Jeremy's Den: May 2021

31/05/2021

Signal Lost In All The Noise

The world is a horrible, miserable, desperate place

And our only respite will be boozing juiced oblivion

Or the blessed relief of death, so welcome to the show

(Jon Richardson, Ultimate Worrier: Modern Living, 2019)

Labour’s problem in Hartlepool has been that they’ve had a problem

Matching Peter Mandelson’s working class credibility

(John Pienaar, Have I Got News For You?, 2021)

I guess that Labour spads are now watching polls closely, to identify the possible residual pockets of Brexit Party/Reform UK vote. Which they should if Keir Starmer is indeed preparing for the worst while hoping for the best. Which might take much more to materialise than Labour's current strategy, if the most recent polls are any indication. Then we must all remember that nothing of this would have happened if Theresa May had not totally fucked up her Brexit Deal, and dragged the UK into the totally unnecessary European Parliament election in June 2019. This, and this only, gave birth to the Brexit Party, which totally ridiculed the Conservatives, just seven months after its foundation. This election also conclusively turned the tide of voting intentions, and probably sealed Jeremy Corbyn's fate as much, if not more, as the controversies over Labour's anti-Semitism. It freed hundreds of thousands Labour voters from ancestral allegiances, over the one issue that dominated the political discourse for years, and made them legitimate targets for Johnson's opportunistic Conservatives. The Brexit Party bubble burst within six months, far quicker than the UKIP bubble before it, and the December election also shockingly burst Labour's bubble. Then Brexit's vote shares at both 2019 elections in each meta-region show they still had some backing around the Red Wall. Which is not surprising as they did not stand in Conservative seats at the December general election, but also a sign that Leave-backing Labour voters were still unhappy with their old party's Brexit equivocation. Which was as much Starmer's responsibility as Corbyn's, to be fair.

In December 2019, the Brexit Party polled above 15% in only 15 constituencies all over the UK. But the most ominous result, from Labour's perspective, is that the Brexit Party polled higher than Labour's margin of victory in 39 constituencies, 28 of them in Northern England, 5 in the Midlands, 4 in Wales, 1 each in London and Southern England. Hartlepool was one of these. If the Brexit vote dissolves into the Tory vote everywhere, à la Hartlepool, Labour is definitely fucked beyond salvage for the foreseeable future. There is right now every reason to believe it will, and the next generation will ask: "Red Wall? What Red Wall?". The Guardian's view on Labour's electoral future, or their conclusion that trends seen at local elections will spread to parliamentary elections, might sound a bit far-fetched for now, but stranger things have happened since 1832. So the whole situation is like a fucking time-bomb, and there probably aren't many ways to avoid the trap, unless Keir Starmer ups his game and learns to speak working-class. There is also massive irony in the fact that some of the most endangered seats are currently held by so-called "Blue Labour" MPs such as Jon Cruddas, who advocated the "faith, flag, family" strategy that failed so miserably already under Ed Miliband. Then Sly Keir will also have to admit that the Bright Young Things from Hampstead Heath are not the best-equipped to read the Andy Capps of this realm. Or the Scots. Not that we can expect that from the man who brought Peter Mandelson back into his Ninth Circle. The same Mandelson who once infamously said the Northern working class were Labour's forever because they had nowhere else to go. Well, now they have.

Labour keep on telling people off for getting it wrong

Instead of listening to the reasons they don’t like Labour any more

(John Pienaar, Have I Got News For You?, 2021)

Labour has got an enormous job, it’s gonna need, if not a new leader

A whole new way of doing it, and possibly both

(John Pienaar, Have I Got News For You?, 2021)

Another alarm for Labour is, or should be, how Keir Starmer does in the Preferred Prime Minister polling that is published at least weekly, and sometimes thrice weekly, by various pollsters. And it definitely does not look good. Fucking awful is actually an apter description. It's quite karmatic indeed that Bozo's fate might depend on Dominic Cummings's personal vendetta and how he can paint the First Minister of England as the mendacious buffoon he is. Though I'm honestly not holding my breath and don't share the Guardian's optimism. I tend to agree with that anonymous Tory MP who thinks it will all go away, again. So far, some of Bozo's supporters have switched to undecided, but it failed to benefit Keir Starmer, who is still on a downward spiral. There are also multiple signs that Keir's leadership, or lack thereof, has left many in Labour in complete disarray. There is massive irony in the Corbynistas now backing Andy Burnham in an hypothetical leadership challenge. Which of course is only second to Andy rebranding himself as the champion of grassroots progressivism, after his fights with the English Government about Covid delivered him a massive victory in the Greater Manchester mayoral election. This could create an interesting situation as Andy Burnham is no longer an MP, and some will unavoidably argue it disqualifies him as a potential party leader. Which is of course total bollocks as there is no rule anywhere saying this. What passes as a constitution in the UK is just a hotchpotch collection of bits and bolts that has no foundation in anything but unwritten conventions and Old Boys Club's agreements between the main English parties. Ironically, a precedent was set by the Scottish Conservatives when they had dual leadership between Ruth Davidson and Douglas Ross, even if it was just a temporary arrangement. But the person who least wants to see Burnham as Labour leader is probably Burnham himself, which makes the point moot.

A leadership challenge in the short term is probably not Keir Starmer's main concern, though it possibly should be, and could come from where you least expect it to. Sadiq Khan might try his luck, even if his reelection as Mayor of London was lacklustre compared to Andy Burnham's in Manchester or Steve Rotherham's in the Liverpool City Region. And of course Khan's opponents would argue that he is even more a symbol of the out-of-touch ultra-woke metropolitan elite than Starmer on his worst days. Then Keir's biggest worry these days should be his lack of effectiveness as Leader of the Opposition. Many said that the public's positive appraisal of his forensic interrogations of Boris Johnson at PMQs would soon wear off, and it is exactly what happened as he seems unable to pull serious punches. As the saying goes, it doesn't look good if you can talk the talk but can't walk the walk. Then Sly Keir's problem right now is that he can't even talk the talk. At least not the one his potential voters are expecting. It says a lot that Keir's rating as Preferred PM is lower than Labour's voting intentions, while Bozo's is higher than Conservatives' voting intentions. To add insult to injury, Keir Starmer is not doing any better when polled against Rishi Sunak. It's interesting to see Wunderkind Rishi being the pollsters' favourite, and indeed their only alternative to Bozo, at a time when the Treasury's dominance on the whole political agenda has been seriously dented, by an odd mix of actual political choices and reality checks on the actual requirements of facing a major crisis. Not coincidentally, this coincides with some within Labour advocating what you could call "Council Keynesianism", in opposition to Brownite orthodoxy.

