First post-Boris Scottish poll puts the SNP on course for dramatic gains from both the Tories and Labour

The wait is now over for the first full-scale poll of Scottish voting intentions since Boris Johnson entered Downing Street.  It's YouGov that have broken the duck, and the figures they've produced neatly prove the point I made a few days ago that you can get a reasonably good idea of the state of play simply by averaging several YouGov subsamples. The following is strikingly similar to the average I published...

Scottish voting intentions for Westminster (YouGov):

SNP 43% (n/c)
Conservatives 20% (n/c)
Labour 15% (-2)
Liberal Democrats 12% (+3)
Brexit Party 6% (+2)
Greens 4% (+1)

The seats projection suggests the SNP would take 51 seats (up 16), the Liberal Democrats 4 seats (no change), the Conservatives 3 seats (down 10) and Labour 1 seat (down 6).

On the face of it, there may appear to have been no change in public opinion since Theresa May was in office.  However, the percentage changes listed above are from the last comparable poll in late April, when the Brexit Party surge hadn't quite reached its height yet.  The reason things looked so desperate for the Scottish Tories just before May's departure is that Farage had eaten directly into their support in a way that the SNP hadn't been able to.  It looks like the Boris effect has clawed back some of that ground, which is why the Tories are 'only' looking at ten losses, rather than eleven, twelve or the whole lot.  Perhaps they might still be able to limit their losses further, but for that to happen they're going to need to squeeze Brexit Party support some more (or hope that Farage doesn't put up candidates in selected seats) and they'll also need to hope that the SNP lose ground to Labour and/or the Liberal Democrats, possibly due to some sort of Swinson bandwagon effect over the course of the campaign.  But as things stand, the SNP are polling an impressive six points higher than the result they achieved in June 2017.

It has to be said that Scottish Labour appear to be staring down the barrel of a catastrophe, one that they might never recover from.  Yes, they made a mini-comeback after being reduced to one seat in 2015, but on that occasion they had a much healthier 24% of the popular vote to use as a base to rebuild from.  If they slump to anything like 15% of the vote, surely some of their remaining voters are going to start to wonder if the game is up this time.  But they turned things around over the course of the short campaign in 2017, so we certainly shouldn't exclude the possibility that they'll do the same again.  Their fate is probably in the hands of the London leadership - it's hard to imagine Richard Leonard spearheading much of a fightback.

There are also Holyrood voting intention numbers in the poll.  Oddly, the summaries that have appeared on social media provide the constituency percentages and an overall seat projection, but not the regional list percentages.  So until the datasets appear, we'll probably just have to surmise the list numbers from the seat projection.

Scottish Parliament voting intentions (constituency ballot):

SNP 45% (-1)
Conservatives 23% (+1)
Labour 13% (-3)
Liberal Democrats 12% (+5)
Brexit Party 3% (-1)
Greens 2% (-1)

Seat projection:

SNP 64
Conservatives 25
Liberal Democrats 15
Labour 12
Greens 10
Brexit Party 3

Pro-independence seats: 74
Anti-independence seats: 55

PRO-INDEPENDENCE MAJORITY OF 19 SEATS

This is a very timely illustration of the point I've been making about the proposed Wings party, ie. that it's intended as a solution to a problem that doesn't actually exist.  We have a pro-independence majority at the moment, and current polling suggests that we're on course to hold on to it - indeed that we could increase it substantially.  The one and only thing missing from the YouGov seat projection is an outright majority for the SNP - they fall short by just one seat, and of course the only way of squeezing out that extra required seat would be to vote SNP.  Voting for a smaller party wouldn't help.

I've been puzzled as to why Stuart Campbell is so convinced that the pro-indy majority is likely to be lost in 2021.  Having spoken to him, it seems to be partly due to a misunderstanding of how the voting system translates votes into seats - he believes that if pro-independence parties have less than 50% of the vote between them, they can't win a majority of seats unless there's some kind of gaming of the system.  That isn't true, and indeed you can see in this poll that there's a very comfortable pro-indy majority in spite of the fact that the SNP and Greens only have a combined 47% of the constituency vote.  Although AMS is a proportional voting system, it's far from being perfectly proportional, and if the SNP remain dominant in the constituencies, it's entirely possible that a handsome Yes majority can be won on less than 50% of the vote, without any gaming at all.

Stuart also appears to be concerned that the Alex Salmond trial may turn voting intentions upside down before the Holyrood election takes place.  All I can say is that there's lots of "what ifs" between now and May 2021, and the Salmond trial is only one of them.  The closest thing to a precedent is the Jeremy Thorpe trial in 1979, which did have a negative impact on Liberal support, but not as big an impact as had been feared.  And that was in spite of the fact that Thorpe was still actively involved in frontline Liberal politics in a way that Salmond is not currently involved in frontline SNP politics.  Incredible though it may seem, Thorpe's trial was postponed specifically so he could stand as an official Liberal candidate in the 1979 general election, a factor which must have pulled down the party's national support.  Mr Salmond, by contrast, is not currently even an SNP member, which may help to minimise any fallout.  But time will tell, and none of us have a crystal ball - not about the Alex Salmond trial, and not about the economic impact of Brexit, which is more likely to work in the SNP's favour.

You might remember that when the Ashcroft poll a few weeks ago showed a slim majority in favour of independence, I pointed out that there was no earlier Ashcroft poll to compare it to, and that we therefore didn't know whether there had been a very recent boost for Yes caused by the advent of Boris Johnson and the rising chances of No Deal, or whether regular polling by Ashcroft would have shown much the same picture during the closing months of Theresa May's tenure.  The new YouGov poll gives the impression that the latter is more likely to be true, because public opinion on independence appears to be unchanged since April.

Should Scotland be an independent country?

Yes 49% (n/c)
No 51% (n/c)

In some ways that's a good thing, because it suggests the boost for Yes reported by Ashcroft isn't a transitory bounce caused by a new PM, but instead has been with us for months and has been sustained.  49% for Yes remains well above the 'normal range' of 43-45% that YouGov reported throughout 2017 and 2018.  As I always point out, Panelbase and YouGov are both on the No-friendly end of the spectrum, so if YouGov are showing 49%, it's perfectly possible that another pollster (like Ipsos-Mori or Survation) might show 51% or 52%.

At the end of the day, YouGov and Ashcroft are essentially reporting the same thing: that the public are split down the middle on independence, and that the race is a statistical tie, ie. it's impossible to know who is really ahead due to the margin of error.