Our proprietor says correctly that a snapshot is not a prediction, but the last polls of a campaign tend nonetheless to be treated as forecasts. So YouGov’s final MRP poll from 2017 has come to be treated. It was in the words of the Times, which today publishes the first YouGov MRP survey of this campaign, “the only survey to forecast [a] hung parliament in last election”.
It finds a Conservative majority of 68 – with the Conservatives on 359 seats, Labour on 211, the SNP on 43 and the Liberal Democrats on 13. This would represent the biggest majority for a Government in Parliament since Tony Blair’s landslide of 2001, narrowly pipping at the post his 66-seat majority in 2005. It would be the biggest Tory win since Margaret Thatcher’s third victory, when she won a majority of 102 seats in 1987.
Quantum theory has it that the observer affects the observed simply by observing it. And what is said of electrons may prove true of elections, or at least of this one. Because the YouGov MRP was right last time, so to speak, some can’t help assuming that it will be right again (which helps to explain why the Times splashes on the survey this morning).
As it happens, other MRP polls were wide of the mark two years ago. And one conventional pollster, Survation, got it right: in other words, its final poll mirrored the actual result. (MRP or the “multilevel regression and post-stratification model”, to give the method its full moniker, is a particular form of modelling applied, in this case, to all constituencies in this election). This is not even the final YouGov MRP: another is expected before polling day.
However, today’s poll is, as John Rentoul puts it, “comfortably in the middle of the range of outcomes suggested by the standard opinion polls”. Together with the legend of that 2017 MRP, this may be enough to effect the progress of this campaign in ways that will be unhelpful to Boris Johnson. We can think of at least four.
First, everyone wants to lay low a tall poppy, and the Conservative campaign is already tall, because the Tories are running a tight ship and Labour a loose one. A consequence of the MRP poll is that the media will take a scythe to that poppy even more enthusiastically – probing for errors, hunting for anti-Conservative stories, and making a meal of Donald Trump’s coming visit (thus ignoring other polls which suggest that Labour may be reducing the Tory lead).
Second and simultaneously, it will be written that “the wheels are coming off Labour’s campaign”, especially in the light of Jeremy Corbyn having no meaningful response to the party’s institutional anti-semitism. The Sunday papers are a traditional venue for this kind of journalistic exercise. This could create a feedback loop whereby Labour’s ratings drop lower.
Third, such a development has the capacity to affect turnout: if voters who fear Corbyn come to believe that he won’t win, their terror may abate – and they may thus not bother to vote (though there is a counter-theory for everything: one of these has it that electoral bandwagons create their own momentum, which would have the opposite effect). That possibility is certainly worrying Downing Street and CCHQ.
Finally, and as we wrote last week, there is a danger of Conservative expectations getting ahead of the electoral facts. A Tory majority of 15 would represent a better result than that David Cameron achieved in 2015 – itself the first blue majority in over 20 years. But it might well disappoint if Conservative MPs come to expect a majority of nearer 50 – which in turn would make the difficulties of government even more problematic.
So paradoxically, this excellent poll for Johnson will only agitate him and his team. Expect, if current trends continue, a frenzy of briefing and counter-briefing about who is likely to be sacked from and promoted to Cabinet. And other inconvenient noises off. The only consolation this pro-Tory site can offer is that, all considered, good polls are better than bad ones, and success, if that is really what Johnson currently enjoys, is better than failure.