It's also interesting to note that no pollster is polling Keir Starmer vs Michael Gove as the Preferred Prime Minister. Sarah Vine and the Daily Mail may be rooting for Sleekit Mickey, but obviously the pollstertariat and the punditariat don't consider him a serious contender for the top rung of the food chain. Then we might be stuck with Rishi as the heir apparent for some time, as the Murdoch press realise that both Mickey and Keir are definitely the wrong horses to bet on. There might be some interesting fallout though after Dominic Cummings' last batch of finger-pointing about herd immunity. Maybe Sleekit Mickey will come out of it as the next tenant of Number Ten, being the only one left unscathed by Dom's strafing runs, for reasons I cannot even try and begin to fathom. Can you? But we just have to remember that it is another game of smoke and mirrors, where everybody lies, and everybody were up to their oxters in it, including Cummings himself as chief adviser to the First Minister of England. Then it might also end up as just another ripple on the ocean, and business as usual will resume, while the endless procession of sycophants are all over the media telling us what a good man Carrie's husband is. And we can sit back and watch while mandatory photo ID and the next Periodic Review of Westminster constituencies make sure that the Conservatives stay in power for all of eternity, the English Variant of the Thousand Years' Reich.

Rishi Sunak is not competent, he just seems competent

Because everyone around him is profoundly incompetent

If you were in a lift with ten people and one of them farted

You’d think “that’s disgusting”, but if you looked at the other nine people

And saw they had liquid shit dribbling out of the bottom of their trousers

You would think the guy who farted is a champion of decorum

Rishi Sunak is a lift farter in a sea of pant soilers

(Nish Kumar, The Last Leg, 2021)

Here’s the thing with leadership

Once you have seized power, really all you have to do is coast

And preside over the decline and blame others

(Richard Ayoade, 8 Out Of 10 Cats Does Countdown, 2019)

There is another poll Labour spads should study more closely, the one conducted by Savanta Comres between 7 and 9 May about what they call "party characteristics", but should be more fittingly described as people's attitudes towards the parties. Of course Savanta Comres totally fucked up the last instalment of their Scottish Tracker for The Scotsman, as I reminded you in my postmortem of the Holyrood election. But their general election polls do look more reliable, or at least not blatantly at odds with the findings of other pollsters. Their first batch of items are broadly about who is better represented by either the Conservatives or Labour. And, while Labour score reasonably well on a number of items, the Conservatives also do, and sometimes in a surprising way. Labour spads should be worried about a tepid result as "the party of the young" and a truly lousy one as "the party of aspiration". Both being items on which you want to score high if you plan on painting yourself as "the party of the future". Also intriguing is that almost as many voters think that Labour is not "the party of the working class", as who think the Conservatives are. Almost as bad as nearly a third of respondents thinking that the Conservatives are "the party of the left behind", despite their abysmal record on welfare and benefits. Definitely some rethinking to do here for Team Red. Now I doubt Labour's credentials as either "the party of the young", "the party of the left behind" or "the party of the working class" will be boosted by Sly Keir's bizarre decision to stand up for the posh toffs in his own backyard. Because nothing says "I'm a man of the people" as loudly as "step off the grass, oiks", does it?

The second batch of questions is more about the bigger picture, and how voters think the parties fare on a number of generic issues. And these ones deliver some really damning results for Labour. The Conservatives outpoll Labour on all items you can spin positively, most of the time with a significant margin. The worst is that 61% of respondents think Labour have lost their way, including 57% of their own voters. Even SNP voters have a more positive opinion, with "only" 53% thinking Labour have lost it. This is of course linked to the commonly-held belief that Labour lack both a credible leadership and a clear sense of direction. Not that the Conservatives really do better, with respondents equally split about them on both items. But Labour's dire performance still make the Tories look somewhat good by comparison. Which sheds a worrying light on the voting intentions. People are not actually choosing whom appears to be the best at the job, but whom would be the least bad of two unsavoury options. This can actually be a winning situation, as Nicola Sturgeon did win an election while sending the subliminal message that "at least we're not as bad as the other lot". The Conservatives can't demonise Starmer in the same way they demonised Corbyn, so that sort of messaging could work too, with an electorate that appears to be driven more by resignation than enthusiasm.

Now there is one item missing from the poll's list: Labour represent people like me. But you can easily figure it out from the crosstabs on some of the key items. And none of it looks really good for Labour. 29% of Leave voters and 51% of Remain voters agree that Labour represents all of Britain, with 61% and 41% respectively thinking they don't. Not really a resounding success with either camp, and the Leavers' position will be more of a problem in future elections. Just think of Batley and Spen, who voted 60-40 Leave just days after Jo Cox's brutal murder, and how some in Labour are already spinning a by-election defeat there as the end of Starmer's leadership. Just as worryingly, 57% of Labour's 2019 voters think the party have lost their way and 33% think it doesn't have a leadership they can believe in. Also 41% of the 18-34s think the Conservatives are the party of young people and 45% think they are the party of aspiration, while only 53% and 55% respectively think it also applies to Labour. Again not a ringing endorsement from one of the key demographics Labour need to win any future election. Next stop is Batley and Spen, a Labour seat since 1997, that voted 43% Labour, 31% Conservatives and 18% UKIP in 2015. Then 56% Labour, 39% Conservative and 2% UKIP-in-disguise in 2017. And finally 43% Labour, 36% Conservative and 15% combined UKIP-in-disguise and Brexit in 2019. By-election there is on 22 July and it does have the ingredients for a repeat Hartlepool. Though of course the Conservatives have more than enough time for some massive fuckups that would save Labour's arse. Or not.

It's not only that we're not in the room

We don't even know where the room is

(Anonymous Labour insider, Have I Got News For You?, 2021)

People should win things because they’re good

Not because everyone else is more rubbish

(Tom McNair, Being Human: Sticks And Ropes, 2013)

The trends of voting intentions remain extremely negative for Labour, and even more so when you factor in the regional crosstabs in England outside London, and I will come back to that in more detail later. There are even quite a few polls predicting that the Conservatives would do just as well as in 2019, while Labour would again lose a few votes here and there. The trendlines show Labour leaking from both ends, losing a few to the Liberal Democrats and more than a few to the Greens. Both of which are of course beacons of liberal progressivism. Naw, just kidding. And we again have The Guardian indulging in some wishful thinking about the Conservatives standing on feet of clay, which sounds like pissing down the wrong tree in the wrong forest. Of course, the Conservatives' contradictions will some day come back to bite them in the arse. But it happened to New Labour only during Tony Blair's second term, and you could argue that the clock has been reset by the 2019 election, so now the deadline is something like six or seven years out. Besides, what hurt New Labour most was not the public becoming aware of their contradictions, which had already happened during Blair's first term. It was the Iraq War and the bundles of lies peddled by Labour about it, and even then losing votes did not result in losing the next election. It took another full term, and then probably as the result of power-weariness more than anything else. Transpose that onto the current situation, and Johnson might not last as long as Putin, but quite possibly longer than either Thatcher and Blair.

As I hinted earlier, my current Poll'O'Polls shows no massive change from the 2019 results. If anything, we just have the weeest of swings from the Conservatives to Labour. Which in reality is more like the cumulative effect of swings from Labour to the Liberal Democrats, the Greens and the SNP, while the Conservatives have a near zero-sum-game mix of gains and losses, depending on which nation and region you're looking at. What I have here is the aggregate of the last four polls, coincidentally all fielded on 27 and 28 May. The super-sample is 5,714 with a theoretical margin of error of 1.3%. Compared to the 2019 results, the increase of the Greens' vote share is quite close to the fall of the LibDems' vote share. But this definitely does not point to a direct transfer from one to the other. There are more subtle patterns at work here, with voters crisscrossing from one party to the other in pretty much all possible directions. The only surprise is that the instability actually helps the Conservatives, as most of the movement is between parties in the center or the left of the compass. There might even be an even bigger time-bomb for Labour in there. The Conservative vote is definitely an acquired taste for many in former Labour heartlands, which I can't call the Red Wall any more as it is now in no better shape than the Berlin Wall. Anyway, we have seen many instances where acquired tastes stick, and people never look back and regret changing course. Like when you switch from omnivore to vegan. After a while, you no longer miss cheese and eggs, because you don't even remember how they tasted like, or do you?

