error lands at 88 percent. The Kalshi market at 92 percent is slightly more confident; we hold a bit more room for a late rally tied to an Iran ceasefire or falling gas prices.
Between the lines
What people spent versus what they said, and what was left out.
Costly signal
Democratic-aligned House Majority PAC has $224 million in ad reservations against $174 million for the Republican Congressional Leadership Fund.
Reservations are real money committed to specific markets. A $50 million edge for the out party is unusual and means Democrats can contest Republican seats rather than just defend their own.
Personnel move
Record retirements and primary defeats mean close to 100 members from this Congress will not return.
Retirements are the most honest poll in Washington. Members with inside data are choosing not to face this electorate, and open seats are easier to flip than incumbent-held ones.
Tone shift
Republicans in competitive races, including Senate nominees, are publicly calling for the Iran war to end as gas prices climb.
Distancing from the party leader is a costly move inside a primary-dominated party. Candidates do it only when their own polling shows the national message is sinking them.
What was left out
The president has pledged $5,000 to every U.S. adult if his party wins the midterms.
When the majority party stops running on its record and offers a conditional payment instead, it is signaling that the record is not selling.
Our number after adjusting the base rate for what is different this time: 88%. It sits inside the historical interval.
Check our statistics yourself
- Reference class
- 18 of 20 (90%)
- Exact 95% CI (Clopper-Pearson)
- 68.3 to 98.8%
- Bayesian 95% credible (uniform prior)
- 69.6 to 97.0%
- Binomial test of our 88%
- p = 1.000
Our 88% sits inside the historical interval: the forecast is consistent with how cases like this have gone.
Reproduce in JASP (free, jasp-stats.org)
- Download base-rate.csv (one row per case in the reference class, outcome 1 or 0) and open it in JASP.
- Frequencies, then Binomial Test. Variable: outcome. Test value: 0.88. Tick Confidence interval. JASP reports the same p value and Clopper-Pearson interval shown here.
- Frequencies, then Bayesian Binomial Test, with a Beta(1, 1) prior. The posterior 95% credible interval matches ours.
Reference class: The Conversation's count of the 20 midterms since 1946, cross-checked against the American Presidency Project table of seat changes and Gallup approval by midterm year.
Base rate
90%
of postwar midterms (1946 to 2022) in which the president's party lost House seats[4][7]
The Conversation's count of the 20 midterms since 1946, cross-checked against the American Presidency Project table of seat changes and Gallup approval by midterm year. (n = 20)
Closest historical parallels
Trump's first midterm (2018)
Alike: Same president, approval in the high 30s to low 40s, a large generic ballot deficit and a wave of Republican retirements.
Outcome: Democrats gained 41 seats and the majority.
Where it breaks: The 2026 maps are more Republican after mid-decade redistricting, so the same vote margin buys Democrats fewer seats.
Bush's second midterm (2006)
Alike: An unpopular war abroad, a president in the high 30s and a majority weighed down by its own record.
Outcome: Democrats gained 30 seats and took the House.
Where it breaks: In 2006 Republicans held a large cushion; today Democrats need only three seats, which makes a flip easier than in 2006.
Clinton's first midterm (1994)
Alike: A president's party with a thin record on the top voter concern facing an energized opposition.
Outcome: Republicans gained 54 seats.
Where it breaks: 1994 involved a long-term realignment in the South; 2026 is a referendum without that regional sorting underneath it.
How it could play out
Every path we see, sized by probability. Paths that resolve YES add up to our 88%.
- Working majority: Democrats gain 3 to 14 seats Yes48%
Republican-drawn maps hold several seats, but Democrats win most of the 23 Toss Ups, matching Cook's 220 to 230 seat range.
- Wave: Democrats gain 15 or more seats Yes40%
The generic ballot stays near D+8 or widens as gas prices stay high and the Iran war continues into November.
- Republicans hold narrowly No12%
A late rally, such as an Iran ceasefire and falling fuel prices, plus a polling miss and redistricting gains let Republicans keep 218 or more.
What would change our number
Generic ballot average falls below D+4 by mid October
-10 pointsCongressional Leadership Fund cancels reservations in Republican-held Toss Up seats to protect Lean Republican seats
+4 pointsAn Iran ceasefire with national average gas prices falling sharply before early voting peaks
-6 pointsCook moves five or more additional Republican-held seats toward Democrats
+3 pointsTPG 88% vs Kalshi 92%
We are four points below the market. Traders price the national environment heavily; we keep extra weight on the new Republican-drawn maps and on October surprises tied to the Iran war, which could move gas prices quickly in either direction.
Market price recorded September 26, 2026 from Kalshi. Analysis, not investment advice.
