Scottish Westminster update - October 2018

Background: a new batch of Scottish polls

For this update on Scottish Westminster voting intentions and seat projections I will rely on the same three October polls I used for my Holyrood update. As a reminder they're here: Survation for the Sunday Post (fielded 28 September to 2 October), Panelbase for the Sunday Times (fielded 28 September to 4 October) and Survation for the SNP (fielded 3 to 5 October). 

Here is what the three individual polls say and their weighted average. The result is slightly less good for the SNP than what we had in June-July but again better than the 2017 results.

Bear in mind all three polls were in the field between the Labour Conference and the SNP Conference. So they tell us nothing about a possible post-Conference bounce for the SNP but they tell a lot about the Magic Dick Effect for Labour. Which is a 2% swing from Labour to the SNP compared to the 2017 results. So much for Richard Leonard being the Next Scottish Wonder and Killing Off The Nats.

Then this only confirms the trend we have seen all along in 2018 polls: the SNP always the first party and Scottish Labour always unable to do really better than their 2017 performance. Surely some polls have been better for the SNP than what we have now. But being 2% up from 2017 is still better than being 2% down. Or isn't it, Richard?

So now might be a good time for the Scottish Branch Office of the English Labour Party to reassess their priorities for good. What does Scottish Labour actually stand for when bedding with the Tories in Scottish Councils? And shouldn't they tone down the 'SNP Bad' rhetoric when 31% of their own voters support Scottish Independence according to the last two Survation polls?

Turns out Scottish Labour have only themselves to blame for making themselves irrelevant. So now Scotland has turned into a Tory-SNP one-on-one. Which is exactly what the SNP should wish for.

But who's gonna be my MP now?

Here is what these three October polls would deliver. Seats by SLLM rating first and then the summary.

On this polling six seats would change hands. Five from Labour to the SNP and one from the Conservatives to the SNP. Again these results show that Conservative and LibDem MPs prove harder to unseat than Labour ones. But also that some of the projected SNP gains are quite fragile.

For the record here is the seat projection each 2018 poll would have delivered individually. Some far better for the SNP than current polling and some far worse. But in the end the big picture points to a SNP surge which was certainly not a done deal when facing all the recent anti-SNP propaganda.

Ending up six seats up from the last GE is obviously an excellent result for the SNP though we have seen better not too long ago. But as always the SNP need a fair assessment of which of their seats are actually in the danger zone and which opposition seats are ready to fall, and how this should dictate the allocation of resources and manpower in the next campaign.

So what could go wrong?

Right now 12 Scottish seats qualify as marginals. Noticeably down from the 2017 result when we had 20 such. And with only one sitting SNP MP in the danger zone but then he always was.

Oddly current polling being less favourable for the SNP than the June-July polls has also moved the frontier between marginals and safer seats deeper into Unionist territory. Seven Unionist seats are now endangered, including two LibDem seats that looked safe last time aboot. You never know what can happen in marginals.

After reallocating the marginals here are the possible ranges of seats current polling delivers for each party. Good again for the SNP as even the worst case scenario has them up from 2017. And as always nothing should be taken for granted, especially seats that might shift on some hundred votes.

And finally the full breakdown of seats, as usual from the worst to the best projected results for the SNP.

What to expect now?

Probably no more fresh Scottish polls for a long time, unless we finally get a late November snap GE. Which I don't think of as the most likely scenario right now. Then current good polling should not get the SNP into the false comfort of over-confidence. We all know how quickly a seemingly favourable situation might take a turn for the worse. So the SNP have to be ready for any upset and get prepared accordingly.

First and probably most important step is a massive GOTV effort by the SNP. Never forget that turnout dropped by 261k between 2015 and 2017 and there is every reason to believe these missing voters were massively SNP-leaning. If these voters had showed up the SNP would likely have cut their losses by half. And we also have repeatedly seen how differential turnout can hurt the SNP in Council by-elections, even in seats that on paper looked like a done deal.

Second point is to make sure the next GE campaign stays focused on the proper issues. It will not be about how the SNP do in Edinburgh but solely how Tories do in London. So let's not repeat 2017 and get cornered into discussing the Scottish Government's performance on devolved matters. The SNP has to be on the offensive on reserved matters, most prominently how their awful handling by the English Government also deeply hurts Scotland. 

And as always stay tuned for further broadcasts.

Cha togar m' fhearg gun dìoladh

© Calum Macdonald, Rory Macdonald 1987

Holyrood projection - October 2018 update

More Scottish polling at last

Seems there has been quite a frenzy to feel Scotland's pulse again recently as we had three new polls in quick succession. Oddly all were fielded before the SNP Conference so we will have to wait for further polling to know if there's any Conference bounce, or not. Interestingly all three follow the 'poll till you boak' pattern that Survation already used, with detailed data on the Westminster vote, the Holyrood vote and the Second Independence Referendum including Brexit options.