Unsurprisingly some are using Labour's abysmal performance in recent elections and current polls to promote proportional representation (PR). Unsurprisingly too, this is relayed by The Guardian, who also advocate a "progressive alliance" to muddy the waters just a wee smitch more. At face value, that is bringing back a bigger variant of the Unite To Remain alliance at the 2019 election, which was an utter fucking disaster. Remember that the only success of this hotchpotch alliance was gaining Brecon and Radnorshire at a by-election in August 2019, and then predictably losing it back to the Conservatives at the general election. And, while there was a fuckload of fanfare and gloating about the gain, especially in The Guardian, all the ones involved were remarkably silent about the defeat four months later. And now they can't even make it work again for the next by-election in Chesham and Amersham. It's even a bit pathetic to see three prominent English MPs pleading for it, again in The Guardian, when they know it won't happen and there's nothing they can do about it. Then if you want an update on the mischiefs of PR, just look at what's happening right now in Scotland, where the electoral system is only partially proportional. The Scottish Greens are basically blackmailing the SNP into reneging on some key manifesto pledges, in return for votes on the budget the Scottish Government would have got anyway, and it totally makes my point. PR is not about offering a fair representation of the people's choices, it's about giving hyperactive minorities a free pass at blackmailing the majority into policies their electorate never voted for. Just saying.

Keir Starmer is the perfect husband

You just wouldn't want to have sex with him

(Anonymous Labour insider, Have I Got News For You?, 2021)

We keep seeing videos of people saying “it’s time for a change”

“It’s time for a change, I voted for a change, I voted Tory”

They’ve been in power for eleven years now

It’s not a great change, is it?

(Ian Hislop, Have I Got News For You?, 2021)

You might expect that voting intentions so close to the 2019 result would also deliver a very similar seat projection. And you would be right. At the end of the day, the Conservatives would end up with a net loss of four seats, and Labour with a net gain of one. Which is oddly reminiscent of the pattern seen at the 2001 election, when New Labour cruised to a second landslide, against an unconvincing and ineffective opposition led by a bland leader who inherited his position at the wrong moment. This would lower the Conservatives' working majority from 87 to 79, which are the right numbers when you remember you always have to factor out the Sinn Féin MPs. Of course you also have to factor out the Speaker and the Deputy Speakers, but there are two each for the Conservatives and Labour, so that does not change the majority. This pedanticish technicality being now behind us, the projected results would also widen the rift between Scotland and England, with the SNP ending up close to their 2015 landslide. But that's a story in itself and I'll come back to that later. Things are not looking too good for the Preshoos Yoonion in Wales either, with Plaid Cymru projected to increase their vote share to 15% and gain Ynys Mon from the Conservatives. Plaid Cymru's challenge is obviously to switch and hold that part of the electorate that vote for them at the Senedd elections, but return to Labour at the Commons elections. Though they would also need a more evenly distributed vote, as even 20% at a Commons election would only get them six seats, as they are either very weak or even non-existent in many constituencies.

Based on the current weighted average of voting intentions, we would also have twenty-six Conservative marginals and fourteen Labour marginals that could deliver a slightly different result with only a wee swing in voting intentions. If all the Tory marginals went to either Labour or the SNP, the Conservatives would end up with 335 seats and a 27-seat working majority, oddly similar to the uncomfortable situation John Major or post-Coalition David Cameron found themselves in. But of course the odds are not in favour of that happening. Even if it did, there is little to support a scenario where Labour would then enjoy a succession of by-election gains that would reduce the Conservatives to a minority government, as they managed between 1992 and 1997. And there is also no reason to believe the next First Minister of England, whoever he is, would gamble all on a snap general election that could deliver a 2017ish setback just as plausibly as a 2019ish landslide. The precautionary principle would definitely rule here, even for a party that so miserably failed to abide by it during the pandemic. And if you want some more useless trivia that proves that politics go round in circles. this projection says the Conservatives would bag 157 more seats than Labour. Which is the same number by which Labour outnumbered the Tories at the 2005 election, 355 to 198. Don't tell me it's all just coincidences. Or is it? Just bear in mind that 2005 was considered a lousy result for New Labour, while something similar would be considered quite a success for Bozo. Not every song remains the same.

The future used to be something people got excited about

That tells you everything you need to know about the world

(Jon Richardson, Ultimate Worrier: The Future, 2019)

The song Gab famously never played at the 2009 Oscars ceremony

They asked him to do a one-minute digest and he told them to fuck off

There’s times I think maybe we should just blow up the world

Like in Star Wars when Princess Leia sees her home planet destroyed

She’s a bit sad but part of her is also like

“Yeah, there’s a lot of arseholes there”, you know?

(David O’Doherty, 8 Out Of 10 Cats Does Countdown, 2018)

As usual, it's interesting to see how the UK-wide trends translate in England, as they have the most seats and usually the largest variations from one region to the other. No seat is predicted to change hands in London, as all three main parties there would lose votes in very similar proportions, and that would mostly benefit the Greens, who are in no position to gain a seat or even influence the result in any there. So the main events will happen in the other three English meta-regions, and not necessarily in directions you would expect, or were already there when I wrote my last general election article seven weeks ago. The Reform UK vote remains high in the North, which might just be a polling bias, from the polls' implicit assumption that all parties would stand in all constituencies. Polls are also notoriously unable to predict local variations such as tactical voting, so there might be few more Hartlepools waiting to happen. Regional crosstabs also show that the three meta-regions are not as monolithic as you might think. There is evidence, for example, that Yorkshire and the Humber may be in the same league as the Midlands, offering Labour some opportunities for second chances. While the rest of the North looks more likely to swing even further towards the Conservatives. The South West is also more likely to grant the Greens a high vote share, plausibly up to double digits. Which is probably the cumulative effect of dissatisfaction with Labour, and opposition to the Conservatives' project to reform planning laws. Protest from within the Tory ranks had it put on the backburner some months ago, but it's back now and might lead to further rebellion down the road.

