Nothing has changed: vote-splitting in Holyrood elections is still a mug's game

So I've just been catching up with today's news (gleefully announced by Kenny "Devo or Death" Farquharson) that Wings Over Scotland editor Stuart Campbell is pondering the idea of setting up a new pro-independence party to stand in the 2021 Holyrood election, in direct competition with the SNP.  As you know, I've always been very sympathetic towards Wings, but I'm not going to be a hypocrite about this: vote-splitting in Holyrood elections is still a mug's game, irrespective of whether we're talking about a Wings party or RISE or any other small party.  What do I mean by vote-splitting?  I mean people who want an SNP government, but are lured into wrongly thinking they can somehow maximise the number of pro-independence seats by only voting for the SNP on the constituency ballot, and giving their list vote to another pro-independence party.

You might recall that analysis by John Curtice suggested it was possible that "tactical voting for the Greens" was directly responsible for costing the SNP their overall majority in 2016 - without vote-splitting by SNP supporters, the SNP could potentially have won an extra two list seats, which would have given them an overall majority of exactly one.  Vote-splitting enthusiasts like Kevin Williamson had been absurdly claiming for months before the election that the SNP were absolutely guaranteed to win at least 65 of the 73 constituency seats, and therefore didn't need any list votes at all.  Kevin was proved hopelessly wrong about that, as many of us had pointed out was pretty likely.  You just can't know in advance how many constituency seats a party will win - opinion polls are snapshots, not predictions, and often they're not even accurate snapshots.  A few percentage points one way or another can make the difference between winning 50 constituency seats and winning 20.  And if you don't have a clue how many constituency seats a party is going to win, by definition you also don't have a clue whether that party will be in desperate need of as many list votes as it can possibly get.

How much difference would it have made if the SNP had got their overall majority in 2016?  It's impossible to know, but it would at least have made a psychological difference, and the debate over whether the mandate for a second independence referendum is a "real" mandate might have followed a slightly different course.  (Doubtless the unionist parties would have still come up with some excuse for denying the mandate, but they'd have really been scraping the bottom of the barrel.)  I don't want us to repeat the mistake of 2016 by giving the unionist parties any more gift-wrapped excuses.

The biggest danger of the proposed Wings party is that it might fall between two stools, ie. it could take enough votes on the list ballot to do the SNP and the Greens significant damage, but still fall below the de facto threshold for winning any seats itself - in other words, it could lead to a net increase in the number of Tory, Labour and Liberal Democrat seats.  We don't yet have enough information to judge whether that is likely to happen, but I have to say I'm a tad sceptical that the Wings party would top 5% of the list vote. Stuart has today pointed to Panelbase polling showing that Wings is a highly recognised 'brand', with only 45% of respondents saying they had never heard of it.  But the reality is that online polling is likely to produce skewed figures on that sort of question, because people who read a great deal about politics are disproportionately likely to join volunteer online polling panels.  Don't get me wrong, there's no doubt that Stuart has an absolutely enormous following - but Esther Rantzen and Robert Kilroy-Silk are also both household names, and they still failed to break the mould of British politics when they attempted to do so.  It's always a mistake to underestimate people's tendency to revert to the major parties in a key election.

I'd imagine Stuart would point out that his proposed initiative isn't just about attempting to game the Holyrood system - it's also being mooted because the SNP aren't pursuing independence strongly enough at this moment of national crisis, and are also in danger of disappearing into a US-style identity politics quagmire.  Voters, he would say, are crying out for an alternative.  And I'm not going to deny that if we ever reached the point where it was rational to conclude that the SNP are never going to be serious about delivering independence, I'd probably be looking for an alternative myself.  But we are a long, long, long way from reaching that point, especially when the bulk of the SNP membership are itching for action on independence as soon as humanly possible.  I'm not any keener on the identity politics stuff than Stuart is: for example, it's now (at least on paper) SNP policy to introduce the Swedish model on prostitution law, which I've always felt infantilises women and is discriminatory against men.  But there's a bigger picture here, and that sort of thing would never make me walk away from the SNP.  You're never going to find a party with a set of policies that you can agree with 100% on every dot and comma.