For full results and crosstabs just click on the links: Survation for the Sunday Post (fielded 28 September to 2 October), Panelbase for the Sunday Times (fielded 28 September to 4 October) and Survation for the SNP (fielded 3 to 5 October). Don't worry too much about the sometimes significant discrepancies between the three as I will be using weighted averages which are designed to iron out the outliers.

So with three different and equally meaningful issues being addressed in these three polls I had to choose which one to start with. And that will be the Holyrood voting intentions because I have this hunch they're the ones that currently matter most.

So what do Holyrood voting intentions tell us?

First and foremost that the Early Autumn polling is pretty close to the Early Summer polling. Like all the shiteslinging that happened since just left no trace, be it 'Salmondgate' or the truckloads of lies and insults hurled at the SNP by the various Unionist factions and their friends in MSM.

Here is what the three individual polls say. As always there are some discrepancies like on the list vote for the Greens. Also Survation insist on publishing their result with one decimal pace, which is quite irrelevant when you have like 3% MOE on any given poll, but I always oblige them.

With all these caveats, here are the weighted averages we now have:

Not as good aw we could hope for the SNP as they are down on both votes from 2016. But still good enough for promising results as we shall see later. Even the breakdown of the list vote by region has barely moved and South Scotland remains the odd one out with Tories in the lead as on the Early Summer polling, again for the first time since 2003.

Finally the trends on both votes show that, despite ups and downs, the Scottish electorate are still happy enough with the SNP to grant them sizeable and improving leads. Though of course the SNP should never take anything for granted, especially votes when the highest factor might very well be the differential turnout again.

The constituency results

On this polling here is how the 73 constituencies project on SLLM rating.

Only one seat would change hands: Perthshire South and Kinross-shire (Roseanna Cunningham) switching from the SNP to the Conservatives. Guess this one would offer the BBC some Portillo Moment re-enactment but it would be only a marginal Tory gain, so not a given if the SNP hear the warning and campaign hard enough in that seat. Then we would also have seven marginal seats that might go either way.

Raw math says that John Swinney and Richard Lochhead, among others, might also have a harder time holding their seats than expected. But the upside could also be unseating Trident-lover Jackie Baillie. You never know what might happen in marginals.

The regional lists results

As usual the bizarre workings of AMS would trigger more changes on the list seats. Here are the results we get by region on current polling. This time the Liberal Democrats and the Greens would come out as the winners with respectively three and two gains and no losses.

For the other parties there would again be some sort of musical chairs with seven seats overall changing hands. Right now the list MSPs notionally losing there seats would be Maree Todd, Paul Wheelhouse (SNP), David Stewart (Labour), Alison Harris, Dean Lockhart, Tom Mason and Maurice Corry (Conservatives).

The big picture

Finally here is what the next Scottish Parliament would look like on this polling, and how the parties would perform in terms of net gains and losses. The Liberal Democrats and Greens would score only gains, the SNP only losses, while Labour and the Conservatives would have a mix of both.

These results are generally good for both the SNP and the Yes movement. The SNP is still the only party able to form a government and the pro-Independence majority gained one seat from where we were on June-July polling, just one seat short of the 2016 result. And we have further confirmation that the SNP are on an upward trend, at least as far as Scottish Parliament elections are concerned, and no longer the downward trend we saw in 2017.

Full breakdown of seats by region

The usual stuff. Not much to add here.

Just note that some surprises may happen with the seventh list seats as five of them would be won by less than 1,000 votes. Surprisingly the shakiest would be the LibDems' seventh in Highlands and Islands, won by just 100 votes in a region where they otherwise do quite well.

And now what next?

You might remember that a month ago, the conclusion of my last Holyrood projection was that I did not expect 'Salmondgate' or the various 'SNP Bad' made-up controversies to have any effect. And that's exactly what happened, with the SNP even slightly up on both votes.

Not that the SNP should lapse into complacency because this batch of polls is good. The next one might not be so. And we all remember how well "We don't take anything for granted…. except your votes" worked in 2016 and especially 2017. So let's just avoid that next time.

More than ever stay tuned for further broadcasts.

Nemo Me Impune Lacessit

Westminster projection - 09.10.2018 update

Eyes Wide Open

Conference Season often brings out the worse out of the parties and this year was no exception. So the LibDems want a second referendum but don't want the other. Tories still haven't made up their minds about austerity and when to throw their own PM under the bus. And Labour are like "we don't rule out not ruling out killing off the Nats, or not". Nice.

So the party currently in charge have lost all sense of direction except the one taking the whole UK over the cliff. And the ones challenging them know no better than bashing Scotland because they finally realized many of their English members actually have a sympathetic view of the SNP and what they stand for. Sad.

