Dramatic BMG poll suggests Yes surge is real : pro-independence vote at 48%, retaining most of last month's big gains

After last week's dead-heat Ipsos-Mori poll (actually a narrow Yes lead on one measure), I suggested that if the next poll reported a Yes vote of 49% or higher, that would make it very likely that the apparent recent boost in support for independence has been real and not an illusion caused by random sampling variation.  Tonight's BMG poll falls just short of that target figure, putting Yes on 48%, but the conclusion to draw may not be radically different.  Yes have retained most of the 3.5% jump in support from the BMG poll last month, with the sharp 'reversion to the mean' that many people expected failing to occur.

Should Scotland be an independent country?

Yes 48% (-1)
No 52% (+1)

This is the fourth poll in the regular(ish) BMG/Herald series, and the sequence of results for Yes so far has been 45%, 45.5%, 49%, 48%.  Those numbers are not inconsistent with a static position - ie. if the true figure has remained constant at around 47%, it may just be pure chance that the first two polls produced an understatement and the next two produced an overstatement.  It would all be comfortably within the margin of error.  However, the telling boost for Yes in the Ipsos-Mori poll, coupled with the fact that three of the last four polls from all firms have put the Yes vote above the 45-47% range we've become accustomed to, probably tips the balance in favour of there having been a genuine recent swing to Yes.  As ever, we need more information, and the next poll may cause a hurried reassessment, but that's how it looks just at the moment.

Make no mistake, though - regardless of whether there has been a recent swing, the last few polls have constituted a significant setback for the anti-independence forces.  They'll put a brave face on tonight's result, emphasising the continuing No lead...but it was only a few months ago that they were crowing about a YouGov poll that suggested Yes support may for the first time have fallen below the 45% achieved in the 2014 referendum.  At that point, it genuinely seemed possible that Brexit was perversely putting support for independence into a tailspin, and that the threat to unionism had almost passed.  Instead, the Tories, Labour and the Liberal Democrats are now having to readjust to a reality in which Yes support is almost certainly higher than the 2014 result, and in which it's far from guaranteed that there is currently even a No lead.  Tonight's poll is technically a statistical tie, meaning that the standard margin of error leaves a degree of doubt over which side is actually in front.

The report in the Herald claims that respondents in the poll were opposed to a second indyref taking place "before Brexit" by a relatively narrow margin of 49% to 39%.  However, that's likely to be a misrepresentation of the question that was actually asked, which last month was about whether there should be another referendum before Brexit negotiations are concluded - not the same thing as "before Brexit".  It would be perfectly possible to hold a referendum after the completion of negotiations but before the actual date of Brexit, which means that this poll leaves open the possibility that there is majority support for an indyref before Brexit.

Incidentally, the result on that question is essentially identical to last month's, making a complete nonsense of the claims in the Express a week ago that there has been a sharp decline in support for a referendum.  That silly fiction was based on a BMG poll commissioned by an anti-independence and pro-Brexit campaigner, which used a radically different and highly leading question.

One small but important point that needs emphasising - as of last month, BMG were still not weighting their results by country of birth, which is standard practice for most of their competitors.  We'll see when the datasets emerge whether they've put their house in order for this poll, but if not, it's possible that the Yes vote is being slightly underestimated.  For some reason, there generally seems to be a disproportionately high number of English-born people in volunteer polling panels, meaning there is often a small inbuilt skew towards No unless the imbalance is corrected for.

*  *  *

SCOT GOES POP POLL OF POLLS

Should Scotland be an independent country?

Yes 48.0% (-0.3)
No 52.0% (+0.3)

(The Poll of Polls is based on a rolling average of the most recent poll from each firm that has reported at least once within the last three months. The firms included in the current sample are Panelbase, BMG and Ipsos-Mori. The aggregate YouGov figures for August to December are excluded, because they don't really constitute a standalone poll. The most recent proper YouGov poll was completed just over three months ago.)

*  *  *

To return briefly to the subject of a few days ago, I've been taking a look at Jim Sillars' "open letter to the Yes movement", which strikes me as being distinctly odd in a number of ways.  Firstly, he blasts the SNP for dictating the pace of a second independence referendum, and demands that other voices within the Yes movement be heard, such as himself.  Now I realise that we have to be careful about being too exclusive in any definition of what constitutes part of the Yes movement, but I don't think it's unreasonable to say that a basic minimum requirement is that the people involved should actually be planning to vote Yes.  As I understand it, that is not Jim Sillars' current position - he might vote Yes in certain circumstances, and he might abstain in other circumstances.  That makes him not so much a Yes supporter as a floating voter.

Secondly, he complains about the characterisation of Scotland's enforced departure from the EU as being a project driven by right-wing Tories -

"I am not a right-wing Tory. I voted leave in a referendum that, according to the ballot paper, was about the UK and the EU (not about Scotland and the EU), as did many SNP members and voters. I expect the UK Government to give effect to the result of my majority vote, which is not dragging us out, but negotiating a deal if common sense will apply in Brussels."

That is nothing short of extraordinary.  The "majority vote" he is talking about is a British one, which is in direct conflict with the Scottish majority vote.  As a British elector, he demands that his British vote must take precedence over Scottish self-determination, and that Scotland cannot and must not act until the British will is imposed upon it.  That is the quintessential unionist (or British nationalist) worldview.  How anyone can write those words while posing as a stalwart of the Yes movement is beyond me.

Thirdly, he feigns confusion over the question that would be put in a second indyref -

"What is the proposed referendum question to be? Membership application to the EU if they will let us in? Or seeking membership of the EEA via EFTA? Or plain independence from the UK with no guidance to where and how we shall trade?"

I suspect he was paying as close attention as the rest of us when the Scottish Government made perfectly clear that their preferred question will once again be "Should Scotland be an independent country?".  The EU will not feature as part of a referendum question on independence, and it would be downright bizarre if it did.  That does not mean guidance will be absent, however - it looks as if the view will be (not unreasonably) that Scotland has already decisively opted to remain within the EU.  None of this will preclude opponents of EU membership from pursuing their legitimate objective in future - all they need to do is secure a mandate in a democratic election.  There will, I expect, be regular democratic elections in an independent Scotland.

Perhaps the referendum question Sillars is really hankering after is this one -

"Do you think devolution was a bad idea in 1970, a good idea in 1979, and a unionist trap in 1997?  Do you think independence was a bad idea in 1976, an absolute imperative in 2014, and kinda 'meh' in 2017?  And do you think 'independence in Europe' was a great slogan in 1992, but an abomination in 2017?"

To which there are, naturally, only two possible answers - either "No", or "I'm Jim Sillars".

*  *  *

If you've enjoyed my writing in recent months and feel a strange inexplicable urge to 'buy me a hot chocolate', bear in mind that my fundraiser from two years ago is still open for additional donations - it can be found HERE.