The regional variations here also support the common-sense theory that there is a socially conservative nationalist working class vote that will go to the dominant populist right-wing party, when they feel the traditional left has let them down with their combination of economic globalism and social liberalism. Same thing that happened in the United States with the Rust Belt voting for Trump, or in France with the post-industrial North shifting to the National Rally. In the UK, UKIP and then the Brexit Party were the natural outlets for this electorate, until they were past their shelf date, and it was then far easier to shift to the anti-European and pro-British-values Conservatives that to go all the way back to the woke metropolitan Labour of the post-Brexit era. The effects on the seat projections in each meta-region are quite interesting, as they show Labour still face an uphill battle in most of the North, and might find it harder than expected to reap the fruits of demographic changes in the South. Oddly, when you compare this to what we had seven weeks ago, their best case scenario is now in the Midlands, where the surge in the Green vote is least likely to impact them negatively. But obviously the big picture again says Labour still have a long way to full recovery, and probably some of the proverbial tough choices will have to be made in the future. It's just not totally clear what they should be, other than pointing out that pursuing conflicting goals to accommodate different parts of the electorate is not part of the solution. Uniting tribes with conflicting expectations might have worked for New Labour for about ten years, but it's definitely outwith New New Labour's reach for now.

On current polling, Labour would gain back four seats in the Midlands (Derby North, Gedling, High Peak, Stoke-on-Trent Central) and three in the North (Dewsbury, Keighley, Pudsey, not coincidentally all in Yorkshire), but lose a further four to the Conservatives in the North (Lancaster and Fleetwood, Oldham East and Saddleworth, Warrington North, Weaver Vale), where former LibDem leader Tim Farron would also be unseated by the Tories in Westmorland and Lonsdale. We also have almost a status quo in the Leafy South, which is not really good news for Labour, compared to earlier projections where they were back to 30+ seats doon there. Their new problem is that, as projected vote shares show, the "new young suburbites" are more likely to vote for the Greens than for Keir Starmer, and in large enough numbers to jeopardise a Labour surge in the leafy commuter belt. So Labour would now gain back Peterborough from the Conservatives while losing Bristol West to the Greens. Labour clearly suffers some "non-gains" because of a fragmented opposition, but it's difficult to see which magic trick could solve this. The Westminster System is clearly not designed to accommodate pre-election coalitions, which would be easy targets for the governing party as "coalitions of chaos". Thusly, Labour's only road back to power is to attract a sizeable part of "soft Tory" voters, which seemed to work early in Keir Starmer's tenure, but has now been totally ineffective for several months. This should raise concerns at Labour HQ about their inability to capitalise on Johnson's Cabinet's unprecedented and world-beating list of failures, law-breaking and lies. Even more so when the last Opinium poll says that more people trust Johnson and Hancock to tell the truth, than trust Cummings' revelations. And fewer trust the Leader of the Opposition than the Government. Ouch.

Dominic Cummings really is the kind of Prince of Dark Arts, isn’t he?

He’s our very own, or was, when he was at the heart of government

Our very own Littlefinger, the kind of Bastard of Barnard Castle

He really knew how to make sure he had all the power in his hands

And I just think that everybody’s desperate to have him back

To tell us exactly, you know, what he wasn’t doing

When he should have been doing his job

(Baroness Sayeeda Warsi, Have I Got News For You?, 2021)

To be fair, the whole of Scotland was watching Cummings going

"Keep going, keep going, Dominic, dig a bigger hole"

"Give Dominic a shovel, get tore in, fella", because the whole of Scotland

Was like "This is the reason we don't want anything to do with these people"

(Janey Godley, Have I Got News For You?, 2021)

Finally, let's have a look at what the polls have to say about Scotland's voting intentions at the next Westminster election. In which Scotland will definitely take part, as the odds of Boris Johnson saying Yes to a Section 30 Order are about as high as those of Boris Johnson saying Yes to a Section 30 Order, or somewhere South of the square root of the SFA. This could have been avoided if the Scottish Government had not instructed the Lord Advocate James Wolffe to sabotage Martin Keatings' People's Action on Section 30. It's quite obvious that the SNP are sending a clear message here: oiks, don't meddle with this as it is our unalienable mandate to totally fuck it up again. Martin's full account and comments can be read here, and I encourage you to read it in full, so you're fully aware of all the gory details of the SNP's deceit and obfuscation. There will be a price to pay for this some day, most likely in the shape of the SNP losing their bid for a fifth term in 2026 if Independence has not been secured by then. But let's take this one baby step at a time, as the SNP do with Independence, and look at first at the trends of Scottish voting intentions. I factored in only the Full Scottish polls and not the subsamples of GB-wide polls, with one exception I will come back to later. The trends speak for themselves, with the SNP continuously predicted to bag twice as many votes as the Conservatives, and Labour in disarray. The most recent polls also credit the Greens with a higher vote share than in 2019, aboot 3 to 4%. This works only in theory, if the Greens field candidates in all constituencies, which they probably won't. The interesting part is that, like in England, this Greenish surge has dented Labour's vote share, but so far has not hurt the SNP.

Based on the most recent survey of Scottish voting intentions, the next general election would be a massive success for the SNP, though not as overwhelming as the 2015 landslide. Then I guess nobody is actually expecting an exact repeat of that one. This poll is the exception to my rule, as it is the Scottish subsample of a GB-wide poll. But this was the massive poll conducted by Find Out Now for the Sunday Telegraph just after the various May elections. Their sample was 14,715 with a Scottish subsample of 1,280. Which is some 25% larger than the usual Full Scottish poll, so definitely qualifies as one to keep. Besides, there was nothing in this poll that contradicted the general trends, or what we could infer from the Scottish Parliament election results. Here we again see the Conservatives doing worse under Doogie Ross than under Jackson Carlaw, losing 3% of the popular vote and two seats. Doogie's own seat in Moray would return to the SNP and the Conservatives would hold only the Leave-voting Banff and Buchan, and the three Borders-and-Border seats. Of those, Alister Jack's seat in Dumfries and Galloway would be held only by the weeest of margins, and would be the next one to fall on a good day for the SNP. At face value, the Liberal Democrats would hold only Orkney and Shetland, but with a similar margin to the 2015 election, when the late Danus Skene nearly sent Alistair Carmichael packing. So fair winds would also shift this one to the SNP quite easily.

There is of course one big flaw in the Scottish polls we have: they don't include the Alba Party. Though it would be challenging to factor in Alba in any projection, as we're definitely treading into terra incognita here. You can definitely expect the SNP to go all the way to "if I can't have it, then neither can you" in both Alba constituencies, and field candidates against the defectors, Neale Hanvey and Kenny MacAskill. Which of course would be a dick move as it would gift Labour with two easy gains on a silver platter. And is exactly what you can expect from New SNP, as it would allow them to again whine about "transphobes splitting the vote", and blame Alex Salmond for Culloden and the Clearances. I don't expect the Alba Party to respond in kind and go tit-for-tat. Though it would be quite enlightening to see how luminaries like Kirsty Blackman, Alyn Smith or John Nicolson would fare if facing a direct challenge. But I don't think we will ever see that as I'm convinced the Alba Party will act responsibly, if New SNP don't, and will take one for the cause. Like asking Neale and Kenny to stand down, and clear the way for SNP "holds", so the Unionists and their affiliated media won't get any oven-ready soundbites about devastating blows and downward spirals. Then the next general election is three years out, which is both an eternity and the blink of an eye in politics, and I can't rule out the Ultimate Game Changer happening in the meantime. Like Nicola Sturgeon pulling a Blair on us: anointing Kate Forbes as her heir apparent, and quitting mid-term. Which could be a major step towards the reconciliation of the clans. One can dream, can't he?

With my teenage self, what advice would I give?

