For the second time in a few days I have to query the editorial priorities of the Herald in dealing with opinion polls - their take on the new Survation poll is "Nicola Sturgeon faces strong public opposition to Indyref 2", which is a tad odd, because the poll actually shows that 60% of respondents want a second independence referendum at some point, rising to 65% if Don't Knows are removed. The only negative side of the poll for independence supporters is when respondents say the referendum should be held, with many saying it shouldn't be for at least five or ten years. However, for my money that has a lot to do with the way the question is posed - it's a five-option question, which will lead strong supporters and opponents of a referendum to choose the option at either extreme, with everyone else likely to be drawn to an option somewhere in the middle to show their moderation. And of course it was entirely at Survation's discretion to decide what "in the middle" would actually look like. If you want a meaningful verdict on the referendum the SNP are proposing, you have to ask a more direct question - "do you support or oppose the SNP's call for an independence referendum once the terms of Brexit are known?" We can only guess what the outcome would have been if Survation had posed that one.
On the voting intentions for Westminster and Holyrood, it's pretty much unalloyed good news for the SNP and the independence movement more generally.
Scottish voting intentions for Westminster:
SNP 40% (+1)
Conservatives 24% (-2)
Labour 23% (-1)
Liberal Democrats 8% (n/c)
Like the recent Panelbase poll, the above figures represent a swing to the SNP from both the Tories and Labour since the June 2017 election, but the swings are bigger, and the SNP's own vote share is up on 2017 as well. In the unlikely event of a totally uniform national swing, this would see the SNP win 46 seats (up 11), the Tories just 8 (down 5), the Lib Dems 4 (no change) and Labour 1 (down 6).
In a perverse way, Labour might think their 23% vote share is not too bad, because it's within their normal range for recent Survation polls, albeit at the lower end of it. Panelbase, by contrast, had put Labour on a post-June 2017 low, which appeared to reflect the impact of the Independent Group breakaway.
Scottish Parliament constituency ballot:
SNP 43% (+5)
Conservatives 24% (-2)
Labour 22% (-3)
Liberal Democrats 9% (n/c)
Scottish Parliament regional list ballot:
SNP 32% (n/c)
Conservatives 22% (-1)
Labour 19% (-4)
Greens 11% (+2)
Liberal Democrats 11% (+2)
Remarkably, the Scotland Votes predictor suggests that the pro-independence majority in the Scottish Parliament would slightly increase on these numbers - the SNP and Greens between them would have 70 seats, rather than the 69 they won in 2016 when the mandate for a second independence referendum was secured. That once again gives the lie to Robin McAlpine's claim from a few weeks ago that all recent polls have suggested the pro-indy majority would be lost. In all honesty, though, Survation's methodology is making it very difficult to work out what would really happen, because it's been blindingly obvious for ages that the question wording they use for the list vote ("your second vote") is confusing respondents and producing distorted results - the SNP's vote share is likely to be a few points too low and the Greens' vote share is likely to be a few points too high.