
Ah. Whoopee. Another Survation. Whether due to an unwillingness to commission other pollsters, those firms still working on their post-GE methodology tweaks, or sheer force of habit, Survation and Norstat fair have the Scottish polling landscape sewn up between them at the moment. This one (link to tables) perhaps falls in the “habit” category, coming from regular Survation-commissioners True North.
I hasten to add this isn’t a criticism of either Survation or True North, because Ballot Box Scotland has commissioned two polls in my existence… both from Survation! They’re a perfectly good firm. It’s more just that we could really do with some Savanta, Ipsos and/or YouGov output to spice things up. Having previously expressed a wish for more polling diversity, I apparently did so on a monkey’s paw, as the last poll of 2024 ended up coming from FindOutNow and was so, ahem, bizarre it led to me adding another firm to my “not worth bothering with” list (note that pre-GE24 they also had one of the biggest misses of any firms, in an election defined by big polling misses for England).
All of that said, there is one upside to having such a limited pool of current pollsters: it’s giving a pretty consistent series to compare against. Two Survation polls separated by two months nonetheless follow the same methodology and have the same house effects, and thus we can say substantial movements between them are likely real. Two polls from different firms separated by the same period may show genuine movement, or may simply show a difference in methodology.
The previous Survation covered the 1st – 15th of November 2024. Changes are shown as (vs that poll / vs last election).
Regional Vote
The general theme for this poll, as in so many lately, is of significant pain for Labour. The mirrored 4 point gain for the SNP and loss for Labour takes what had been a mere 2 point margin and widens it into a whole 10% chasm. Both figures are also outwith the margin of error, meaning we can be quite confident they are reflective of some degree of real movement rather than just statistical noise.
Everyone else though is in margin of error territory. For the Conservatives and Greens that’s down a point each; Survation have been a lowball pollster for the Greens lately, and this continues that trend. The Lib Dems go in the opposite direction with a single point gain, whereas Reform pick up 2% to functionally tie with the Conservatives.
The continued poor showing for Alba here justifies my decision not to include FindOutNow polls in my tracker. Whilst I’ve previously bemoaned the lack of polling diversity, part of that limited spread has been having one pollster with a known and notable over-exaggeration of Alba support. With FindOutNow putting them on an eyebrow-raising 6%, that would mean two clear Alba-exaggerating firms sitting in the average. If that was two amongst a half-dozen regularish pollsters, it might have been fine. If this poll had also shown a strengthening of Alba, it would have been evidence of a genuine shift. Since FindOutNow would have made the average majority Alba-exaggerators, and this one hasn’t shown growth, I think that would have distorted the average too much.
I appreciate Alba folk already think I have a vendetta against them, but they have previously simply not amounted to anything. The idea that they weren’t profiting from the period of SNP collapse whilst led by the prominent-if-controversial Salmond, but are suddenly gaining in a period of SNP stabilisation and after Salmond’s death has left them without any serious figurehead simply does not pass the smell test. A few carefully selected council by-elections where they could benefit from high effort-low turnout dynamics don’t compare to a national election.
Constituency Vote
On the non-proportional side of things that is currently proving the most consequential, it’s a similar story of a sharp Labour slide against almost equal SNP recovery. Again that substantially widens the prior gap of 5% into a much chunkier 13%, the kind of lead that translates to dominance under First Past the Post.
For the other parties there are some slight differences with the list vote, with the Conservatives, Greens and Alba static, and the Lib Dems actually losing a point. I’ve said it before, saying it again: the Lib Dems have yet to have an election in reality where their constituency vote doesn’t lead their list vote quite substantially (in relative terms), so I’m always a little dubious when a poll is this way round. Reform on the other hand make slightly larger gains here that make their constituency vote equal to their regional, and again put them in what is functionally a statistical tie with the Conservatives.
Seat Projection
Projecting that into seats might give us something like this:
Please see this page for how projections work and important caveats.
My one-syllable summary of this projection is “ooft”. After the most recent Norstat saw big gains for the SNP and a sharp decline for Labour, I did wonder if that was something of an outlier. Given Survation are now backing them up, it seems it wasn’t. This isn’t as strong for the SNP or as weak for Labour as that Norstat, but it’s close, and it includes an eye-popping and proportionality-busting gain of 19 seats for the SNP relative to the last poll.
