Latest UK Opinion Polls

The latest UK polls are shown in the live voting intention table and chart at the top of this page, which update as new polls are published. It lists every GB Westminster poll from YouGov, Opinium, Survation, More in Common, Find Out Now and the other major UK pollsters, with a 7-poll moving average of the most recent seven.

The current leader is shown at the top of the 7-poll moving average table and chart on this page, which update as new polls are released. The average uses the seven most recent GB Westminster polls from YouGov, Opinium, Survation and other major UK pollsters, so one unusual poll cannot change the leader on its own.

If there was a general election tomorrow, the winner would be decided seat by seat under first-past-the-post, not by national vote share alone. PollCheck translates the current 7-poll average into a constituency-level seat projection, updated as new polls are published, showing which party would be largest and whether it would reach the 326 seats needed for a Commons majority. The live projection is on the PollCheck homepage and Seat Calculator.

A 7-poll moving average, sometimes called a poll of polls, is the average of the seven most recent polls. It evens out the noise in individual polls, so one unusual poll moves it less and the trend is easier to read.

Restore Britain are tracked on the chart and in the table above, with their own line in the 7-poll moving average. Not every pollster asks about them, and those that do usually ask in two stages: Restore Britain are not on the main list of parties, so a respondent picks "another party" first and is then shown Restore Britain as an option. YouGov, Ipsos and Survation all ask this way. YouGov have tested putting Restore Britain on the main list instead and found similar levels of support either way.

Pollsters who do not ask about them fold their support into Others, so the average here is built from the seven most recent polls that named them.

Every poll that has named them, plus pollster coverage, is on our Restore Britain polling page.

UK polls show vote share but due to the first-past-the-post system, seats don't directly match vote percentages. PollCheck projects seats with a constituency-level demographic model built from British Election Study survey data, the Census and each seat's 2024 General Election result, matched to the polling average. Try our Seat Calculator for full constituency-level projections.

Poll accuracy varies depending on methodology, sample size, and timing. The 30 most recent GB voting intention polls (as of 21 October 2026, excluding MRPs) sampled between 1,031 and 3,069 people. At 95% confidence the margin of error on a party's share is at most 1.96 x the square root of (0.25 / sample size): about plus or minus 3.1 points for 1,000 people, 2.2 for 2,000 and 1.8 for 3,000. Aggregate polling (like 7-poll averages) tends to be more accurate than individual polls. Historical accuracy can be viewed on our Pollster Check page.

Pollsters use different methods - how they contact people, how they weight their samples, which smaller parties they prompt for. These choices mean some pollsters consistently show a party a point or two higher or lower than others do. That consistent gap is called a house effect. It doesn't mean a pollster is wrong: nobody knows which method is closest to the truth until an election happens.

PollCheck calculates it by comparing every poll with the average of the seven polls published closest to it in time. That way each poll is judged against what other pollsters were showing at the same time, not against a whole-year average - important because support moves over a year, and a pollster that mostly polled early or late in the period would otherwise look biased when it wasn't. A pollster's house effect is the average of those gaps across all its polls from the last 12 months, adjusted so the effects across all pollsters average out to zero. Only pollsters with at least five GB-wide polls in the period are shown; MRPs and UK-wide polls are excluded.

The margin of error tells you how much a poll result might differ from the true population value due to random sampling. For example, if a party polls at 30% with a MOE of plus or minus 3%, the true support likely falls between 27% and 33%.

Larger samples produce smaller margins of error. A poll of 1,000 people has a MOE of about plus or minus 3.1%, while a poll of 2,000 has a MOE of about plus or minus 2.2%.

Major UK pollsters release new polls several times per week. PollCheck updates its data when new polls are published. The 7-poll moving average is automatically recalculated as new polling data becomes available.

In PollCheck's 7-poll moving average of pollster crosstabs to 19 October 2026, Labour lead among voters aged 18-24 on 31.1%, with the Greens second on 26.9% and Reform UK fourth on 13.7%. Conservative support is lowest in this age band.

See the Demographic Breakdown section above for the latest 7-poll moving average by age, drawn from sample-weighted crosstabs across major UK pollsters.

Pollster crosstabs consistently show the Conservatives and Reform UK leading among voters aged 65 and over, with Labour and the Greens polling weakest in this age band.

The age-group breakdown is in the Demographic Breakdown section above, updated as new polls are published and drawn from the 7-poll moving average of pollster crosstabs.

PollCheck aggregates regional crosstabs from major pollsters into a sample-weighted 7-poll moving average across the 11 GB regions. In the regional averages to mid-October 2026, Reform UK lead in four of the eight English regions outside London (East of England, East Midlands, Yorkshire and the Humber, and the South West, the last two by under a point), Labour lead in London, the North West, the North East and the West Midlands, and the Conservatives lead in the South East. Scotland is dominated by the SNP and Plaid Cymru polls strongly in Wales.

See the Demographic Breakdown section above for current regional standings.

Regional figures use each pollster's reported voting-intention breakdown across the 11 standard GB regions, aggregated into a 7-poll moving average weighted by regional sub-sample size, with no bin remapping. Regional sub-samples are smaller than the GB-wide samples behind the headline figures, so the estimates carry wider error bands.

Age bins from different pollsters are standardised to five categories: 18-24, 25-34, 35-49, 50-64 and 65+, with each band mapped to the standard bin it overlaps most. Each data point uses a 7-poll moving average with sample-size weighting, drawn from the GB pollsters that publish the crosstab. The polls in the current window are listed under Recent polls in the Demographic Breakdown section.

Gender crosstabs are standardised to Female and Male. Each data point uses a 7-poll moving average with sample-size weighting, drawn from the GB pollsters that publish the crosstab. The polls in the current window are listed under Recent polls in the Demographic Breakdown section.

The 2024-vote view shows where each party's current voters say they voted in 2024 (composition) and how many of each party's 2024 voters would still back it (retention). It uses a 7-poll moving average weighted by 2024 sub-sample size, counting actual voters moving between parties rather than just percentages, using recall data from the GB pollsters that publish it. The polls in the current window are listed under Recent polls in the Demographic Breakdown section.