Then all have been expecting some sort of Conference bounce. But looks like the Brighton bounce was wiped out by the Liverpool bounce which in turn died down after the Birmingham bounce. And in the end we're pretty much where we were not so long ago. Again.

Polls continue to paint a bizarre picture, swinging back and forth between the two main parties almost at random. Though they now do show a smallish Conference bounce for Labour, but not enough for a decisive victory. Yet?

Trends definitely show Labour closing in on the Tories but at a sluggish pace. They still end up in second place and at this rate they would need months to finish first. And they certainly don't have that sort of time if the snap GE actually happens in the wake of Brexitocalypse. Or earlier if Mogglodytes finally make up their minds on their home movie version of Night of the Long Knives and when to stage it.

Now more than ever it's up to Little England to make up their minds. London, Wales and Scotland have already chosen. London and Wales want to be represented by Labour. Scotland wants to be represented by the SNP. Only Little England is still sitting on the fence and still won't swing decisively one way or the other. More on this later.

Pictures of a City

My updated super-sample includes six polls fielded between 26 September and 5 October, so partly overlapping the Conference season. Sample size is 10,364 with a 0.93% MOE. And here is what we get:

Statistically we have a tie between Labour and Tories. We've been down this road before and already know it doesn't end well. And again only the SNP and the Liberal Democrats are doing well enough to improve their representation significantly.

Of course the current uncertainty also has a lot to do with how people see the Lady In Charge and the Pretender. This one is still really not good for Jeremy Corbyn as 'Preferred PM' polling shows. And these are the numbers after undecideds have been cleaned off. Because the actual preferred PM is either 'None of the above' or 'Don't know' depending on which poll you read last.

So public opinion send a mixed message. Not really sure they still want Tories in charge but still sure Jeremy Corbyn is a worse pick than Theresa May. Public opinion not really sure of what they want is no novelty and MSM these days are certainly not helping. When all you're after is soundbites all you get is soundbite politics and a soundbite answer to that.

One More Red Nightmare

Now here is the projected House of Commons on this polling. Tories 25 seats down but still the first party and Labour only 7 seats up so not yet on a winning streak. As was the case a while ago when we already had a tie between Tories and Labour, the net effect would be to make the UK ungovernable unless some unholy alliance happened. Like a formal Lab-Lib coalition loosely supported by the SNP, Plaid Cymru and the Greens.

Such a result would only be an invitation to hold another snap election quickly to get a working majority. So 2019 could become one of those Years Of Two Elections like 1910 and 1974. Not that a second election would necessarily deliver a much different result, December 1910 did not. But Tories and Labour would probably agree on holding one anyway.

For the record, and one last time, here is what Electoral Calculus' Regional Predcitor projects when fed with the same polling data. Better for the Conservatives at face value but it's not like five seats would make much of a difference in this kind of stalemate situation. They would if they were the last five seats needed for a majority. Which in this case they aren't.

Just like last time only a few seats away from my models, and still showing mine just a wee smitch better for the SNP and LibDems.

Fallen Angel

Under current polling just 32 seats would be changing hands. Still far below what is needed for a clean change. And again the SNP being 11 seats up and the Liberal Democrats 6 seats up would make them the winners in comparison to the 2017 results. Unfortunately for Labour these would probably make them gaining only 7 seats overall less of a headline.

Even with Labour starting to make inroads deeper into Tory territory no senior member of the Cabinet would be in real danger. Just a few PPSs or PUSSs going down would not be a political earthquake especially when they hold positions the general public doesn't even know exist. And would anyway wonder 'what the fuck is that for?' if they knew they existed. And don't even get me started on what people might think a PUSS does for a living.

So MSM would have to be content with former glories like Soubry, Villiers and Rudd going down for their headlines. Though we all know some sort of totally unexpected upset might happen and probably will under the proper circumstances. Possibly in Wales or London if Labour somehow get their act together there.

Due to the lack of even a few juicy Portiballs moments, the most interesting situation here might be East Devon. There Councillor Claire Wright already stood twice against scandal-prone Tory MP Hugo Swire on an anti-austerity and environmentalist platform. Third might be a charm if she stands again so she might very well unseat Swire next time.

Technically Wright stands as an Independent but she sits in the same group as the Greens on Devon County Council, which is why I added her to the Greens' tally in my summary. And if she chooses not to stand next time, there is still a strong possibility Liberal Democrats could swing the seat as their voters provided the bulk of Wright's votes both times she stood.

The Law of Maximum Distress

Right now we have 52 ultra-marginal seats that might swing the election either way, 30 of them in England outside London. But it's not necessarily good news for Labour as Tories might have better chances for a majority in most of the English marginals.

