New Reuters/Ipsos Poll: Democrats Lead 53-44 on Generic Ballot as Trump Approval Stays at 32%

Introduction

One week before Election Day, a new national poll has given Democrats their clearest national advantage of the midterm campaign so far. The October Reuters/Ipsos Core Political survey, conducted October 22-26 and analyzed by Ipsos on October 26, 2026, finds Democrats leading Republicans by nine points among likely voters on the generic congressional ballot: 53 percent of likely voters say they would vote for or lean toward the Democratic candidate in their district, compared with 44 percent who say they would vote for or lean toward the Republican candidate.

The same survey puts President Donald Trump's overall job approval at 32 percent among all Americans, which Ipsos describes as remaining at a near low for his presidency. His ratings on the economy are lower still: just 25 percent of Americans approve of how he is handling the economy, and only 20 percent approve of his handling of the cost of living.

The poll does not decide any race. A generic ballot asks voters which party's candidate they would support in their own district, without naming the actual candidates running there, and control of Congress is decided district by district and state by state. But as a measure of the national mood one week out, the survey captures the two forces that have defined the closing stretch of this campaign: deep frustration over affordability, and a Democratic electorate that reports more enthusiasm about voting than the Republican electorate.

This article explains what the new Reuters/Ipsos poll actually found, how it was conducted, how it compares with earlier polling in this cycle, what it does and does not predict about control of the House and Senate, and what both parties are watching in the final seven days before November 3.

The headline numbers

The headline result is straightforward. Among likely voters — registered voters who say they intend to vote or have already voted in the midterm election — the Democratic candidate leads the Republican candidate 53 percent to 44 percent on the generic congressional ballot, a nine-point margin.

Ipsos published the finding on October 26, 2026, based on its analysis of the October Reuters/Ipsos Core Political survey. The survey was conducted October 22-26, 2026, using Ipsos's probability-based KnowledgePanel. The full sample was 1,679 adults aged 18 or older, nationally representative. Reuters's own reporting of the same survey noted that the sample included 1,280 registered voters.

The margin of sampling error for the full adult sample is plus or minus 2.4 percentage points at the 95 percent confidence level, according to Ipsos, which notes that the margin is higher for subgroups. Percentages are rounded to the nearest whole number, so columns may total slightly more or less than 100 percent.

Two details about the headline matter for how it should be read. First, the 53-44 figure includes voters who lean toward a party's candidate, not only voters who firmly choose that candidate. Ipsos's wording is that likely voters "would vote for or lean to" the candidate of the party they name. Second, the likely-voter population is a modeled group, not a count of ballots cast. Ipsos weights respondents who are registered and express an intention to vote, or say they have already voted, to targets drawn from the L2 national registered voter file and the 2022 Current Population Survey Voting and Registration Supplement, using age, gender, region, race and ethnicity, education, and county voting history from 2016, 2020 and 2024.

That modeling is standard polling practice, but it is worth stating plainly: a likely-voter screen is an estimate of who will vote, and it can miss late shifts in turnout. The poll is a snapshot of opinion in late October, not a forecast of the final vote.

A shift on the economy

The most striking change in the survey is not the ballot margin itself but the issue underneath it. For much of the Trump presidency, Republicans held an advantage on the question of which party handles the economy better. That advantage has now reversed, according to Ipsos.

In the new survey, 34 percent of Americans prefer the Democratic Party on the economy, compared with 28 percent who prefer the Republican Party. Ipsos notes that this is a major change from January 2025, when the Republican Party led on the economy by 17 points. Reuters's reporting of the same survey, using a voter-based cut of the data, put the economy-trust figures at 37 percent for Democrats and 33 percent for Republicans. The two sets of numbers describe different populations in the same survey — all Americans in the Ipsos release, voters in the Reuters report — and both point in the same direction: Democrats now lead on an issue Republicans long owned.

The president's own economic ratings help explain why. Only 25 percent of Americans approve of Trump's handling of the economy, and only 20 percent approve of his handling of the cost of living. Reuters reported that 73 percent of voters in the poll said their cost of living is heading in the wrong direction.

