SCOT goes POP!: 2026-04-05

Wednesday, April 8, 2026

Catastrophe for Starmer as Labour slump to joint all-time low in weekly YouGov poll, level with the Greens - the SNP have huge 18-point lead in Scottish subsample

YouGov have suddenly and belatedly started including Rupert Lowe's Restore Britain and Jeremy Corbyn's Your Party as options in their weekly GB-wide polls, which on the whole is a good thing because to some extent we've been flying blind about the impact of those two parties (especially Lowe's). The curious thing is, though, that you'd expect a non-trivial vote share for Restore Britain to reduce Reform's vote share significantly, but in this poll Reform's vote share has actually increased. A recent Find Out Now similarly found that including Restore made only a trivial difference to the Reform vote. The only logical explanation I can think of is that most Restore supporters were previously selecting 'other party' when answering polls - which would indicate that they're quite committed in their choice of party, it's not just a casual preference affected by what they see in front of them in a menu of options.

So on the whole you'd have to say this is good news for Farage, because it looks as if his support hasn't been exaggerated by recent polls in the way that had seemed a logical possibility.

GB-wide voting intentions (YouGov, 6th-7th April 2026):

Reform UK 24% (+1)

Conservatives 19% (-)

Greens 16% (-3)

Labour 16% (-2)

Liberal Democrats 13% (+1)

Restore Britain 4% (n/a)

SNP 3% (-)

Plaid Cymru 1% (-)

Your Party 1% (-)

Scottish subsample: SNP 33%, Reform UK 15%, Labour 14%, Greens 11%, Conservatives 11%, Liberal Democrats 9%, Restore Britain 2%, Your Party 1%

In one sense a three-point drop makes this a very disappointing result for the Greens (and of course it's possible that Your Party's inclusion may have been a factor in that), but on the other hand what they probably care most about is where they stand in relation to Labour, and at the moment they're tied, which is good enough to be getting on with. I constantly have to remind myself what Labour's record low vote share is (because it keeps changing), but I'm pretty certain 16% is the lowest they've ever been with YouGov, so this is a return to an all-time low. No poll from any polling company during Jeremy Corbyn's leadership ever had Labour lower than 18%, so Starmer is unambiguously performing worse than his predecessor ever did.

The Iran war has (undeservedly) seen a partial recovery in Starmer's personal ratings, and some supplementary questions in polls have delivered a not too dreadful verdict on his response to the crisis. Some suggested that meant he and Labour were "having a good war", but it would be hard to argue that case now.

To reiterate the point that people always miss about YouGov's Scottish subsamples, they can be taken more seriously than subsamples from other firms because they're correctly structured and weighted. However, they do still have a big margin of error due to the small sample size, so you will still sometimes get wildly misleading results, as happened a few weeks ago when there were two subsamples in a row showing the SNP below 30%. This week's result is more routine, and is pretty encouraging on the whole - the unionist vote is split in an ideal way, and there's a substantial Green vote that the SNP will hope to squeeze on the constituency ballot next month, where in most cases the Greens are not standing candidates.

* * *

My latest constituency profile for The National is Glasgow Baillieston & Shettleston - a seat I can claim a tenuous connection to, because my dad was once an assistant headmaster of the long since demolished St Gregory's secondary school in Cranhill.

* * *

If you are enjoying Scot Goes Pop's election coverage so much that you start to feel an inexplicable urge to buy me a hot chocolate or a ham-and-cheese toastie, donations are very welcome. There are three main options:
a) you can donate by card HERE
b) you can make a direct PayPal donation to my PayPal email address, which is: jkellysta@yahoo.co.uk
c) you can make a donation by bank transfer - for the necessary details, please drop me a line at my contact email address, which is: icehouse.250@gmail.com

* * *

Over the last few months, I've been building up the Scot Goes Pop channel on YouTube - you can check it out HERE, and don't forget to subscribe.

