Enormous lead for the SNP in first post-June Scottish poll

"The findings are likely to dismay Nicola Sturgeon" says Jason Allardyce of the Sunday Times about a new poll which he must know perfectly well is going to absolutely thrill Nicola Sturgeon, because it shows the SNP have shrugged off the hammering they've been taking in the media since June, and have retained an enormous lead in Scottish Parliament voting intentions.

Holyrood constituency voting intentions (Panelbase) :

SNP 42%
Conservatives 28%
Labour 22%
Liberal Democrats 6%
Greens 2%

No percentage changes are listed because it appears that this is the first Scottish Parliament poll that Panelbase have conducted since the Holyrood election almost eighteen months ago.  They've done Scottish voting intention polls since then, but for obvious reasons the voting intentions being measured were for Westminster or local council elections.  And that brings me to the most important health warning about these numbers.  I rarely disagree with Stuart Campbell of Wings, but he's attempting to use this poll as evidence to support his belief that the pattern shown by the Scottish subsamples of Britain-wide polls (basically a tight three-way contest, with the SNP and Labour battling it out for first place) is completely meaningless.  Much more likely, I'm afraid, is that the reason for the disparity is that those subsamples relate to Westminster voting intention only.  

Nobody disputes that an individual subsample should be regarded as unreliable, but an aggregate of a large number of subsamples is likely to produce figures that are at least within the correct ball-park.  What they've shown is in line with what seems intuitively likely, and indeed what was apparently borne out by the local by-elections on Thursday - ie. that Labour have built upon their mini-recovery at the general election.  It's not clear whether Panelbase even asked for Westminster voting intentions in the new poll (there's no sign of Westminster numbers on Twitter, and I don't pay the Murdoch levy so I haven't been able to read the Sunday Times article in full), but my strong suspicion is that a full-scale Scottish Westminster poll would show the SNP significantly lower than the Holyrood numbers suggest, and Labour significantly higher.  To believe anything else, you would have to believe that there has been a large swing from Labour to SNP since June, which flies in the face of all logic given what happened on Thursday.  

I suspect that the traditional gap between Westminster and Holyrood voting intentions, which has existed for the vast majority of the eighteen years since devolution, has reasserted itself.  If so, the explanation is the usual one - there is a significant minority of voters out there who like both Labour and the SNP, but think Labour is the more natural choice for Westminster and that the SNP is the most natural choice for Holyrood.

In many ways that's not bad news.  The question of whether a second independence referendum is held will ultimately be decided by Holyrood arithmetic, after all.  This is just one poll, but if other firms corroborate Panelbase's findings, it means that the SNP have only lost around 4% or 5% of support since their decisive victory in May 2016.  If another Holyrood election was held now, it looks highly likely that the SNP would be effectively re-elected as a minority government - probably without a pro-independence majority in parliament, but they're not a million miles away from holding on to even that.  With no election due until May 2021, the unionist parties can scarcely look at this poll and think "all we have to do is hold out until the next election, and then the parliamentary majority for an indyref will automatically disappear and we can all get back to normal".

And as far as the prospects for actually dislodging the SNP from government are concerned, this poll is an absolute hammerblow for the unionists.  The Tories seem to have a natural ceiling of around 30% support, so realistically if any party is going to defeat the SNP it'll have to be Labour - and yet even after a very favourable summer, Labour still seemingly find themselves twenty points behind the SNP, and in a dismal third place.  It appears that no matter how good things get for Labour in Westminster terms (or even perhaps in local government terms) there will always be gravity holding them back in Scottish Parliament elections.  Perhaps the only way to overcome that handicap would be to find a truly inspirational leader - but instead they're going to be stuck with either Anas Sarwar or mystery man Richard Leonard (and I suspect it may well be Sarwar, who is the worse of the two).

Panelbase also asked a voting intention question on independence -

Should Scotland be an independent country?

Yes 43% (-1)
No 57% (+1)

For my money, the SNP leadership will be very pleased (or at least monumentally relieved) about those figures.  Although we haven't had a post-election party political voting intention poll until today, there was an independence poll from Survation in mid-June which suggested a substantial drop in support for independence.  There appeared to be a danger that the momentum was running away from Yes, in which case there was a chance that things might have worsened significantly over the course of the summer.  Instead, the situation seems to have stabilised.  The percentage changes listed above are from the Panelbase poll conducted in the days leading up to the general election, so it appears that there hasn't been a statistically significant swing since then.  Yes remains firmly in the game.