The context, as reported by Reuters, is the surge in inflation and fuel prices since the United States ordered military strikes on Iran in February, which sent gasoline prices higher. Before that war began, Reuters/Ipsos polling showed voters preferring Republicans over Democrats as the better stewards of the economy. The latest survey shows that preference has flipped, and the flip tracks closely with the affordability squeeze voters report in their own household budgets.

That squeeze is also visible inside the Republican base. Among Republicans in the survey, 62 percent approve of the president's handling of the economy, but just 48 percent approve of his handling of the cost of living. Republicans are now more likely to say they approve of his job performance "somewhat" (42 percent) than "strongly" (34 percent), a change from the start of his second term, when strong approval was higher. Ipsos presents this softening — strong approval giving way to softer approval — as one of the clearest signs of strain within the president's own coalition.

Enthusiasm: Democrats more eager to vote

The survey also finds an enthusiasm gap that could matter as much as the ballot margin. Reuters reported that 46 percent of Democrats said they were very enthusiastic about voting in the midterms, compared with 30 percent of Republicans. Ipsos, in its own release, made the same point in general terms: as has been the case throughout the fall, Democrats continue to express more strong enthusiasm than Republicans toward voting in this fall's midterm elections.

Enthusiasm is not votes, and a 16-point enthusiasm gap does not translate mechanically into a 16-point turnout gap. But in a midterm election, where turnout is typically lower than in a presidential year and where the composition of the electorate can swing a close race, which side's voters are more motivated to show up is one of the most watched indicators in the final week.

The enthusiasm finding also fits with what early voting data has shown so far. More than 31 million ballots had already been cast by October 27, according to CNN's analysis reported earlier in this campaign, and Democrats have led in party registration among early voters in states that report that data. Early-vote data shows only how voters are registered, not how they voted, and the early pool tends to skew older and less Republican than the full electorate — so it cannot be read as a result. But the direction of the enthusiasm gap and the direction of the early-vote registration data are consistent with each other.

How this poll compares with earlier polling

The new survey is an update, not a first. Earlier in this cycle, a Reuters/Ipsos poll found Democrats leading Republicans 47 percent to 35 percent among registered voters on the generic ballot, with Trump's approval at a then-record-low 31 percent. The new poll's likely-voter figure of 53-44 and its approval figure of 32 percent describe a different population and a later moment, so the two results should not be compared as if they were the same question asked to the same group.

What can be said is that the direction has been stable: Democrats have led on the generic ballot in Reuters/Ipsos polling throughout the fall, and Trump's approval has stayed in the low thirties. The new survey's approval figure of 32 percent is described by Reuters as a rise from the record low, and by Ipsos as remaining at a near low for his presidency. Both descriptions are consistent with a one-point move inside the margin of error: essentially flat, at a historically weak level for a president one week before a midterm election.

Other national averages point the same way. RealClearPolling's generic ballot average, cited in recent forecasts, has put Democrats ahead by about 8.3 points. That average blends many pollsters and many populations, while the Reuters/Ipsos figure is a single survey with a single methodology. The two should not be averaged together, but the fact that they land in the same range — a high-single-digit Democratic lead nationally — is why forecasters have been comfortable describing the national environment as Democratic-leaning.

The national environment, however, is not the election. House control is decided in 435 separate districts, and Senate control in a handful of competitive states. A nine-point national lead is consistent with a Democratic House majority, but it does not guarantee one, and it says even less about the Senate, where Democrats must gain at least four seats to win control of a chamber Republicans now lead 53-47.

What the poll does not predict

This section matters because generic-ballot polls are routinely over-read in the final week of a campaign. The Reuters/Ipsos survey measures national preferences for Congress. It does not identify winners in individual districts or states, and it should not be translated directly into a seat count.

There are three reasons for that caution, all grounded in how American elections work.