I would say, introduce more fibre to your diet

And don’t strain too hard during a bowel movement

If alcohol and drugs don’t get you, haemorrhoids will

(Kevin Bridges, 8 Out Of 10 Cats Does Countdown, 2020)

18/05/2021

At The End Of The Counts

What the Americans call moonshine, the Irish call poteen

And the Scottish call hand sanitizer

(Jimmy Carr, 8 Out Of 10 Cats Does Countdown, 2017)

About 13% of English words are not spelt the way they sound

And that percentage increases to 100 in Glasgow

(Jimmy Carr, 8 Out Of 10 Cats Does Countdown, 2017)

So now it's done. All the new MSPs have been sworn in and the Scottish Parliament is open for business again. You remember I had two big hopes for this Parliament. First, that the SNP would get an outright majority on the constituencies, so they could tell the Greenish Inquisition tae git tae fuck, and they missed it, though I guess there is little they could have done. Second, that the Alba Party would make enough of an impression to get one MSP in each region, and that one was quite wildly off the mark, wasn't it? And of course, I really wanted Joan McAlpine holding her seat, and that was a big miss too. I'm in absolutely no mood to blame anyone or sling mud to the four winds, unlike some SNP zealots who tweeted some truly despicable stuff and showed everyone what a sore winner looks like. As Bob Dylan once wrote, "what's lost is lost, we can't regain what went down in the flood", so let's just move on. But first let's move back a bit and see how the well the last batch of polls before Election Day did, or not, and how it compares with the actual result. Here there are some odd similarities with the last poll before the 2016 election. Back then and again this year, they overestimated the SNP on the constituency vote, and by almost the same amount. And, again like in 2016, this year's polls underestimated the SNP and the Conservatives on the list vote, and overestimated the Greens. Troubling.

So the end result was also off by a couple of seats, just as it had been in 2016. Interestingly, if you feed the actual results into an uniform swing predictor, it will deliver 63 SNP seats, 60 constituencies and 3 list seats. Just saying that uniform swing can be quite crap too. Interestingly, uniform swing would have given the SNP Dumbarton, which they missed. But wouldn't have given them Edinburgh Central, Ayr and East Lothian, which they gained. Now it's interesting to check which pollster was the closest to the actual results. Panelbase's last poll, published a week before the election, definitely wins on the constituency vote as they had the SNP on 48% and 27% ahead of the Conservatives. That same poll also wins on the list vote, as they had the SNP on 39% and 17% ahead of the Conservatives. Panelbase was also one of only two pollsters who got the Green list vote right on 8%, Opinium being the other one, and were also closer with the Conservative and LibDem list votes, on 23% and 6% respectively. So I guess it's a tie between Panelbase and Opinium here, rather than an outright win for Panelbase. YouGov, who were the best in 2016, definitely fucked it up this year, wildly overestimating the SNP constituency vote and the Green list vote. And the worst by far this year was Savanta Comres, missing the SNP vote by 6% on both the constituencies and the lists in their last poll for The Scotsman, that predicted the SNP down to 59 seats overall. Ouch. I think it's fair to say that the identity of the client introduced a bias here that goes well beyond the usual sampling variations or house effects. But I can't even challenge them on this as they blocked me on Twitter for no reason but questioning their compliance with British Polling Council disclosure rules. Weird election indeed, where one of the pollsters turns out to be the sorest loser.

Some constituencies did not fit any pre-determined pattern. Once such case is Rutherglen, which appeared as a weak SNP hold with Labour breathing down their neck, but saw the SNP increasing their majority. The only explanation that works is a combination of Unionist tactical voting and a shift of "soft Yes" voters from Labour to the SNP. In Rutherglen, the latter vastly outnumbered the former and proved there was no Margaret Ferrier Factor punishing the SNP, no matter how often BBC One Scotland brought it up. Then the SNP had a close shave with a total trainwreck in Banffshire and Buchan Coast, just as I hinted in my last pre-election article. Losing 10% of the vote and prevailing by only 772 votes looks awful in what was once Alex Salmond's seat, where the SNP bagged a majority of the popular vote at all previous Holyrood elections. But, as I said last week, the SNP would definitely have asked for it if they had lost the seat. We also have the Greenies arguing they had been "robbed" list seats by some obscure far-right group, without offering any solid evidence. As always twisting the basic math of AMS, and hoping their sequacious cultists will swallow it. Sore winners turning into sore whiners is just what we need right now, innit?

Nicola Sturgeon is the hardest politician

You wouldn’t want to get into a ring with Nicola Sturgeon

(John Pienaar, Have I Got News For You?, 2021)

There are five million CCTV cameras in Britain today

That’s one for every person in Scotland

(Jon Richardson, Ultimate Worrier, 2018)

There are many lessons to be learned from this election. First one is how quick the metropolitan enlightened English press, that is The Guardian, were to help the SNP in their savaging of the Alba Party and personal smear against Alex Salmond. The main reason for this being of course that they support the same kind of social liberal policies, most prominently with their unquestioning promotion of identity politics, which have replaced Independence at the top of the SNP's manifesto for now. There is no doubt that the Unionist parties and media quickly realised that the Alba Party was more of a threat to them than to the SNP, and behaving as the SNP's little helpers was the surest way to nip it in the bud. The enemy of my enemy is my friend and they all lived happily ever after. The Unionist parties also used the full potential of anti-SNP tactical voting. This is made quite painfully obvious by the long list of seats they held against all odds (Dumbarton, Edinburgh Southern, West Aberdeenshire, Dumfriesshire), thusly depriving the SNP of an outright majority on the constituencies, which was always the one recipe for the truly unquestionable victory they needed, instead of the quasi-status quo we had. The chart below also shows how much tactical voting there was even in seats that appeared out of the reach of the SNP even on very good day (Eastwood, North East Fife, Edinburgh Western). The results there obviously do not reflect the true strength of the winning party in such constituencies, but show that the Unionists have mastered "better safe than sorry" much better than the SNP.

Here I included all seats held by Unionists before the election on one side, then only those they held at this election on the other side, as tactical voting was quite obviously more prominent in these. As the Liberal Democrats held all of their four constituencies, I set aside the two mainland seats, as Orkney and Shetland have a life of their own, where general patterns found elsewhere don't apply. Interestingly there was also pro-Independence tactical voting in Edinburgh Central, the key marginal that a Green vanity candidacy handed to the Conservatives on a sliver platter in 2016. There the SNP benefited from a combination of the Green vote going down, and a fair share of Labour voters switching to the SNP. Other Green vanity candidacies in other seats did not significantly impact the SNP's results. In some cases like Glasgow Pollok, the SNP's vote share fell slightly, but Labour voters switching to the SNP almost compensated for votes lost to the Greens. In the most controversial case, Galloway and West Dumfries, the Green vote proved irrelevant in the end, after fears it might cost the SNP a possible gain. The SNP would have missed it anyway due to the extraordinary amount of Unionist tactical voting. There were also some hints of Unionist tactical voting for the Tories on the regional lists. But it's far less convincing as we know a fair share of Labour constituency voters switched to the Greens on the lists, so the numbers actually don't make the case one way or the other. I can't think of any credible situation where tactical list voting clearly gained the Tories a seat from the SNP or the Greens. The only example I have is actually one where it backfired, with Labour losing a seat to the Conservatives in Highlands and Islands, and only due to the high proportion of LibDem voters who switched to the Tories on the list.