How this question resolves
Resolves YES if Democratic candidates win at least 218 of the 435 U.S. House seats in the November 3, 2026 general election (including any later runoffs), as called by the Associated Press or shown in certified state results. Resolves NO if Republicans win 218 or more seats, or if Democrats finish below 218.
Closes: January 3, 2027. Judged by: Associated Press race calls and certified state results, confirmed by the party breakdown reported by the Clerk of the U.S. House when the 120th Congress convenes
Questions people ask
Who will win the House in 2026?
+
As of September 26, 2026, Democrats are strong favorites to win the House, with an 88 percent chance in our forecast. They need a net gain of only three seats, lead the generic congressional ballot by about eight points, and face a president with approval in the 30s. New Republican-drawn maps in several states are the main thing keeping Republicans in the race.
How many seats do Democrats need to flip the House in 2026?
+
Democrats need a net gain of three seats to reach the 218-seat majority, according to the Cook Political Report's September 2026 analysis. Cook rates 23 races as Toss Ups, 17 of them held by Republicans, and expects Democrats to gain between five and 15 seats, which would give them 220 to 230 seats.
Does the president's party usually lose seats in the midterms?
+
Yes. Since 1946 the president's party has lost House seats in 18 of 20 midterm elections. The only exceptions, 1998 and 2002, came when the president's approval was in the 60s. In every midterm since 1950 with the president's late August approval under 50 percent, his party lost House seats, with a median loss of about 29.
Did redistricting help Republicans keep the House in 2026?
+
Mid-decade redistricting helped Republicans on paper. Texas, Missouri, North Carolina, Ohio and other states drew new maps that add Republican-leaning seats, partly offset by California's new Democratic map. The gain raises the bar for Democrats, but with a generic ballot lead near eight points, most forecasters still expect Democrats to win the majority.
Sources
Every source is dated on or before the September 26, 2026 evidence cutoff. Nothing published later was used.
- [1]Cook Political Report projects Democrats will take control of the House as races shift. Washington Examiner, Sep 25, 2026.
“Democrats currently hold 215 House seats and need a net gain of just three seats to secure a majority.”
- [2]Cook Political Report projects Democrats will take control of the House as races shift. Washington Examiner, Sep 25, 2026.
“The Congressional Leadership Fund has $174 million reserved, while the Democratic-aligned House Majority PAC has $224 million reserved.”
- [3]Generic Congressional Ballot: Latest Polls. Silver Bulletin, Sep 25, 2026.
“It applies a likely voter adjustment, which in this case benefits Democrats and moves their lead from D +8.1 to D +8.8.”
- [4]For 80 years, the president's party has almost always lost House seats in midterm elections. The Conversation, Jan 19, 2026.
“In 18 of them, the president’s party lost seats in the House of Representatives.”
- [5]Poll: Trump's approval hits a new low of 33% as Republicans sour. Strength In Numbers (G. Elliott Morris), Sep 23, 2026.
“In our new poll, 33% of Americans approve of the job Donald Trump is doing as president and 63% disapprove”
- [6]2025 to 2026 United States redistricting. Wikipedia, Sep 26, 2026.
“Net 12 Republican and 1 Democratic seats established.”
- [7]Seats in Congress Gained/Lost by the President's Party in Mid-Term Elections. The American Presidency Project, UC Santa Barbara, Jul 13, 2026.
“Beginning with Obama, job approval is the average job approval during the noted half month period.”
- [8]State legislative special elections show shift toward Democrats compared to previous regular elections. Ballotpedia News, Feb 26, 2026.
“In the state legislative special elections that have happened since January 2025, there was an average shift of 5.6 percentage points towards Democrats.”
- [9]Who in Congress is not running for reelection in 2026?. NPR, Sep 16, 2026.
“close to 100 lawmakers who started this Congress will not be around for the next one”
Version history and integrity
Published forecasts are never edited. When the evidence changes we publish a new version beside the old one, each locked with a sha256 hash of its full contents. When this question resolves it is scored in public on our track record.
| Version | Cutoff | Forecast | Lock hash |
|---|---|---|---|
| v1 | Sep 26, 2026 | 88% Very likely | 735a5ac40eeea15fa48fadcc599846548f2b3462e39f2ea8b57f08928d6e5a35 |
Download every forecast: JSON or CSV (CC BY 4.0). Public sources only. Analysis, not investment or legal advice.
Related forecasts
Will gas prices be above $4.50 a gallon on Election Day 2026?
33%
Unlikely
Will Democrat James Talarico defeat Ken Paxton in the 2026 Texas U.S. Senate race?
52%
Toss-up
Will President Trump invoke the Insurrection Act to deploy troops inside the United States before January 1, 2027?
8%
Very unlikely
Will Vivek Ramaswamy win the 2026 Ohio governor's race?
45%
Toss-up