On holding a referendum to accept or reject Brexit deal (Survation) :
The poll's Scottish subsample is tiny - but for what it's worth, it shows a somewhat bigger majority of 43% to 33% in favour of a referendum. As the entire rationale of Indyref 2 is to give people the choice between Hard Brexit (assuming that's still what we're heading towards) and the only viable alternative, it would seem from these numbers that there are reasonable grounds for optimism that people will be receptive to the arguments in favour of holding a vote, if they are presented in a thoughtful way.
There's mixed news for Remainers in the other questions in the poll - the centre of gravity of public opinion appears to be support for remaining within the customs union from outside the EU. Single market membership seemingly wasn't asked about, and that's arguably the more important aspect of the Soft/Hard Brexit divide. (Norway is inside the single market but outside the customs union - and the reverse is true of Turkey, which speaks volumes.) Extraordinarily, in spite of her catastrophic loss of general support, a narrow majority of respondents would still trust Theresa May's judgement if she decided that no deal was better than the deal on offer - although unsurprisingly respondents in Scotland take the opposite view by an overwhelming margin.
* * *
I noticed a comment on Wings earlier today saying there had been a poll showing that the SNP would gain seats in an early general election. I did a search and drew a complete blank, so I assumed it was probably just some sort of 'voodoo poll'. But oddly enough, when I checked Survation's Twitter feed tonight, there was indeed a projection that the SNP would win 41 seats in a new general election (a gain of 6), and it was supposedly based in part on the notorious poll commissioned by the Record the other day. That's peculiar, because as far as I can see there were no voting intention figures in the Record's reporting, and there are none in the published datasets either. Could there have been voting intention results which have been withheld by the Record? If so, did they do it because they didn't want anything to contradict the "disaster for SNP" narrative?
I must stress this is just wild speculation. But if by any chance there has been an unpublished Survation poll of Scottish voting intentions showing the SNP making ground rather than losing ground, that would be very significant. The results would have been weighted back to 2017 recalled voting, which means they ought to be more accurate than the pre-election polling was. I've been quite concerned that any new poll might actually show the SNP slipping behind Labour - that seems an odd thing to say given that Labour finished third only ten days ago, but shock election results can generate momentum like nothing else can.