Combined with an unchanged 10 for the Greens, that would give a slender but possibly workable majority of 67 to 62 for the Pro-Independence bloc. It gives a similar lead for the SNP plus Lib Dems, who have been fluttering their eyelashes at one another over a possible budget deal. Whether the Lib Dems would be willing to offer such support after 2026 isn’t clear, but it’s an option, and perhaps one I should consider adding to the non-exhaustive chart below.
Speaking of the Lib Dems, same caution I always add to projections that think 10% lands them a Glasgow MSP: I’m not convinced. We had a Lib Dems on 10% election last year and they didn’t get any more than 4% in any constituency within Holyrood’s Glasgow region. The model says what it says though, and it’s a useful reminder models can never be made perfect. Note too that at 14 seats, this is a joint-lowest projection for the Conservatives from me this term.
The big problem for Labour here isn’t actually the lack of Pro-Union majority, but instead just the general direction of travel towards a big SNP lead over them. If, and that’s a big if, this general shape of things holds into 2026, even if the balance tilts marginally in the Pro-Union bloc’s favour, it’s very hard to see how Anas Sarwar credibly leads a government.
At some point Scotland may find itself in the circumstance of being led by a party that didn’t win the most seats. That is an entirely legitimate outcome of a proportional voting system. However, you can just about sell that to the public on a 50 SNP vs 40 Labour seat tally or something like that. Trying it on what could be a 50+ vs 25 split just looks staggeringly undemocratic to most people, realities of PR be damned, never mind what could be a very messy reliance on Reform UK to actually govern.
Possible Majorities
Note: these majorities relate simply to passing a vote in the Scottish Parliament. They do not imply the formation of a full coalition government.
- Traffic Light: Labour, Lib Dem and Green
- Independence Bloc: SNP, Green and Alba
- Grand Coalition: SNP and Labour
- Union Bloc: Labour, Conservative, Lib Dem and Reform UK
In a sign of just how bad the rest of the poll has been for Labour that a net swing of 3% to the SNP for Westminster feels relatively mild. Certainly, another -4% dip for Labour will concern them, but hey, it’s “only” a 9% lead for the SNP – not even in double digits! Given the spread of Labour’s vote it’s also possibly the kind of margin that leads to a near tie in seats rather than a big SNP majority.
What’s most remarkable here is that Reform narrowly overtake the Conservatives to pull into third place. This is simply the opposite case of their Holyrood figures, and is similarly a statistical tie, but it’s a moment that will undoubtedly have the Conservatives spooked, following on from identical figures for both parties in the most recent Norstat. The Lib Dems were perhaps abnormally low in the last poll and bounce back to something more in line with their Holyrood support, whilst a mild Green oscillation ticks them down a point.
In the tables published thus far (Thursday lunchtime, when I had time to write the last bits of this up and schedule it), there do not seem to be any Independence figures. If it turns out Independence wasn’t asked about at all I’d be very surprised and frankly a bit annoyed, but the Parliamentary stuff is the most important at the moment so it wouldn’t be the end of the world.
Hypotheticals
As ever, the last little bit of analysis concerns those hypothetical and more proportional voting systems that BBS likes to play about with. The use of pure FPTP at Westminster is an affront to democracy, and though Holyrood fares far better, AMS is still deeply imperfect. The examples here simply transpose the poll findings onto more proportional voting systems – the reality is that different systems would of course result in different voter behaviour.
I often let these sections pass largely unremarked these days, but occasionally it’s worth actually talking about. This purely proportional model shows how much Labour’s losses and the SNP’s gains in the constituency vote are distorting proportionality. The SNP simply should not have gained 19 seats from +4% of the proportional vote, and Reform UK should not have lost 2 seats from +2%.
The only reason that happened was the SNP absolutely mangling the intended proportional workings of the system with strong projected constituency leads. When it’s fully proportional however, the SNP’s seat gains become a much more modest 6. Likewise, Reform actually make gains here as they should with a higher vote share.
Scandinavian Style Westminster
If you find this or other Ballot Box Scotland output useful and/or interesting, and you can afford to do so, please consider donating to support my work. I love doing this, but it’s a one-man project and takes a lot of time and effort. All donations, no matter how small, are greatly appreciated and extremely helpful.
(About Donations)