In Edinburgh, they have salt and sauce

In Glasgow, we have salt and vinegar because we’re classy

(Susan Calman, QI: Revolutions, 2020)

Instead of scuppering the Scottish nationalists

A Scottish Parliament has made them more popular

Now, that might only be a blip but, with her own politics, her own news,

Today’s Scotland feels a lot more than 400 miles away from Westminster

(Andrew Marr, History Of Modern Britain, 2007)

Now the final test is how my last projection compares with the actual results, on a seat by seat basis across all regions. It's also an opportunity to assess how my model fared, in comparison to the more widely used method of uniform national swing. Uniform national swing on both votes, based on the same batch of the last five polls, would have delivered 65 SNP seats, just like my model. But these would have all been constituencies, with no list seats on top, while my model was a wee smitch closer to the actual result with 64 constituencies and 1 list seat. Relying on the regional crosstabs of the polls avoided me two misses in West Aberdeenshire and Dumbarton, which uniform swing allocated to the SNP. But uniform swing allocated Dumfriesshire to the Conservatives, while I had it switching to the SNP. And both missed the Labour hold in Edinburgh Southern, as tactical voting is obviously something you can't readily incorporate in any projection model, as it would require some seat-by-seat tweaking of the projection, which defies the fundamentally statistical foundation of the model itself.

Overall I have scored 2 misses and 71 hits on the constituencies, a 97% hit rate. I will certainly not fault either my model or the polls for that, as I have already explained in some detail why I missed Dumfriesshire and Edinburgh Southern, and I'm quite sure everybody would have. Then I have 7 misses and 49 hits on the list seats, a 88% hit rate. This is based solely of the number of seats, and not the order in which they were allocated, which is pretty much irrelevant, given the number of factors at play here. I will not fault my model here, as the formula to allocate seats under the Scottish variant of d'Hondt is straightforward and basic maths. There is no way you can get that one wrong, unless you are politically motivated to make it look more complicated than it is and misrepresent it. The misses here are generally the result of polling errors, as polls regularly underestimated the Conservatives and overestimated the Greens on the lists. Polls also underestimating the SNP's list vote is far less of a factor here, as they were bound to not bag any list seats in six regions anyway. Missing a third of the list seats in South Scotland, my worst result, is also a ripple effect of missing one of the constituencies there. So I consider my overall result of 9 misses and 120 hits, or a 93% hit rate, quite satisfactory. After all, some people have won the fucking plaque on Four In A Bed with lower scores than that, haven't they?

Now that I'm done patting myself on the back, there is one last and important message from this election's results. When you have all the results in, even a 10-yr old can figure out the average and the standard deviation of every party's vote share, can't they? Or call Rachel Riley to the rescue, she might need a wee fee to cover all her legal bills. The standard deviation doesn't say much by itself, as a higher average vote share will almost automatically generate a high standard deviation. But the ratio of standard deviation / average does matter. The lower it is, the more evenly spread your vote is across all constituencies. Here the SNP is a clear winner with a ratio of 0.17, compared with 0.51 for Labour, 0.54 for the Conservatives and 1.49 for the Liberal Democrats. This might sound fine at face value, but it begins with a blessing and it ends with a curse. Because that's the textbook situation where a party with this sort of results is relatively immune to small swings, but can be fatally hurt by larger swings. Of course any swing against the SNP would have to be huge to make them lose the next election, but we've seen this before, haven't we? Labour, 2015, 24% swing. Never say never again. Just saying.

There’s nothing noble about being on the side that loses

You find the biggest kid in the playground and you stand next to him

(Werewolf Milo, Being Human: The War Child, 2012)

I know a thing or two about prophecies, they’re bullshit and mind games

Seems to me that they only get dangerous when you actually start believing in them

(Annie Sawyer, Being Human: The Graveyard Shift, 2012)

Now the last page in this chapter was the Airdrie and Shotts by-election, for Neil Gray's former Commons seat. The actual results shows that both my hunch and my calculations were way off on this one. And this is one of a few cases where I still don't know if I'm fully gutted or just slightly miffed to have been proved wrong. My last projection was a close call, with Labour gaining back the seat by 1% on a high level of tactical voting from the Conservatives and LibDems. But also based on the 6 May results, when Labour polled 12% higher than their national average in the overlapping Holyrood constituency, and the SNP only 3% higher. So the combination of all factors made a Labour gain quite likely. This was also supported by expectations of a very low turnout, which statistically hurts the SNP more than the Unionist parties. Then everything went a different way to what earlier trends and recent Commons polling in Scotland indicated. Turnout fell from 40k to 22k but the SNP slightly increased their vote share. The impact of tactical voting was a 6.9% swing to Labour, twice that at the Holyrood election a week earlier, but still massively below what would have unseated the SNP. Then the most extraordinary thing here is that the SNP were able to keep their troops mobilised for the by-election, after what was undoubtedly a strenuous Holyrood campaign. Guess they also heard the warnings about Labour withdrawing resources from Hartlepool when they realised it was already lost, and throwing the Full Monty and his python at Airdrie and Shotts. The message got through that nothing should ever be taken for granted, and it worked.

Despite this convincingly brilliant victory in Airdrie and Shotts, the SNP can't escape the simple fact that Labour still have quite a voter base all over the Auld Strathclyde area, even before tactical voting is factored in. On 6 May, Labour overperformed by 9% in Central Scotland and Glasgow, and by 6% in West Scotland on the constituencies. Which I think is a quite fair assessment of their strength in these regions, as tactical voting went in totally different directions depending on the constituency, and looking at the bigger picture across the whole region irons out the differences. And, even after losing part of their votes to the Greens, they still overperformed by 6% in Central and Glasgow, and 4% in West on the lists. The only other region were they overperformed on both votes was Lothian, and by smaller margins. There are in fact enough regional variations all over Scotland to make the case that uniform national swing is probably no longer the safest way to predict the results of future elections. I also believe that some patterns seen at this Holyrood election could easily be duplicated at a Commons election, despite the different nature of the two votes. Unionist tactical voting is just one of them. You also have to account for the ancestral Scottish tribalism and ferality, which are probably more of a factor on the Yes side, if recent events are any indication. As usual, only time will tell. Or some unforeseen by-election.

Now eagerly waiting for the first poll about the 2026 election, just to check the level of buyer's remorse from SNP voters who switched to the Greens on the list, instead of Alba.

No constitutional chopper has come down between Scotland and England

But it feels as if a slow, gentle separation is taking place

Like two pieces of pizza pulled apart, still connected

But only by strings of molten cheese

(Andrew Marr, History Of Modern Britain, 2007)

06/05/2021

What's New On Election Day?

