14 December, 2019| GE2019

How well did the polls do in projecting the 2019 election result?


After somewhat poor showings in 2015 and 2017, pollsters in 2019 performed admirably, with all final polls close to the actual result.


Thursday 12th December saw close to thirty-two millions votes cast, resulting in a sizeable majority for the Conservatives, and the worst defeat for Labour since 1935.

The 2019 election campaign bore witness to pressure groups utilising forecast models for tactical voting gain, as well as political parties exploiting polling so as to squeeze the voters of third place candidates into backing them as opposed to their first choice. Undoubtedly election campaigns are a profitable scene for pollsters, but when it came to projecting how voters would vote, how well did they do? And, more importantly, how well did the Britain Elects Poll Tracker do?

The answer to both? Quite well, actually.

After criticism for somewhat poor showings in the 2015 and 2017 general elections, pollsters, particularly Opinium, Ipsos Mori and Survation, can walk away from the 2019 election with their heads held high. None of the ten final election polls were wildly off the mark of the actual result.

On the detail, only one pollster overstated Conservative support in their final election poll, whereas eight overstated Labour, with two of those eight overstating support by more than 3pts.

The Britain Elects Poll Tracker also performed well, projecting a Conservative lead of over 9pts, when in reality the Tories secured a lead of Labour of over 11pts.