Earlier this month, Daily View reported that the Conservatives had opened a five-point lead over Reform UK in a More in Common poll, their biggest since the 2024 election. In the very same week, YouGov had the two parties a single point apart, and Deltapoll put Reform four points ahead of the Tories.
If you follow politics on social media, you will have seen all three numbers, each shared with total confidence by someone who liked the result.
So which one is right? Quite possibly all of them. The trick is knowing what a poll can tell you, and what it can’t.
A poll of around 2,000 people, the usual size for national voting intention surveys, comes with a margin of error of roughly two to three points either way for each party. That is not a flaw. It is simply what happens when you ask a few thousand people to stand in for 48 million voters.
So when YouGov put the Conservatives on 21% and Reform on 20%, the honest reading is not “Tories ahead”. It is “too close to call”. The true figures could easily be the other way round.
A five-point gap, like More in Common’s, is more meaningful. But it is still one poll, and one poll can be an outlier. Any pollster’s results will occasionally wobble a few points from week to week through chance alone.
Pollsters don’t just ask people how they would vote and add up the answers. Each one adjusts its raw results to match the population, and they make different choices about how.
Some give more weight to people who say they are certain to vote. Some adjust for how people voted last time. They handle “don’t knows” differently: some leave them out, others assume many will drift back to the party they backed before. And they differ on whether to list smaller parties by name when they ask the question.
Those choices add up. The clearest example this month is Restore Britain, Rupert Lowe’s new party. More in Common had it on 6%; YouGov had it on 3%. Neither is necessarily wrong. They are measuring in slightly different ways, and a new party is exactly where those differences show most.
The practical rule: compare a pollster with itself. A change in YouGov’s numbers from one week to the next tells you far more than a gap between YouGov and Deltapoll.
The best defence against a misleading poll is lots of polls. The Electoral Reform Society averages the latest survey from every major pollster each month. Across ten companies in October, Labour averaged 29.7%, Reform 22.1% and the Conservatives 21.5%. That picture is far steadier than any single headline.
The direction of travel matters too. Labour’s lead in that average has doubled three months running since Andy Burnham became prime minister. Reform has slipped, and both YouGov and More in Common now have the party at or near its lowest score in their current trackers.
When several pollsters, each using its own method, move the same way at the same time, that is a trend. When one poll jumps and the others don’t, it is probably noise.
No party has leaned on opinion polls quite like Reform. It overtook the Conservatives in January 2025, and by that September its average lead stood at 13 points. Poll leads became the backbone of its message: that it was the real opposition, and that a Reform government was only a matter of time.
That argument is harder to make now. Reform’s October average was down on the month before, and the party is at or near its lowest point with two major pollsters. Nigel Farage has admitted Reform is losing support to Restore Britain, set up by Rupert Lowe, one of the MPs elected for Reform in 2024. He is also under investigation by Parliament’s standards commissioner over how he declared a £5m donation and other payments, though he denies any wrongdoing.
A party that treated every good poll as proof of destiny can hardly dismiss the bad ones as noise.
To be fair to Reform, around one in five voters is still a remarkable level of support for a party with a handful of MPs. Deltapoll still has it ahead of the Conservatives, the next election may be years away, and polls have been wrong before. Its supporters are entitled to say the story is not finished.
But the lesson cuts both ways, and it applies to everyone. Labour should not mistake a “Burnham bounce” for a permanent realignment, and the Conservatives should not build a comeback narrative on one survey. More in Common’s own director, Luke Tryl, called it “far too early” to talk of a return to two-party politics.