You can propose as many alternatives as you like in your referendum

We will propose only one option in ours

We deserve a vote too on whether we want to keep subsidising the Scots

Who are always complaining about being occupied

(Jim Hacker, Yes, Prime Minister: Scot Free, 2013)

The rest of the UK doesn’t see why the students in Scotland

Should get free university education but English students don’t

Even if they go to a Scottish university which we are paying for anyway

They don’t see why they should pay prescription charges but the Scots don’t

I have no doubt that the UK will vote for full independence now

(Jim Hacker, Yes, Prime Minister: Scot Free, 2013)

Election Day it is, and in-person voting is already underway as I put the finishing touches to this. The upside of delayed counts is that you can still post a late rant and projection on Election Day itself. The other upside of course being that you don't have to stage an all-nighter with the dug asking for extra walkies out of the usual schedule. The downside is that we won't know for sure what actually happened until probably lunchtime on Sunday when all the numbers are crunched and the final list results announced. It does not mean we can't have high hopes, but just keep in mind that, in 2016, even the very last poll was somewhat off, overestimating the SNP vote by 1.5% on the constituencies and underestimating it by 0.7% on the regional lists, and underestimating the Conservatives by 3% on both votes. Which looks acceptable at face value as it was within the standard margin of error, and it was much closer to the actual result than any earlier poll. But, if it had been duplicated on Election Day, the SNP would have bagged 67 seats and a 6-seat majority instead of 63 and 2 seats short, including 65 constituencies, and the failed "Both Votes SNP" would have been the least of Nicola Sturgeon's worries. As always, the devil is in the details, and it would be really ironic if we were faced with the same situation this year. Then I will have my fingers crossed, hoping this photo from The National was premonitionary, and we'll get a super-majority or close to it, with the Alba Party doing better than the Greens.

Actually, we could even tweet updates until lunchtime tomorrow as the first counts will start at 12:00, but I definitely won't do that. I will be just preparing for BBC One's coverage, which will start at noon sharp on Friday, when there won't be anything yet to announce. But of course we must allow the pundits some time to draw definite conclusions from what hasn't happened yet, mustn't we? They will have a full hour to do that, before News At One rubs Labour's nose in it with full gory details of how they lost the Hartlepool by-election. Then we'll go Full Scottish again, but only until 8pm as we wouldn't want to cancel EastEnders for just some more election coverage, would we? But of course we will have seven more hours on Saturday, from lunchtime to the Celebrity Mastermind slot. I wonder if the election officials all over Scotland have taken their cues from the Line Of Duty writers, as tomorrow will definitely end on a cliffhanger with no region having returned all constituencies. Oddly, West Scotland will have only one count on Saturday (Renfrewshire South) while the other regions will be split 5-4 or 4-5 between the two days. I'm just slightly gutted that Edinburgh Southern will count only on Saturday, so we will have to wait to celebrate Catriona McDonald ousting Ian Murray's sock-muppet Daniel Johnson. But I'm quite chuffed that election officials in Highlands and Islands have decided to play it safe, with all three island constituencies declared tomorrow. I just hope they kept the ferries at the ready anyway, so we'll get the results no matter the weather. Let's just hope the Election Days finale won't be a letdown as Line Of Duty's was, which was really a pity after Martin Compston had so brilliantly backed the SNP and Independence, and telt Conservatives tae git tae fuck.

Scarborough is very picture-postcard, much prettier than Dundee, I think

(Johnny Vegas, 8 Out Of 10 Cats Does Countdown, 2016)

Did you know that, in Wales, sheep outnumber people three to one?

So there is not much to do except…. well…. things with sheep

(Jimmy Carr, 8 Out Of 10 Cats Does Countdown, 2015)

Before we deal with all the Scottish stuff, let's take a detour through Wales, where they have an election too. For the Senedd Cymru, formerly known as the National Assembly for Wales. The Senedd is elected on the same Additional Members System as the Scottish Parliament, with 40 Members of the Senedd (MS) elected in the constituencies and 20 on regional lists from five regions. The Welsh election has been polled far less than the Scottish election, and not generated the same level of interest in England. They have had only nine polls this year, while Scotland has had thirty-five, and two "last day" polls from Savanta Comres and YouGov, compared to our five from five different firms. Below you have the voting intentions based on the weighted average of these two polls, and the seat projection, factoring in the regional crosstabs of the polls like my Scottish projections. This is not all milk and honey for Welsh Labour, as their lead over the Conservatives has been cut by nearly half since the 2016 election. Five years ago, UKIP was the fourth party in Wales and then it imploded. Their MSs scattered to the four winds, some becoming independents, some joining the Brexit Party and then Reform UK, with a few founding the new nutjob party Abolish The Welsh Assembly (AWAP). But most of their voters switched to the Conservatives, a pattern Keir Starmer has seen elsewhere, hasn't he? Propel, formerly known as the Welsh National Party, are dissdents from Plaid Cymru. I guess the original name, which was rejected by the Electoral Commission, gives you a hint about where they get their inspiration from.. Though their poor performance in the polls makes them more like the Cardiff Variant of the Alba Party.

Despite some rollercoaster polling over the last five years, which at times pointed to a Conservative takeover of the Senedd, the last-day polling does not predict an earthquake in Cardiff. UKIP's seven seats will quite logically go to AWAP and the Conservatives. Labour are predicted to lose four constituencies to the Conservatives (Vale of Clwyd and Wrexham in the North, Vale of Glamorgan and Gower in the South) but will compensate in part with list seats, ending just two seats down. Plaid Cymru are predicted to gain one seat, and the Greens will enter Senedd for the first time with a list seat in the South Central region. Remember they are not the Welsh Greens, but part of the Green Party of England and Wales, the same as Caroline Lucas. And will gain their seat in a region that is mostly Cardiff and Cardiff's commuter belt. Guess they attract the same kind of metrocosmopolitan middle-class enlightened rainbow-socked youth as the ones we have North Of The Wall, or those they have Doon Sooth in the outskirts of Hampstead Heath. With such a make-up of the Senedd, it will be interesting to see how the coalition negotiations will go. Labour started the last term as a minority government in a coalition with the lone Liberal Democrat MS. Then First Minister Mark Drakeford beefed it up to a two-seat majority when a dissident from Plaid Cymru joined his government. Over the years, Welsh Labour have tried all the available options (minority government, coalition with the Liberal Democrats, coalition with Plaid Cymru), and are more open than Scottish Labour to exploring all possible options. So, if the election goes the way that the last-day polls indicate, my tenner is on a Lab-Plaid coalition for the next term.

Game Of Thrones is set in Westeros, a lawless country full of poverty,

Violence and dragons, like an upmarket Wales

(Richard Osman, 8 Out Of 10 Cats, 2019)

I recently bought a riverboat in Glasgow

Or, as we call it in England, a shopping trolley

(Jimmy Carr, 8 Out Of 10 Cats Does Countdown, 2014)

Pollsters have been quite active surveying the Scottish Parliament election and the hypothetical Second Independence Referendum, but also devoted some effort to the Commons voting intentions in Scotland. Which have not yet ceased to be relevant as I don't see any plausible scenario in which Scotland would be independent before 2 May 2024, which is by law the date for the next general election. Though of course the First Minister of England may choose to ignore it. Either by repealing the Fixed-Term Parliaments Act, or by circumventing it in the same manner as in 2019. The same stunt, that was suggested by the Liberal Democrats and the SNP then, would work again if Boris Johnson saw a way to use it to his advantage. Which could definitely happen if UK-wide polls continue to show the Conservatives in the lead and Labour bound again for a disastrous crash-landing. The last snapshot of Scotland comes from Opinium, among dozens of items they surveyed between 28 April and 3 May for Sky News. It's fairly good for the SNP, though not their best this year. The Tories getting the same vote share as in December 2019 means there is technically a swing from them to the SNP, smallish but enough to switch their most vulnerable seats. But of course the SNP should be more worried about Labour increasing their voting intentions. The regional crosstabs also show Labour doing better than their national average in Glasgow, Central Scotland and West Scotland. So Labour is still alive and kicking in the heartlands of the Auld Strathclyde, which could lead to some nasty surprises for the SNP.

