Rob Davidson

Polls can't predict this, yet

Last week, a 'big poll' predicted that the Liberal Democrats would lose almost all of their seats at the next General Election.  This caused quite a stir on social media. Liberals love to bash their own team and polling results are a regular self-flagellation. But while polls can be very accurate near an election, there are good reasons why we shouldn't fret over them 3 years out - and a clear task for us to focus on instead.

"MRP Polls" like the one published last week can be astonishingly accurate and have successfully predicted shock results such as Trump's victory in 2016 and Theresa May's 'loss' in 2017.

But the power and flaw in these polls is their sensitivity - they can vary significantly depending on when people are surveyed, how the question is framed and so on. The campaign group, Best for Britain, were rightly ridiculed when their initial 2019 MRP poll suggested that the Lib Dems were the best tactical choice to oust the Tories in Filton & Bradley Stoke.

This latest poll has come clear red flags like the Filton & Bradley Stoke prediction. But regardless of specifics in this poll's findings, there are 4 basic reasons why no General Election prediction polls can be relied on this far out:

1. Polls are snapshots: In a post-election analysis, the Liberal Democrat's stated that an MRP poll from June 2019 heavily influenced their December election strategy. Just as Best for Britain found with their early snapshot, the MRP from June 2019 didn't hold up all the way to election day.

In another post-election analysis, Social Liberal Forum points out, "MRP modelling needs to be more timely, and must be regularly updated as voters shift." So let's be clear - no MRP poll from Jan 2021 will predict an election in 2024.

2. The squeeze: As late as October 2019, the Lib Dems were still polling above 18%. The main reason for the last-minute fall from grace wasn't the divisive "Stop Brexit" message or problems with their new leader, it was the same thing that happens every election under the UK's "first past the post" system - squeeze.

The public give honest responses weeks out from and election but, when crunch time comes only one choice matters: do you want to kick out the government or keep out their main opponents?

This can work well for the Liberal Democrats at times - the (in)famous "Lib Dem Bar Chart" can convince voters to switch, where Lib Dems are believable as the challenger. But in most seats the tactical reality benefits a two party system and the polls shift massively as election day nears.

3. Tory money: The Conservative Party has more money than anyone else and spends more on elections than anyone else. That's only going to get worse as Boris Johnson has announced he intends to raise the election spending limit from £19 to £33 million. Make no mistake, this is purely aimed at allowing his own party to vastly outspend any challengers. In the 2019 General Election period, the Tories received donations of £19m while Labour only received £5m. During the 2017 General Election, the Tories spent £18m to Labour's £11m and Lib Dems £6m (2019 figures not available yet for some parties.)

4. Tory gerrymandering: It's not just money. The Tories are finally going to get around to the boundary changes that have been on the cards for years now - the effect, a 10-seat boost to Johnson.

Johnson has also introduced a Bill to scrap the Fixed Term Parliaments Act. The Bill may be somewhat flawed and we saw in 2019 that it could be bypassed with a simple majority motion but without the Bill, Johnson can set the date of the next time when it suits him best and he won't have to explain why he's breaking the now-common, 5-year cycle.

And, of course, he had his choice of Director General at the BBC - the Rupert-Murdoch-affiliated Tim Davie has already attacked 'left wing' comedians. The UK's media environment may be quite different come 2024.

This mix of factors is both good and bad for Liberals. On the good side, Lib Dems are in 2nd place in more seats than ever before meaning we can make 'squeeze' work for us. But this isn't a done deal. To outperform these early MRP polls, each seat will need to install a good candidate PDQ and that person will need to work their socks off so 'everyone knows' that they are the clear challenger.

But, on the dark side, the Tories may build the funds and rig the game enough to fight on two fronts: holding onto Labour's old Red Wall seats in the North and fending off the Liberal challengers in the South.

It's a hard road ahead but the polls today are not the signposts some think. Our job is pretty clear, let's not beat ourselves up unless we fail in that task.


Want to keep in touch?

Sign up to our weekly briefing - with all the latest news, opinion and analysis from justLiberals and our partners.

We'll never share your email with anyone else.