Imagine what would have happened if Jeremy Corbyn, John McDonnell and the rest of the Labour left had prematurely concluded during the Blair years that the game was up and that they should set up a new socialist party to compete with Labour.  Would they have achieved anything?  Well, the new party might have recorded a respectable 3% or 4% of the vote in general elections, thus making it easier for the Tories to win.  And that would have been about it.  They'd never have got anything like as close to power as they did in June 2017.

Every Loser Wins: How "democracy" works in Scotland as part of the UK

Scottish Labour seem to be slipping deeper and deeper into an Alice Through the Looking Glass world where the acceptance of reality is, at best, an optional extra.  First we had a letter from Labour candidates waffling about the "uncertainty and economic upheaval that leaving the UK would cause", and then adding as an afterthought that the Tories are also "threatening our place in Europe".  Er, the Tories aren't "threatening" anything, this isn't David Cameron in 2012 making vague noises about an EU referendum in the medium-term.  The Tories are actually taking us out of Europe, probably without any deal, in a matter of a few short weeks.  Scotland is indeed facing "uncertainty and economic upheaval", and that'll be caused entirely by the prospect of a No Deal Brexit, which itself has been caused indirectly by the mistake of voting to stay in the UK in 2014.  We've essentially reached the point where it's intellectually dishonest to claim (as the likes of Alex Cole-Hamilton do) to be in favour of both EU membership and remaining part of the UK.  Those two things are now irreconcilable, and a choice will have to be made.  And that hasn't happened as a result of some sort of dastardly SNP plot - the people of England freely chose in the 2016 referendum to change the nature of the 'deal' that was on offer for Scotland as part of the UK.

Then we had Richard Leonard "slapping down" John McDonnell (a slightly odd thing for a branch office manager to be doing to the Shadow Chancellor) by insisting that the majority of people in Scotland are opposed to a second independence referendum.  To state the bleedin' obvious, he's making himself look a bit bloody ridiculous saying that sort of thing, because the last two opinion polls - including one published just two days ago - showed a slim majority in favour of holding a second indyref within the next couple of years.

To reiterate, though, the whole notion of a Scottish party branch pulling its errant London leadership "back into line" is really rather peculiar.  The logic of saying that Nicola Sturgeon can't just hold a referendum when she wants to is that "constitutional matters are for the United Kingdom government to decide", not for devolved politicians.  In other words, if there's a Labour government in the near future, it's for Jeremy Corbyn and John McDonnell to decide.  And yet the likes of Ian Murray and Richard Leonard are indignantly saying: "Oh no no no, this is not a matter for the UK government, but for a Scottish party.  And not for the party that was actually elected to government in Scotland, but for the second-largest opposition party in the Scottish Parliament."

What do you have to do to decide the future of Scotland?  Stand for election and be soundly beaten, it seems.

*  *  *

UPDATE: And the ever-reliable Stephen Daisley has joined Scottish Labour through the looking glass...

"Every so often, Jeremy Corbyn pops up to throw the SNP a bone, much to the horror of his Scottish foot soldiers, who know how toxic the independence issue is with their voters."

Well, it can't really be all that toxic, can it, Stephen, given that the Ashcroft poll shows 40% of Scottish Labour voters want an independent Scotland?

Has Ashcroft poll turned Our Precious Union to ASHES? Westminster in shock as a MAJORITY of Scots now support independence

Apologies for the slight delay in getting this post up - I was writing an article for The National about the Ashcroft poll, which you can read HERE.  But here are the rather wonderful figures that you've probably already seen...

Should Scotland be an independent country?

Yes 52%
No 48%

It's not possible to give percentage changes from the last comparable poll, because Ashcroft hasn't been polling on independence in recent years.  That means, strictly speaking, that we shouldn't talk about "Yes moving into the lead", because it's conceivable that previous Ashcroft polls (if they had existed) would also have shown a Yes lead.  However, even the most Yes-friendly pollsters had No consistently in the lead last year, so it does seem overwhelmingly likely that the Ashcroft methodology would have shown a No lead giving way to a Yes lead at some point - but what we don't know is exactly when that would have happened.  Is the Yes breakthrough a direct consequence of Boris Johnson becoming Prime Minister?  We'll only find out for sure if Panelbase (the only firm to have polled fairly regularly on independence this year) shows a Yes lead in their next poll.

