It’s actually not “the second-highest Yes vote ever recorded” – but 54% for independence in a remarkable new poll is still plenty high enough to be getting on with

As you can see above, the Find Out Now poll with the Yes lead was commissioned by The National, who are claiming in their write-up that 54% for Yes is the “second-highest level ever reached”.  I can’t make any sense of that.  54% is certainly unusually high, but it’s identical to the previous poll from the same firm, and there have been multiple previous polls from other firms putting Yes anywhere between 55% and 58%.  The National are citing Mark McGeoghegan as the source of their claim, which is odd, because although he’s an abusive troll on social media and an identity politics extremist, he generally does know his stuff about polling.  But if he’s been quoted correctly, he seems to have got it wrong on this occasion.

If I had seen the Survation poll on its own today, I would have been fairly despondent, because taken in combination with the ComRes poll at the end of last year, it would have looked very much like the effect of the Supreme Court ruling had eased off and we were back to the status quo ante of a small-to-moderate No lead.  However, Find Out Now’s figures have very much muddied the waters and we’ll thus have to wait to see which version of public opinion is corroborated by future polls.  It’s also possible, of course, that the later fieldwork can partly explain the difference between the Survation and Find Out Now results, because perhaps public opinion has changed due to the veto of the GRR Bill.  However, I doubt if that would have been sufficient to transform an 8-point No lead into an 8-point Yes lead, especially given that only the very tail-end of the fieldwork would have been affected.

There are party political voting intention numbers in the Survation poll, which are a bit of a curate’s egg from a pro-indy point of view.  Although they show the SNP falling a few percentage points short of the self-defined target for victory on the popular vote in a Westminster election used as a de facto referendum, they nevertheless add further weight to the findings of other recent polls suggesting that the Labour surge has either stalled or gone very slightly into reverse – which is crucial, because at least in terms of seats in a first-past-the-post election, Labour is where the threat lies.

The SNP’s projected seat losses can be mostly explained by the regional list vote – which means they should be taken with a pinch of salt, because Survation always seem to understate the SNP on the list ballot.  In fairness, Survation appear to have made an attempt to address that problem by changing the wording of the question they ask about the list, making it clear that you can if you wish vote for the same party on both ballots, thus by implication stressing that the list is not a second preference vote.  But for some reason that hasn’t made any obvious difference to the results.

In case you’re wondering about Alba’s absence from the list results, it looks very much from the datasets as if Survation have dropped Alba as an option – which is really odd, because they included Alba in their recent propaganda poll for Scotland in Dungeon.  Indeed, Alba registered in that poll with 2% of the list vote.