So, as our latest “15 minutes of fame” star, Brenda, has it, loosely, “OMG, why? There’s too much to be done!” Or, as Nick Robinson asked “What was it about the 20 point lead in the opinion polls that first attracted you to the idea of an early election, Prime Minister?”

Of course we don’t necessarily trust opinion polls any longer. They got the last general election wrong “because their sample groups were unrepresentative” and they got the Brexit referendum wrong “because an unexpectedly large number of people who hadn’t voted before turned out to vote.”

Both of those statements are true, to some unknown extent, but we don’t know how far they’ve been corrected and so we don’t know how far should we trust the polls. The nice errors they tell us, ±3% are based on the assumption that the sample is representative.

At the last election the polls overestimated support for Labour by about 5% (the exact figures vary a bit from poll to poll), overestimated LibDem support even more, and a big chunk of that overestimate actually voted Tory in England, in Scotland, even though the pollsters got the annihilation about right, the overestimated votes still went to the SNP, giving even bigger swings than expected. Depending on the steps taken to correct that mistake they might still be over-estimating Labour’s support, spot on, or over-estimating support for the Tories and the SNP by an unknown amount.

If lots of people that voted for Brexit come out to vote again… who knows what will happen? At that referendum there was no polling of those people and there’s no knowledge of whether they’ll vote again or not. You’d think it would be good for the Tories but one of the pluses that Jeremy Corbyn certainly brought to his leadership elections was an upswell of new members of the Labour Party who voted for him. Can he replicate that in the general election? They probably won’t be well represented in the old or the new sampling mechanisms if they exist and could represent a groundswell of Labour voters that might overturn the apparent deficit in the opinion polls.

Finally, there is a new, and honestly disturbing, element to consider this time. Several strategists I’ve heard interviewed have suggested that, alongside the actual issues, the Tories will try to campaign on “it’s closer than you think, you must turn out or you’ll get a Labour government” while Labour will campaign on “you must turn out or there will be a Tory landslide!” Both of those, from the various party’s perspectives, seem entirely reasonable. Over the weekend May and Hammond refused to confirm that they would continue the so-called Triple Lock for pensioners and commit to not raise taxes and NI in the next parliament. Both of these are likely to be unpopular, but are probably sensible (even though I’m not going to vote Tory, I think they’re sensible moves as do quite a lot of other people on all sides of the political commentary spectrum) and if you’ve got a 20 point lead, you can afford to say somewhat unpopular things like that. However, almost too soon for the comments to have registered and way out of line with the rest of the polls, that bastion of good, honest journalism the Daily Mail published a poll (with screaming headlines) that said the Conservative lead had been cut to “only” 10%. We’re supposed to believe that the failure to continue the triple lock and no tax increases has slashed the lead in half of course. But the timings just don’t support that conclusion. I wonder if we’re seeing fake news in the newspapers - a poll and an opinion piece to rally the Mail’s faithful to vote (Mail readers are almost certain to vote Conservative). If so it will be the first time it’s this blatant and a fixing of an opinion poll to suit your message, at least to my knowledge.

With six weeks to go, we’re already hearing discussions of tactical voting - both in small chats and on a national level. There was quite an in-depth piece on it on The Today Programme this morning, for example. I remember tactical voting being mentioned at just about every election that I’m old enough to remember, but (and I am relying on memory rather than anything else) I don’t think I ever remember it being seriously talked about before the last few days before the actual vote. Tactical voting being discussed, and websites and the like set up to let you determine who you might vote for, with this long to go seems to be unprecedented. Whether it will result in an increase in tactical voting remains to be seen but if it does how will that skew the results? There are relatively few seats in our elections (thanks to the multiple parties we have) where any party gets more than 50% of the votes. Generally getting over 40%, certainly over 45%, results in the seat being regarded as safe. But our opinion polls don’t look closely at individual seats. In a seat where the incumbent has 40%, the closest opposition party has 30%, then 20% then a string of smaller parties (which is not uncommon), if half the supporters of the third placed party switch to the second placed party this time, we’ve suddenly got a really close vote. Will that happen? We just don’t know but it might. Tactical voting already happens of course, and probably changes the outcome in a small number of seats, but if it’s really organised this time how much of a difference will it actually make?

The other big unknown is the apparent collapse of the UKIP vote. While Corbyn motivated the masses to vote in his leadership elections, what if Farage motivated them to vote for Brexit but now they’ve been heard, they go back to not voting? UKIP’s support peaked at around 25%, it’s not down to about 10%. What will happen to former UKIP voters? The polls suggest they’ve mostly gone back to voting Tory, but will former Labour-turned-UKIP voters go back to Labour?

If you believe the opinion polls, this election is a done-deal. But there are real reasons to doubt the polls this time.

Yesterday’s twist is that the Tories are setting themselves up as the party of Hard Brexit and the Labour Party as the party of Soft Brexit. That’s playing badly for Labour in the right-wing press of course, who are portraying it as muddle-headed but how much will it appeal to Remainers who want a soft-as-possible Brexit, plus a core of Brexiteers who voted Leave but didn’t vote for the hard core leave or walk away that they’re now being told is what they did vote for? How many Tory voters that voted Remain will tempted by the prospect of a party that will negotiate for a Soft Brexit as the best deal? Or, after two years, can bear to switch to the LibDems and vote for a second referendum? Likewise, how many Labour Leave voters will be too tempted by a Hard Brexit and will vote Tory? We will find out in six weeks time of course. The parties have to commit to their plans and strategies, and many voters will make their decision based on other things as well or instead.

I think, despite the big lead we’re seeing for the Conservatives the outcome for this election is not a sure thing simply because I’m really not sure how reliable the samples are in the opinion polls being used for the reasons laid out above. And even if the Tories do sweep to a comfortable majority it’s certainly possible there will be a change to the established pattern of safe red and blue seats across the country.

No comments

The author does not allow comments to this entry