Of course Labour might have some good moments like gaining back Copeland or Morley and Outwood. Best one could be unseating Iain Duncan Smith. But the general climate right now makes these less likely than Tories holding a few of these marginals by a hair.

Three of a Perfect Pair

Even with all the marginals reallocated to one side or the other, current polling still delivers an unstable situation with all possible majorities coming out of likely shaky deals.

Even a peak Tory performance would fail to get them a majority and the continuation of the DUP deal would only result in a three-seat majority. And in the end it might very well hinge only on DUP actually unseating Sylvia Hermon in North Down and the Lady might prove resilient enough to avoid just that.

Not that a peak Labour performance would look better. Being the first party would deliver fuck all unless they find a way to some loose big tent coalition. Which might be the recipe for an early failure to get anything done and pave the way to yet another snap election.

Fracture

Let's have a look now at the Geographical Divide across the UK. It's just as obvious now as in was in 2015 and 2017, even with voters slowly switching against the Tories. Scotland is still decisively yellow and London and Wales different shades of red. While England has somewhat moved away from Tory heartland to opposition-not-too-unfriendly territory, but not in the way that would benefit Labour most.

I left out Northern Ireland as different rules and voting patterns apply there, as Karen Bradley recently discovered. Anyway their 18 seats would not change the general landscape dramatically except for the amount of bribe the DUP would need and the kind of blackmailing they would allow themselves to indulge in to keep a Tory government on life support on a month-by-month basis.

Even if the Conservatives' lead in England outside London has shrunk dramatically from 2017 (1% down from 8%), Liberal Democrats are the ones actually snatching most of the disgruntled Tory voters (bagging 10% up from 4%) as Labour are barely level with their 2017 vote there. Which of course doesn't help gaining the number of seats needed for a successful UK-wide upset and confirms what kind of new direction Labour strategy should take.

Note too that Scotland has the highest anti-Tory voting intentions in all of the UK outside of London. Too bad Scottish Labour choose to ignore it and make the right decisions as they keep thinking that bashing the SNP and bedding with Tories on Councils is their best strategy. When it obviously isn't as they're still losing ground across all possible crosstabs pollsters can think of.

By the way we also know that Scotland is the most politically aware of all. Official statistics show we have the highest voter registration of all the UK though unfortunately it does not always translate into higher turnout. Which is something the SNP especially have to work on if we want actual election results to reflect the polls. And I can't help mentioning we also have the highest membership population-wise for any party. That is the SNP with 125k members or 3.2% of the Scottish Electorate.

The Divide is actually not just between England and the rest of us, neatly following Offa's Dyke and Hadrian's Wall. It actually neatly cuts midway through England. BMG Research routinely crosstab voting intentions with the NUTS-1 regions, aka the now defunct European Parliament constituencies. And here is what the average of their last four polls delivers, with the usual caveats about smaller subsamples. Plus the fact that BMG tend to be more Labour-friendly than others like YouGov so the actual situation might be 1 or 2 points more Tory-leaning depending on the region.

So we see the Divide again between Southern Little England and the Northern Powerhouse, broadly along the Severn-Wash Line as we've already noticed. And the more you move up towards The Wall, the more you wander into Labour heartlands. From deep blue South to still slightly purplish Midlands to rust red North.

Just bear in mind that Southern Little England has 44% of the English electorate outside London while the Labour-leaning North has only 33% and Midlands 23%. So a massive Tory victory in Little England, which FPTP makes more than likely, might be just enough to keep them in power no matter what.

We'll Let You Know

What will happen next is as always a matter of conjecture. The next batch of polls might actually show a Labour bounce big enough to get out of the current deadlock. Recent polling also shows that Boris Johnson becoming PM would boost the Labour vote. And so would Labour firmly committing themselves to a second EU referendum, whatever this would actually mean.

The combined effect of all these factors would probably be a 3% to 5% swing from Tories to Labour. And all of this happening more or less simultaneously is not that far fetched when you think it through. The irony being that it could actually mean the next GE would be decided by just a few votes and a few seats. And Scotland might turn up being the key to Number 10. Like 1974 all over again.

So the SNP might find themselves in the unique and bizarre situation of having to wish for Labour doing extremely well in England. Like a tsunami of Blairite proportions that would make half a dozen Scottish seats irrelevant as the outcome would already have been decided elsewhere. And even that might not be the weirdest situation facing us. Guess we have some interesting moments coming up before the next GE.

Best educated advice is to stay away from Ladbrokes for now. Odds are roughly 6-to-4 for Tories and going down, ain't good enough to risk serious dough. Just wait for further polling if we see some more solid direction in the next batch.

So stay tuned for further broadcasts.

Wha daur meddle wi' me

Umphrey's McGee's tribute to John Wetton (1949-2017)

© Robert Fripp, 1974