If Trump was Chamberlain: "This morning I handed the Iranian Government a final TruthSocial stating that unless we heard from them by 8pm Eastern that they were prepared at once to open the f****n' Strait you crazy b******s I would wipe our their civilisation. I have to...
— James Kelly (@JamesKelly) April 8, 2026

...tell you now that no such undertaking has been received, but that Iran have sent me a neat 10-point counter-proposal, and that consequently, I might not bother committing genocide after all. Thank you for your attention to this matter!"
— James Kelly (@JamesKelly) April 8, 2026

Tuesday, April 7, 2026

More on the MRP poll: the 67-seat projection for the SNP may even be an underestimate due to the Greens being included in too many seats

Some more details of the Find Out Now MRP poll for The National have been published on the Electoral Calculus website. Whoever wrote the article (presumably Martin Baxter) doesn't seem to be fully versed in devolved Scottish politics, because references are repeatedly made to the "official opposition" party in the Scottish Parliament, which is a concept that does not actually exist. The leader of the largest opposition party is called first at FMQs, but apart from that there's no special status.

However, what matters is whether the numbers in the poll are accurate, and I've been looking through the individual seats projections with interest. The first thing that leaps out is that there is a Green vote share given for every constituency, when in fact the Greens are only standing in a handful of constituency seats. That might conceivably make a difference in two cases: in Caithness, Sutherland & Ross, the Lib Dems are projected to have a four-point lead over the SNP, and the Greens are on 4%, while in Edinburgh Southern, Labour are projected to be three points ahead of the SNP, with the Greens on 4%. So if you take the numbers in the poll absolutely literally, the 67-seat projection for the SNP could even be a slight underestimate.

On the other hand, the SNP are projected to be just two points ahead of the Lib Dems in Skye, Lochaber & Badenoch, which in 2021 returned the largest SNP majority in the country, with the Lib Dems in third. Now, of course that's not totally implausible, because the SNP will be losing any personal vote for Kate Forbes, and the Lib Dems surprised us all by winning the overlapping constituency of Inverness, Skye and West Ross-shire in the 2024 UK general election. But it does make me wonder about the assumptions baked into the poll's methodology. It's almost as if the Westminster numbers are being used as a baseline. I note, for example, that there's no sign of a Lib Dem breakthrough in the projection for Argyll & Bute, which is a similar constituency in many respects but not in one: there was no major Lib Dem recovery there in 2024.

The SNP are projected to hold on to the seats where the Greens are actually standing and are purported to have a decent chance. However, the Greens are in either second place or joint second in Edinburgh Central, Glasgow Southside and Glasgow Kelvin & Maryhill, and are a relatively modest 5-9 points behind the SNP in each. Intuitively, that strikes me as a much more plausible estimate of the state of play than we've been seeing in certain quarters.

Reform UK's best showing is projected to be 24% in Banffshire & Buchan Coast, which is presumably based partly on that being the most pro-Brexit constituency. More startling, however, is the 23% for Reform in both of the North Ayrshire seats (Cunninghame North and Cunninghame South).

* * *

My latest constituency profile for The National is Glasgow Anniesland.

* * *

Over the last few months, I've been building up the Scot Goes Pop channel on YouTube - you can check it out HERE, and don't forget to subscribe.

Even if he doesn't follow through on this clear threat of genocide, it strikes me that the threat itself constitutes the gravest of crimes. What does "terrorism" mean if not acts designed to instill terror in a civilian population? https://t.co/n04RyXnXYB
— James Kelly (@JamesKelly) April 7, 2026

Monday, April 6, 2026

Find Out Now - ZAP WHAM POW - that a new MRP poll shows the SNP on course for an overall majority

As you may have seen, a new Find Out Now MRP poll commissioned by The National is suggesting that the SNP are on course for an overall majority at the Holyrood election. They would take 67 seats, all on the constituency ballot. I've been trying to work out what those 67 are, or to put it more simply which six constituencies they wouldn't win, but so far I'm struggling with navigation in the table of results - if anyone can let me know in the comments section, that would be great. It's probably safe to assume that Orkney and Shetland will both be staying Liberal Democrat, but I wouldn't want to guess which other four constituencies are projected to be in the unionist column. The basic figures are -

SNP 67

Labour 17

Greens 14

Reform UK 14

Conservatives 10

Liberal Democrats 7

I have a new article at The National discussing the track record of past MRP projections, which you can read HERE.

