General Election 2019: Is the exit poll right or wrong?

A combination of pictures shows the broadcaster's exit poll results with Britain's Prime Minister Boris Johnson's Conservative Party winning 368 seats (L) and Jeremy Corbyn's Labour Party winning 191 seats, on the outside of the BBC building in London, as the ballots begin to be counted in the general election, on December 12, 2019. (Photo by Tolga AKMEN / AFP) (Photo by TOLGA AKMEN/AFP via Getty Images)

The exit poll has come out showing a crushing Conservative victory, with the Labour party down below 200 seats – a Labour defeat that not only puts the Conservatives in office but probably secures them in power after the next election as well.

Could it be wrong? The exit poll has got the headline result broadly right at every election since 1987, when it predicted a hung parliament but was 48 seats off – producing a Conservative landslide. So it is within the historical error bar that the Conservatives could end up with a smaller majority or a hung parliament. Past errors in the exit poll – not only in 1987, but the smaller ones in 1992, 2015 and 2017 – have underestimated the Conservatives. It may be that a similar error is at work – or they could have compensated. However it has been a long time since a major miss – and that miss was far off the parameters to turn this exit poll into anything other than a Tory victory. An error within the historic level would still mean a Labour defeat – but one that might not write off the 2024 election.

But for anything other than a Tory victory we’d need an error well in excess of anything the exit poll has got wrong before.