AS THE 2024 UK General Election wrapped up, Keir Starmer’s Labour Party swept to power with an overwhelming majority of 412 seats. In the run-up to the election, Labour’s victory was all but certain, with the question merely being one of degrees.

For the winners, the exit poll was fairly accurate. Labour was slated to take 410 seats, and with nearly all seats declared, have won 412. However, outside of Labour, the polls seem to have read it wrong.

The Conservative Party, in government for the last 14 years, was predicted to pick up 131 seats, but finished 10 below on 121. Similarly for the right-wing rivals, Reform. Although they took over 14% of the vote nationally and were polling with 13 seats, they have only managed to win four seats.

Another surprise is the Liberal Democrats, which took 71 seats, increasing their share in the House of Commons by 63. Despite this win that party leader Ed Davies has hailed as “record-breaking”, the party were polling far lower when stations closed last night, coming in at a still respectable 61 seats.

While a discrepancy of 10 seats may not seem all that significant, in the context of how accurate UK exit polls usually are, it certainly is.

The margin for error in Westminster elections usually sits at around one or two seats, with seven or eight being considered a serious anomaly. In 2010, when the Conservatives first came to power following three successive Labour governments, exit polls had a margin of error of one to three. In 2017 this increased to four, while in 2019 it was two to four.

The fact that the polls were out by 10 seats for three different parties is worth commenting on.

So what happened?

Speaking to the BBC on the release of the polls, the political scientist and polling expert, John Curtice used the prediction for Nigel Farage’s Reform Party as an example. He said that it was reached by taking those constituencies where Reform is considered to be in with a chance and then adding them up.

“What we’ve got is a lot of places where there is a small chance that Reform will win the seat, but we’re talking a 20%, 30% chance,” he said.

“But the result could end up being quite a bit bigger or smaller, all depending on how the cookie crumbles.”

What seems to have happened is that a number of constituencies were quite tight, and the exit poll was forced to come down on one side or the other based on the data that it had. This ended up with a number of the tighter constituencies going to the wrong party in the exit poll, leading to the discrepancies that we saw.