First, the House is not a national election. A party can win the national popular vote for the House and still fall short of 218 seats if its votes are concentrated in districts it already wins by large margins, while the other party's votes are spread more efficiently across competitive districts. Forecasters who do translate polls into seats use district-level polling, fundraising, candidate quality and historical patterns — not the national generic ballot alone.

Second, the Senate map is even less national. Only a third of Senate seats are on the ballot, and control will be decided in states such as Texas, Maine, Alaska, Iowa, Michigan, Ohio, Kansas and North Carolina. A national survey cannot capture the local dynamics of those races: a popular incumbent, a controversial challenger, a late advertising blitz, or a state-specific issue such as the cost of fuel or the price of groceries in a particular region.

Third, the likely-voter model itself carries uncertainty. Ipsos's model weights respondents to 2022 turnout patterns and to county voting history, but midterm electorates change from cycle to cycle, and late-deciding voters — who tend to break against the president's party in midterms, but not always — can move a close race in the final days.

Forecasters, for their part, have been more bullish than the poll alone would suggest. Decision Desk HQ has given Democrats a 79 percent chance of winning the House and a 57 percent chance of winning the Senate. Prediction markets have gone further: Polymarket has priced Democrats at 93 percent for the House and 65 percent for the Senate, and Kalshi at 90 percent and 64 percent respectively. Those market prices are not polls; they reflect traders wagering on outcomes, and they can move quickly on new polls, news and campaign developments. They are reported here as context, not as predictions endorsed by this article.

The state races behind the national numbers

While the national poll gives Democrats a lead, the Senate races that will decide control remain close in the states where they are being fought.

In Texas, the Univision/YouGov poll released October 27 found Democrat James Talarico leading Republican Ken Paxton 50 percent to 45 percent, with Republican Governor Greg Abbott leading Democrat Gina Hinojosa 48 percent to 46 percent. In Michigan, recent polls have put Democrat Abdul El-Sayed narrowly ahead of Republican Mike Rogers, with the RealClearPolitics average at El-Sayed plus 1.9 points. In Georgia, Democrat Jon Ossoff has led Republican Mike Collins in every recent survey, by margins ranging from about seven to eight points. In Ohio, Republican Jon Husted has led in some surveys while the race remains competitive.

Those state numbers are reported in separate coverage on this blog, and they illustrate the central tension of the final week: a Democratic-leaning national environment coexisting with a set of Senate races that are individually close enough that either party could still win the chamber. The Reuters/Ipsos poll does not resolve that tension. It measures the environment; the states will decide the outcome.

The House picture is clearer but not settled. The Cook Political Report's October 23 update moved 19 House races toward Democrats and none toward Republicans, and described Democrats as "favored outright" to win the 218 seats needed for a majority, projecting a gain of between 12 and 22 seats. Decision Desk HQ's October 26 forecast projected 238 Democratic seats. Those forecasts rest on district-level data that a national poll cannot supply, and they remain probabilities, not results.

What to watch in the final week

Five things will determine whether the Reuters/Ipsos snapshot holds through Election Day.

First, turnout. The enthusiasm gap favors Democrats, but Republicans have a large late advertising operation — more than $277 million in a final-six-weeks blitz from the Trump-aligned MAGA Inc super PAC, according to Federal Election Commission data reported by Reuters — aimed at closing that gap in the states that matter most.

Second, the final polls. More surveys will land before November 3, including state polls in Texas, Michigan, Ohio and Maine. Watch whether they confirm the national picture or show a late tightening, which is common in the final days of a campaign as undecided voters make up their minds.

Third, early voting totals. More than 31 million ballots had been cast by October 27. The final early-vote totals, and the party registration of those voters in states that report it, will give a partial read on the electorate — partial, because registration is not a vote, and because Election Day voting still lies ahead.

Fourth, the president's closing message. Trump has spent the final week barnstorming battleground states, including Texas on October 27, with a message that the midterms are a referendum on his presidency — a framing he has embraced explicitly, telling a Port Huron, Michigan, crowd on October 25 to "pretend that I'm on the ballot." Former President Barack Obama, campaigning for Democrats in Michigan, has framed the same election as a chance to say "enough is enough" after what he called "10 years of chaos and cruelty and corruption and incompetence." Both parties are now making the same closing argument in opposite directions: this election is about Trump.