Before I proceed any further, I'd just like to observe again that Mike Smithson, known and loved by thousands of East Dunbartonshire residents as a keen letter-writing impartial Lib Dem election expert, is utterly unspoofable.  Here we have a poll that shows a majority in favour of independence for the first time in two years, that shows a majority want a second independence referendum by 2021, and that shows Nicola Sturgeon is comfortably the most popular leading politician in Scotland, well ahead of Ruth Davidson and Jo Swinson.  And yet what is Smithson's choice of headline?  "Lord Ashcroft poll has Swinson beating Johnson, Corbyn and Farage in Scotland."  Technically true, Mike, but I'd gently suggest to you that there's a bigger picture here that you might be missing through those Lib Dem goggles of yours.

In fact, a direct comparison between the personal ratings of Nicola Sturgeon and Jo Swinson makes for pretty depressing reading for the new Lib Dem leader.  Respondents were asked to give each politician a score between 0 and 100, with 0 being the worst possible figure and 100 the best.  In spite of the hatred (I don't think that's too strong a word) that some unionists feel towards Ms Sturgeon, and in spite of the fact that Ms Swinson is significantly less well-known than Ms Sturgeon, the 30% of voters who gave Ms Swinson a basement rating of between 0 and 10 is virtually identical to the 34% who did the same for Ms Sturgeon.  Meanwhile, a measly 4% gave Ms Swinson a high rating of between 81 and 100, compared to a very substantial 24% who did so for Ms Sturgeon.

Although it's true that Ms Swinson's average rating of 31 (that's 31 out of 100, remember!) is higher than that of deeply unpopular politicians such as Boris Johnson and Jeremy Corbyn, it's not that much higher in the overall scheme of things.  She's only seven points ahead of Mr Johnson on 24, and nine points ahead of Mr Corbyn on 22.  That's not really much to write home about.

Average (mean) rating of each politician out of 100:

Nicola Sturgeon 45
Ruth Davidson 36
Jo Swinson 31
Willie Rennie 30
Boris Johnson 24
Jeremy Corbyn 22
Richard Leonard 22
Nigel Farage 18

Average (median) rating of each politician out of 100:

Nicola Sturgeon 50
Ruth Davidson 26
Jo Swinson 25
Willie Rennie 25
Richard Leonard 10
Jeremy Corbyn 9
Boris Johnson 3
Nigel Farage 2

You can just imagine the mounting panic of unionist politicians and strategists when they first read through this poll.  Normally it's possible for them to find a silver lining to cling to somewhere, but on this occasion the Yes side seem to have managed a full house...

Majority for independence - CHECK
Majority in favour of holding an independence referendum by 2021 - CHECK
Majority who think maintaining EU membership is more important than staying part of the UK - CHECK
Majority who think Brexit strengthens the case for independence - CHECK
Majority who think Brexit makes independence more likely - CHECK
Majority who predict a second independence referendum would result in a Yes win - CHECK
Nicola Sturgeon the most popular politician - CHECK

I'm slightly dubious about the wording of the question that asks whether EU membership or staying part of the UK is more important if it's not possible to have both, because there's an implicit presumption there that it would be desirable to have both, which may have caused some pro-indy people to opt out of the question altogether.  In spite of that, though, 45% say EU membership is more important, and only 43% say remaining in the UK is more important.

A few people have been asking whether it's true that 16 and 17 year olds were not interviewed for the poll.  As far as I can see from the datasets that's the case, so it's possible that the Yes vote should be a little higher and the No vote should be a little lower.  It's unlikely that it would make more than a 1% difference in each case, although that would still be enough to turn a 4% Yes lead into a 6% Yes lead.  That said, it's worth pointing out that there was a poll a few months ago that appeared to exclude 16 and 17 year olds...but that turned out to be an error in the datasets.