With the 2026 midterm election only days away, a new national poll shows Democrats holding an advantage on the congressional “generic ballot,” while the Republican campaign works to protect majorities in both chambers. The Reuters/Ipsos poll found that likely voters preferred a Democratic candidate in their district over a Republican candidate by 53% to 44% when undecided voters who leaned toward a party were included.
That result, released Oct. 26, is a meaningful snapshot of the final campaign stretch. It is not a seat-by-seat forecast. House and Senate control will be decided in individual states and districts, where candidate quality, turnout, incumbency, geography and the particular Senate map can produce results that differ from the national mood.
Facts: What the latest 2026 midterm polling shows
The Reuters/Ipsos Core Political survey was conducted online from Oct. 22 through Oct. 26 among 1,679 U.S. adults, including 1,280 registered voters, according to Reuters. Ipsos’s official poll summary describes the likely-voter result as a 9-point Democratic edge: 53% said they would vote for or lean toward the Democratic candidate, compared with 44% for the Republican candidate.
The underlying topline and methodology document reports a registered-voter sample of 1,279 and a margin of error of about 2.8 percentage points for that group. The exact registered-voter total differs by one between the Reuters story and the Ipsos document, a small reporting discrepancy that does not change the published topline. Poll margins of error also do not account for every possible source of error, including nonresponse or changes in the electorate between the survey and Election Day.
Before leaning voters were counted, registered voters favored Democrats 47% to 36% on the generic ballot, while 11% said they did not know which party they would support. That gap was one point narrower than in the Reuters/Ipsos poll that ended Oct. 19. Including leaners produced the 53%-44% result.
The same survey put President Donald Trump’s job approval at 32% among all adults and 34% among registered voters, with 66% of adults disapproving. Reuters reported that the overall figure was one point higher than the previous week’s record low in its series. The poll also found that 73% of voters said their cost of living was moving in the wrong direction, and that Democrats held a 37%-33% advantage over Republicans on which party would handle the economy better.
The poll’s campaign context is clear: voters were weighing affordability, and the parties were trying to turn that concern into a vote for their own candidates. Reuters reported that Democrats were generally viewed as having a strong chance to win the House, while analysts saw a steeper path for Democrats in the Senate.
Analysis: Why the national lead matters—and what it cannot show
A generic-ballot lead gives campaigns and observers a measure of national party preference. It can signal which party has the stronger environment and help explain why one side is spending more aggressively in competitive races. It can also shape candidate messaging in the final days, especially when campaigns argue that a vote is part of a broader effort to control Congress.
But the measure does not translate directly into a number of seats. House contests are conducted district by district. Senate contests are statewide, and the 2026 map may require Democrats to win races in states with different partisan histories and local issues. A national polling lead can therefore coexist with a narrow Senate path or with uncertainty in a handful of decisive races.
Turnout is another unresolved variable. The Reuters/Ipsos survey found that likely voters favored Democrats by 9 points, 53% to 44%, but turnout models are estimates rather than vote counts. In the official Ipsos topline, 66% of registered voters said they were almost certain or very likely to vote, while Democrats were somewhat more likely than Republicans to say they were very enthusiastic about voting. Those measures indicate a possible Democratic enthusiasm advantage, but they do not establish who will actually cast a ballot.
The timing of the survey also matters. Interviews ended Oct. 26, and the election is scheduled for Nov. 3. Late events, advertising, candidate appearances, early voting and breaking news can change preferences or turnout after the field period closes. The poll should therefore be read as evidence about the race at the end of October, not as a final result.
Campaign status: Both parties still face a path to control
Republicans entered the final week trying to retain their House and Senate majorities. Democrats saw the national environment as an opening, but the polling did not remove the structural and geographic challenges involved in winning the Senate. The Reuters account described control of Congress as consequential for Trump’s agenda during the remaining two years of his second term.
For voters, the practical question is local: which candidates are on the ballot, what issues dominate in the state or district, and who turns out. The national poll supplies context for those contests, but it cannot replace state-level surveys or certified vote totals.
The most defensible conclusion as of Oct. 31 is limited but important. Democrats had a measurable national advantage in the latest Reuters/Ipsos likely-voter ballot test, and Trump’s approval remained below a majority. Those findings suggested a challenging environment for Republicans. They did not prove that Democrats would win either chamber, and the final outcome remained uncertain until votes were counted.
Sources
- Reuters, “Trump approval rises from record low to 32% ahead of midterm election, Reuters/Ipsos poll finds” — Oct. 26, 2026 reporting on the poll, campaign context and limitations of the generic ballot.
- Ipsos, “Latest U.S. opinion polls” — official Oct. 26, 2026 summary of the Reuters/Ipsos likely-voter result.
- Ipsos, “Reuters/Ipsos mid-October Survey Topline & Methodology” — official questionnaire, field dates, sample sizes and margins of error.
