Friday, December 18, 2020

On what planet is a Tory boycott of an indyref, guaranteeing a Yes majority, a bad thing?

I was just catching up with the story in The National from a few days ago about the suggestion that the Tories might boycott a consultative referendum held without a Section 30 order, even if it clears every legal hurdle and has the blessing of the UK Supreme Court.  This, of course, should be a cause for celebration, because it would guarantee a Yes victory, and we'd all have a nice relaxing campaign trying to maximise the scale of the win and the overall turnout.  Whatever mandate is achieved would then be used as leverage to bring us closer to actual independence or at least to an agreed referendum - and leverage is, after all, the whole point of holding a consultative referendum.

Curiously, though, the article finishes with a quote from Malcolm Harvey, who back in the day was one of the leading pro-indy bloggers (he was one of the founding triumvirate at Better Nation, and before that had his own blog Malc in the Burgh).

"Turnout would be about 50% and the thing would be even less legitimate than holding an un-sanctioned referendum in the first place. It's a mad idea."

My jaw dropped to the floor when I realised he wasn't describing the boycott as mad, but the idea that we should hold a referendum that might be boycotted.  Is anyone spotting the slight flaw here?  There'd be nothing to stop the Tories boycotting absolutely any referendum, even one brought about by a Section 30.  It's hard to think of a better definition of 'madness' than giving our opponents a veto on every move we make.

You know the bit at the end of Life of Brian when the Crack Suicide Squad turn up, stab themselves, and say with their dying breath "that showed 'em, huh?" If Malcolm and Pete Wishart had been in charge of the Roman troops that day, they'd have reacted to that seemingly futile gesture by saying "well, that's it then, we can't possibly go ahead with these crucifixions now..."

An average of all twenty-three polls conducted this year puts support for independence at more than 53%

By overwhelming popular demand (well, a couple of people asked) here is an update of this blog's Poll of Polls.  Just a reminder that rather than including all of the most recent polls, it includes only the most recent poll from each individual firm, to ensure that no 'house effect' is given too much weight.  There are six polls in the current sample - one from Savanta ComRes, one from Survation, one from Panelbase, one from Ipsos-Mori, one from YouGov and one from JL Partners.  The percentage changes are measured from the last published update on 10th November.


Should Scotland be an independent country?

Yes 54.8% (-)
No 45.2% (-)

So no change whatsoever!  Of course there's been movement underneath the surface, but it's cancelled itself out.  Panelbase and Savanta ComRes have shown a swing to Yes (a huge swing in the latter case), while YouGov, Survation and Ipsos-Mori have shown a modest swing to No.

It's entirely possible that we've now seen the final Scottish poll of 2020, in which case this is also a good moment to calculate an average of all the year's polls - I can always update it later if another poll unexpectedly turns up.  This isn't as straightforward an exercise as it appears, because there's a debate to be had over what should be considered a 'proper' independence poll.  Twenty-one polls this year have asked the standard independence question, while an additional two have asked about independence using non-standard questions.  There's also been a Scotland in Union propaganda poll asking about "leaving the United Kingdom", which as I always point out can't be considered a question about independence, because it's perfectly possible to leave the UK without becoming independent.  (For example, if Northern Ireland ever leaves the UK, it will almost certainly join another existing state rather than become independent.)  So I've excluded the non-indy propaganda poll, but included the two non-standard indy polls for the sake of completeness - which is hopefully fair enough, because one of the two was commissioned by an anti-indy client (Hanbury) and the other by a pro-indy client (Progress Scotland).

2020 average:

Yes 53.2%
No 46.8%

We've never come close before to a whole calendar year in which there was a majority for Yes, so we've broken that duck in quite some style.  (And in case you're wondering, including the Scotland in Union propaganda poll would have made only a minimal difference - Yes would still have been on 52.8%.)

A few people have also been asking already about future crowdfunded Scot Goes Pop polling.  Obviously I have to be careful not to go to the well too often, because there's always a danger of donation fatigue setting in, but if anyone has any strong feelings about timing, let me know in the comments section below.  The two obvious options are at the start of 2021 when people are coming to terms with the realities of Brexit, and at some point in April/early May, during the Holyrood campaign proper.

Thursday, December 17, 2020

Truly staggering: support for independence reaches all-time high in an online poll of 58%

The Survation poll published a couple of days ago raised the genuine possibility that Yes support might have fallen back slightly, because it was the third poll out of four to show a 2-point drop.  But that theory has been blown out of the water by a sensational new poll from ComRes that by two clear percentage points puts Yes at the highest ever level of support in any poll conducted online by any firm.