UPDATE: OK, thanks to Michael and Keith in the comments section, we now know which six constituencies are projected to elude the SNP. They are:

Caithness, Sutherland and Ross (LibDems)

Edinburgh North Western (LibDems)

Edinburgh Southern (Lab)

Fife North East (LibDems)

Orkney Islands (LibDems)

Shetland Islands (LibDems)

That means the SNP are projected to enjoy a whole string of eyebrow-raising wins elsewhere:

Dumbarton: I'm struggling with this one. It should be a Labour hold on the basis of the swing in national polls, and bearing in mind the track record of tactical voting in the constituency, it's hard to see how the SNP take it.

Ettrick, Roxburgh & Berwickshire: An SNP gain is possible here, but it's like trying to thread a needle - the unionist vote would have to be divided almost perfectly. Unlikely in my view.

Dumfriesshire: A bit more plausible, but still a very tough one - the SNP are starting from ten points behind the Tories.

Galloway & West Dumfries: The most winnable of the three Blue Wall seats in the south, and the SNP have a past track record of success here, although on the basis of national trends they would still be expected to fall just short. But I can accept this one as a plausible SNP gain.

Banffshire & Buchan Coast: The SNP look vulnerable here to both the Tories and Reform, but it's by no means outlandish to think they'll hold on.

Aberdeenshire West: On paper this doesn't look promising for the SNP, but we keep hearing the Tory canvass results in the northeast are dreadful, so yes, this is a possible gain.

Inverness & Nairn: I don't think anybody really knows yet what the impact of Fergus Ewing's independent candidacy will be, so there's a big question mark on this one.

East Lothian Coast & Lammermuirs: Looked like a lost cause not that long ago, but could now be very close.

Edinburgh Central, Glasgow Kelvin & Maryhill and Glasgow Southside have all been touted as Green gains. As previously explained, those predictions are based on a smoke-and-mirrors statistical exercise and shouldn't be taken seriously, but with an effective Green campaign the SNP are not necessarily safe in any of the three, and they're also vulnerable to Labour in Edinburgh Central. Each of the three in isolation looks like a probable SNP hold, but is it really likely their luck will hold out in all three? Even if just one of the three were to go to Labour or the Greens, it would make winning an overall majority very tough.

So as you can see I'm still very sceptical about the prospects of a single-party overall majority, but believe me about one thing: I do want to believe.

* * *

My latest constituency profiles for The National are Fife North East and Galloway & West Dumfries.

* * *

Over the last few months, I've been building up the Scot Goes Pop channel on YouTube - you can check it out HERE, and don't forget to subscribe.

Bombshell GB-wide Ashcroft poll is the first not to show Reform in an outright lead for the first time in almost a year - Greens are in joint lead for the first time ever - Labour are in FOURTH - and the SNP have overtaken Labour UK-wide in the seats projection

So this is a genuine landmark for the reasons given in the title, and it's also worth making the point that the data tables (unless I'm misinterpreting them, but I don't see how I can be) show that the Greens are actually in a slight overall lead over Reform and the Tories - but that seems to have been disguised by the rounding to the nearest whole number.

GB-wide voting intentions for next general election (Lord Ashcroft, 26th-30th March 2026)

Greens 21%

Reform UK 21%

Conservatives 21%

Labour 17%

Liberal Democrats 9%

SNP 3%

Plaid Cymru 1%

I know somebody listed Scottish subsample numbers on the previous thread, but I can't see any in the data tables with the "don't knows/will not votes" removed. However, I've used what I presume was a rough recalculation to fine-tune a UK-wide seats projection, which shows: Reform UK 204, Conservatives 175, Greens 116, SNP 48, Liberal Democrats 47, Labour 33, Plaid Cymru 8, Others 19.

The target for an overall majority is 326, so it's not hard to see why a hung parliament is currently the strongly favoured outcome on the exchanges. Nevertheless, under first-past-the-post not all that much movement is required to transform an absolute guddle into a clear majority, and by the same token not much movement would be required to turn a projection showing a right-wing parliament, as this one does, into one showing a centre-left parliament in which the SNP might just hold the balance of power. Even if they don't hold the balance of power on their own, the huge strength of the Greens is a potential game-changer, because at the very least the English Greens are not opposed to independence.

