Storming start to 2021 as Yes scores second-highest ever vote share in an online poll

The final Scottish poll of 2020 was the first in the new regular series from Savanta ComRes for the Scotsman newspaper, and it was truly extraordinary - it had Yes on 58%, which was 2% higher than in any previous online poll from any firm (although it 'only' equalled the best ever result in a phone poll). The first poll of 2021 is the second poll in the same series, and it's only 1% shy of being a repeat performance - which of course means it's the second best ever for Yes in an online poll.

Should Scotland be an independent country?

Yes 57% (-1)

No 43% (+1)

I know some people may actually be disappointed by this result, because they would have expected the reality of a very Hard Brexit to inevitably increase Yes support from wherever it had previously been.  We've seen hypothetical polling questions over the years suggesting that Scotland's forced departure from the single market and the customs union would indeed have that effect - but the thing is that people are notoriously bad at answering hypothetical questions, ie. they can't necessarily predict with any great accuracy whether their opinions will change in a different context.  In any case, my guess is that people are currently thinking far more about the pandemic than they are about Brexit.

That said, this may not be the ideal poll with which to judge the trend.  Because the previous poll in the series was a record-breaker, it's impossible to discount the possibility that Yes support was exaggerated due to sampling factors, in which case you'd expect some kind of reversion to the mean this time.  We'll have to wait for further polls to be sure of the direction of travel - but we can certainly rest easy in the sense that it looks like support for independence is still running at an exceptionally high level.

Obviously the main cloud on the horizon for the Yes movement at the moment is the open warfare between Nicola Sturgeon and Alex Salmond (and perhaps just as importantly between their proxies/supporters). The poll was conducted between the 8th and yesterday, which means it just about took place in its entirety after the story broke - but the impact may not have been fully factored in on the first day of fieldwork, which is when most people respond to online polls.  The forthcoming Scot Goes Pop poll will have later fieldwork, so that might give us a better clue as to any effect on public opinion.  In all honesty I'm a bit nervous about it, because it'll be conducted by a firm that hasn't been quite as favourable for Yes as ComRes has recently, but fingers crossed there'll at least be some sort of Yes lead.  I'm quite proud of the fact that a Scot Goes Pop poll in June started the run of seventeen/eighteen polls in a row showing a Yes majority, so I'll be a bit miffed if it's also a Scot Goes Pop poll that ends the sequence, but hopefully that won't happen!

ComRes also have Holyrood voting intention numbers - 

Scottish Parliament constituency ballot:

SNP 53% (-2) 

Conservatives 19% (-1) 

Labour 18% (+2) 

Liberal Democrats 6% (-)

Scottish Parliament regional list ballot:

SNP 44% (+2) 

Labour 18% (+1) 

Conservatives 16% (-4) 

Greens 11% (-1) 

Liberal Democrats 8% (+1)

In recent times it's mainly been Survation that has shown Labour ahead of the Tories, so there may be some significance in the fact that another firm now has Labour in the runner-up spot in the more important list ballot.  But of course the real story is one of Tory collapse rather than of Labour recovery - returning to main opposition party status on a pathetic 18% of the vote would be a pyrrhic triumph for Starmer and Leonard by any standards.

On past form the ComRes question will have described the list vote as the "second" vote, which I've always suspected leads to an overstatement of the Green list vote, and an understatement of the SNP list vote (because a minority of indy supporters interpret it as a 'second preference' vote).  I very much doubt that the Greens are really in double figures.