Fifth, late events. A major news development in the final days — on the economy, on foreign policy, or on election administration — can move a close race. The Reuters/Ipsos poll was conducted October 22-26, before the final week's rallies and before any late-October surprises. It captures the mood at the start of the final week, not the mood on Election Day itself.

What's next

Election Day is Tuesday, November 3, 2026. Early voting continues in most states through the end of October, with deadlines varying by state. In Texas, early voting runs through October 30. In North Carolina, in-person early voting ends October 31. In Florida, mandatory early voting runs through October 31, with some counties having opened as early as October 19.

Voters should check their own state's deadline and polling-place information through their state election authority or local election office, rather than relying on national summaries, because rules on mail ballots, ballot return deadlines and voter identification vary widely.

For the campaigns, the final week is a turnout operation. Democrats will try to convert a national lead and an enthusiasm advantage into ballots in the House districts and Senate states that decide control. Republicans will try to use a massive late spending advantage and the president's rally schedule to close the gap and to persuade late-deciding voters that the cost-of-living squeeze is a reason to stay the course rather than change direction.

The Reuters/Ipsos poll suggests the first task is easier than the second — but a poll is not a vote, and the only number that will matter on November 4 is the count.

Frequently asked questions

What does the Reuters/Ipsos poll actually show?

It shows Democrats leading Republicans 53 percent to 44 percent among likely voters on the generic congressional ballot, in a survey conducted October 22-26, 2026. It also shows President Trump's approval at 32 percent among all Americans, with 25 percent approving of his handling of the economy and 20 percent approving of his handling of the cost of living.

Is the generic ballot the same as a prediction of who will win the House?

No. The generic ballot asks voters which party's candidate they would support in their district, without naming candidates. House control is decided in 435 separate districts. Forecasters use district-level polling, fundraising and historical patterns to project seats; the generic ballot alone cannot do that.

How was the poll conducted?

Ipsos conducted the survey October 22-26, 2026, using its probability-based KnowledgePanel, with a nationally representative sample of 1,679 adults aged 18 or older. The margin of sampling error for the full sample is plus or minus 2.4 percentage points at the 95 percent confidence level. The likely-voter figures weight registered voters who intend to vote or have already voted to turnout targets from the L2 voter file and the 2022 Current Population Survey.

Why do different reports of the same poll show different numbers?

Because they use different populations. Ipsos's release reports economy-trust figures of 34 percent for Democrats and 28 percent for Republicans among all Americans, while Reuters's reporting of the same survey put those figures at 37 percent and 33 percent among voters. The headline ballot figure of 53-44 is among likely voters in both accounts. Always check which population a number describes before comparing it with another.

Does a nine-point national lead guarantee Democrats will win Congress?

No. A national lead is consistent with a Democratic House majority, but the Senate is decided in a handful of close states, and a national survey cannot capture those races. Democrats need to gain at least four Senate seats to win control of a chamber Republicans lead 53-47. A poll is a snapshot, not a result.

When is Election Day, and when will results be known?

Election Day is Tuesday, November 3, 2026. Some races may be called on election night, but close races — especially in states with large mail-ballot counts — may take days to resolve. Early and mail ballots already cast will be counted under each state's own rules and timelines.

Sources

Ipsos, "Democrats hold 9-point edge on midterm congressional ballot," October 26, 2026 (Reuters/Ipsos Core Political survey, October 22-26, 2026).

Reuters, "Trump approval rises from record low to 32% ahead of midterm election, Reuters/Ipsos poll finds," October 27, 2026 (syndicated reporting of the same survey).

Reuters/Ipsos polling history and RealClearPolling generic ballot average, as cited in recent midterm forecasts.

Decision Desk HQ, Cook Political Report and prediction-market figures, as reported in October 26-27, 2026 forecasts.