Should Scotland be an independent country? (Savanta ComRes)

Yes 58% (+5)
No 42% (-5)

Even if this poll proves to be an outlier, it's pretty much impossible to reconcile it with a situation in which Yes is moving backwards.  At worst it means that nothing much has changed since the summer, and at best it means we've taken another big step forward.  It's also, of course, the sixteenth poll in a row to put Yes ahead on the standard indy question, and the seventeenth in a row if you add in a Progress Scotland poll that used a non-standard format.

Today's results are from a new series of monthly polls commissioned by The Scotsman - and whatever you think of that publication, it's a very welcome development.  Scotland, at this stage in its political story, is crying out for regular polling, and yet the monthly Herald/System Three series that ran for decades is now a long-distant memory.  Curiously, The Scotsman have opted to slightly blunt the impact of their first poll by billing the 58% Yes vote as merely the joint highest ever - which is true, but only if you count a poll that used a completely different data collection method (an Ipsos-Mori telephone poll in October).

I'm on record as saying many times that if Nicola Sturgeon intends to wait until Yes hits a sustained 60% in the polls before doing anything, she'll wait forever and Scotland will never become independent.  I was asked this morning if the ComRes numbers have changed my mind about that.  As ever, the answer is no: we still haven't reached 60%, and even if we eventually do in the occasional poll, it's unlikely to be on a sustained basis.  But hopefully now we've actually reached these giddy heights, the penny may have dropped with the leadership that we don't need anything over and above this.  And Keith Brown's repeated use of the words "settled will" would appear to support that interpretation.

Scottish Parliament constituency ballot: 

SNP 55% (+5)
Conservatives 20% (-3)
Labour 16% (-2)
Liberal Democrats 6% (-)

Scottish Parliament regional list ballot: 

SNP 42% (+1)
Conservatives 20% (-1)
Labour 17% (-1)
Greens 12% (+1)
Liberal Democrats 7% (-)

Again, these numbers contradict most recent polls which have shown modest movement against the SNP.  However, I would still guess the SNP might be underestimated on the list vote (and the Greens overestimated) due to the Survation-like way in which ComRes pose the list question, which may lead to some people replying as if they were being asked for a second preference vote.

In fairness to Douglas Ross, his net favourability rating of -9 isn't quite as poor as in other polls, but it's still miles behind Nicola Sturgeon's, slightly behind Patrick Harvie's, and no better than Willie Rennie's.  This is plainly not what the Tories hoped for or expected when they defenestrated Jackson Carlaw.  They'll be particularly perturbed that just 16% of respondents regard Ross as "charismatic".  Presumably he would never have been hand-picked unless he had been expected to be seen as charismatic, so the experiment really isn't working out.

Tuesday, December 15, 2020

Analysis of yesterday's Survation poll: the fifteenth or sixteenth in a row to show a pro-independence majority

Apologies for not covering this yesterday, but I'm sure most of you are already up to speed with it.  A new Survation poll is the fifteenth in a row (or arguably sixteenth in a row depending on definition) to show a majority for independence.

Should Scotland be an independent country?

Yes 52% (-2)
No 48% (+2)

Three of the last four polling firms to report have now shown a small recent decrease in Yes support - so that might be significant, but it could still be happening by chance.  The exception was of course the Panelbase poll for this blog in November showing Yes climbing to an all-time high of 56%.  

Westminster voting intention:

SNP 51% (-1)
Labour 21% (+1)
Conservatives 20% (+2)
Liberal Democrats 6% (-2)

The changes here don't appear to be of any great significance, as they just revert to the numbers in the last-but-one Survation poll.  Labour's second place seems to be a Survation 'house effect' - it's been seen in all of the last three Survation polls, but not in polls from other firms.

Scottish Parliament constituency ballot:

SNP 53% (-1)
Labour 20% (+2)
Conservatives 20% (+1)
Liberal Democrats 6% (-2)
Greens 1% (+1)

Scottish Parliament regional list ballot:

SNP 41% (-2)
Labour 20% (+1)
Conservatives 18% (+1)
Greens 10% (n/c)
Liberal Democrats 7% (n/c)

As I always point out, the list numbers from Survation polls need to be taken with a pinch of salt, because respondents seem to be influenced by the way the question is posed.  Some pro-indy voters seem to be left with the impression that they're being asked for a second preference, thus leading to a (possible) understatement of the SNP and overstatement of the Greens.