On the exchanges, the Greens are currently estimated to have a 1 in 8 chance of winning most seats in the general election, but as the above numbers demonstrate, they might not actually need to win most seats to end up with influence.

An intriguing quirk is that the SNP are currently the fourth-largest party in the Commons (albeit only just, and they may soon be overtaken by Reform). The projection from this poll shows they would still be in fourth place, but in a radically different way - they would have five times as many seats as now, they would re-overtake the Liberal Democrats, and they would overtake Labour for the first time. Let's just reiterate that: the SNP would have more seats than Labour, UK-wide.

Ashcroft himself concedes that the reason his results might be different from other pollsters is that he has a completely different approach to the voting intention question - instead of directly asking people how they will vote, he asks them to rate their chances of voting for each party in turn. As I understand it, any respondent who does not estimate a 50%+ probability of voting for at least one party is assumed to be an abstainer and excluded, and everyone else is assigned to the party they gave the highest probability to. That method seems intuitively reasonable to me, but whether the results it produces will be more accurate, or less so, is anyone's guess at this stage.

For weeks after the Gorton & Denton by-election, YouGov were putting "footnotes" of sorts on their polls to give the impression that the Green advantage over Labour must just be a temporary effect caused by the by-election and would fade. There is now some doubt over that, not just because of this Ashcroft poll, but also because last week's YouGov poll showed the Greens moving back ahead of Labour, after having slipped behind for one week.

In case you're wondering, the last GB-wide poll not to show an outright Reform lead was a Survation poll in late April/early May of last year. That showed Labour and Reform tied on 26% apiece.

There is actually some relief for Starmer in the supplementary questions in the Ashcroft poll. It's generally believed that head-to-head leadership polls are more predictive of election results several years in advance than headline voting intentions, and Starmer does have a clear 15-point lead over Farage. However his lead over Badenoch is just three points, which amounts to a statistical tie - and Ashcroft doesn't even bother to ask whether respondents prefer Polanski to Starmer, which many will suspect is because he feared what the answer might be.

There are a couple of results that I actually found quite surprising. When asked whether nuclear power should be phased out, with wind power expanded and the net zero target brought forward a decade, respondents are almost split down the middle - 40% in favour, 45% against. My guess is that Ashcroft asked it as a "shopping list" question in the hope that most respondents would find something on the list to object to, thus producing a result he'd be able to spin as clear and decisive support for nuclear power, but that didn't happen.

And on Europe, there are any number of people who will tell you that if you spell out in a poll question what returning to the EU would actually mean in practice, the pro-EU majority evaporates. It looks to me like Ashcroft set out to prove that theory and spectacularly failed. When asked whether they want to rejoin the customs union, restore freedom of movement and then rejoin the EU itself as soon as possible, 55% supported the idea and only 34% were opposed. That's absolutely remarkable.

Ashcroft did manage to get a result which he can spin as showing massive opposition to scrapping the "nuclear deterrent", but as he lumped "and cut defence spending" into the question, the result is pretty meaningless.

* * *

Loopy billionaire lord tries to convince us that funding the NHS with fair taxation is as impossible as enhancing the size of women's breasts with hypnotherapy

I cannot in all good conscience conclude my discussion of this poll without drawing your attention to the fact that Ashcroft has made a complete blithering idiot of himself with one particular part of his write-up -

"Perhaps more controversially, nearly a third of voters said they felt less favourable towards Polanski when they heard that in his days as a hypnotherapist he once claimed he could increase the size of women’s breasts by hypnosis. Polanski claims to have apologised and put all this behind him, but in a different way he is arguably still at it. Just as there are those who want to change their body shape through the power of mind over matter, there will always be people eager to believe we can fund the NHS by taxing the rich"

Nice try, Mike, but you are believed to be worth £2 billion. That alone would be enough to fund 1% of the entire annual budget of NHS England. Quite plainly, taxing the rich could very easily fund the NHS - and the only use hypnotherapy would be on that front would be for those like you who don't want us to notice or believe a simple arithmetical fact.

* * *

Over the last few months, I've been building up the Scot Goes Pop channel on YouTube - you can check it out HERE, and don't forget to subscribe.