What should worry the SNP is that a number of reputedly safe seats have turned into marginals. Among them is Airdrie and Shotts, and the SNP are definitely very unlucky to have a by-election there a week away. I won't bore you again with how they have really brought this on themselves, but I will not refrain from rubbing their nose in it if they lose it. Or not. You might remember that I rated this one as a 40-40 tie between the SNP and Labour when it was announced that it would not take place on the same day as the Scottish Parliament election. Polls since have confirmed this, as the one I had last week had it going 41-40 to Labour, and the most recent one is a 40-40ish tie. No cheating here, as the numbers purely reflect the national voting intentions, and how the regional crosstabs show Labour overperforming in Central Scotland. The most recent polls have the SNP's lead over Labour there on 10-15%, way below the national average of 25-28%, and of course projections for individual seats reflect just that. Especially in a seat with this profile, that Labour nearly gained back in 2017 with the Corbyn Surge. On top of this, Labour certainly improved their odds when they selected Kenneth Stevenson, Councillor for the Fortissat ward of the North Lanarkshire Council, instead of Pamela Nash, a former MP for the constituency and now CEO of the ultra-Unionist lobbyist organisation Scotland In Union. This might reflect the personal views of Labour members in the constituency, but is also a clever choice in an area that voted 51% Yes in 2014. Even the maths agree with that, with the most recent projection definitely making this a tougher than expected by-election for the SNP.

Oddly, the most disturbing news for the SNP have come from Northern England. A new Survation poll found Labour losing the Hartlepool b-election by 17%, with Keir Starmer also receiving dismal favourability ratings. This came on top of other polls showing Labour on their way to defeat in a couple of mayoral elections, also in Northern England. Keir Starmer tried to put on a brave face and play down too high hopes for Labour at the English local elections, but reliable sources say they basically have conceded defeat already and have withdrawn resources from the North. And have now unleashed the Red Stormtroopers on an unsuspecting Airdrie and Shotts. Gaining back a Commons Scottish seat is just what Sly Keir needs at the moment, when even a huge swath of his 2019 voters turned their back on him in Hartlepool. What actually happens in Airdrie and Shotts will hint at what could happen in some of the new marginal seats. Think Glasgow North East, Rutherglen and Hamilton West for those most obviously in the danger zone. Then Labour might just bide their time and wait until the next general election and the SNP fielding candidates against the Alba Party incumbents in East Lothian and Kirkcaldy and Cowdenbeath, which would switch both back to Labour with no specially strong effort involved. Just saying.

My plan shows that Scotland is an equal partner in the United Kingdom

And it’s full of marginal seats…. I mean depressed areas

(Jim Hacker, Yes Minister: The Official Visit, 1980)

Scotland is not at all that remote

It’s that pink bit about two feet above Potters Bar

(Jim Hacker, Yes Minister: The Official Visit, 1980)

As you surely know by now, the Magic Number today is 65. The number of seats the SNP need for an outright majority and no need for a government coalition, though the decision on this will be purely political and not dictated by the numbers. Then I guess the bookies' favourite this week was whether of not the SNP could reach the Magic Number on the constituencies only. Which is definitely a "mibbes aye, mibbes naw" situation, as polls have been going in all possible directions. Seat projections from the most recent polls say the SNP could bag anywhere from 59 to 68 constituencies. Odds are they will not lose any of those they bagged in 2016, save for special local circumstances. Yet there is also a plausible scenario where "no change" could actually be more of a zero-sum game, with the SNP losing a couple to the Conservatives while gaining a couple from Labour. At face value, the trend of the constituency voting intentions does not predict a doomsday scenario for the SNP as we have a whole batch of polls pointing to a vote share around 50%. If Labour end up on Election Day at the high end of their polling, the SNP might find themselves in tighter than expected races in a few seats like Glasgow Pollok, Glasgow Provan or Rutherglen. At the other end, a strong showing by the Conservatives is unlikely to endanger many SNP seats. except possibly Perthshire South and Kinross-shire. But its main effect would be to jeopardize possible SNP gains, especially in the South Scotland region, which might unexpectedly hold the key to the whole election. Three Conservative seats (Ayr, Dumfriesshire, Galloway and West Dumfries) and one Labour seat (East Lothian) there are clearly on the SNP's top targets list. Gaining all four would put the SNP three-quarters of the way to a majority on the constituencies only, and there are credible hints the SNP's lead over the Conservatives in the region might increase from a measly 8% in 2016, one-third the national average, to double digits. Which would still make it something of a Tory heartland, but would put their vulnerable seats within reach.

Some of key battlegrounds will shed light on one of the SNP's major contradictions in this campaign. Their sheepish acceptance of Green vanity candidacies in twelve constituencies. Will the Little Green Men hand Edinburgh Central to the Conservatives on a silver platter again? Will they jeopardise a possible SNP gain in Galloway and West Dumfries? Will they make the long wait for the results more stressful for the SNP in Edinburgh Northern and Leith, or in Paisley? Then don't expect any retaliation or even significant whining from the SNP if some trainwreck happens. They need cuddling the Greens for an agreement on a government coalition, don't they? Of course, there is another Magic Number out there. 86, the number of seats needed for a super-majority. Though bagging it, on a combination of the SNP, the Greens and the Alba Party, would mostly have symbolic value. Legally, it is needed only to change the electoral law, something no party has seriously proposed so far, and would be highly controversial anyway. Hoping that a super-majority could force Boris Johnson's hand, and make him agree to a Section 30 Order, looks delusional to me, though other political arguments for it do have weight. Then of course the likelihood of such an outcome will be determined by the list votes, and that's another and more complicated story. And this might be the point where the SNP regret having relentlessly savaged the Alba Party all along the campaign. Just saying.

The English don’t care about Scotland one way or the other

When you gave us devolution, you gave us certain powers

You don’t like our choices, so now you’re having a wee tantrum

Threatening to go home and take your ball with you

(Deputy Prime Minister Rory McAlister, Yes, Prime Minister: Scot Free, 2013)

Did you know, for example, hurkle-durkle is an old Scottish term

Meaning "to lounge in bed all day"

Hey, Scottish people, you can't lie in bed all day unless, of course

You've given the Domino delivery guy his own key

(Jimmy Carr, 8 Out Of 10 Cats Does Countdown, 2017)

The trend of voting intentions for the regional lists is definitely troubling for the SNP. Instead of the small surge seen in the constituency voting intentions, here we have an uninterrupted descent into the mid-to-high 30s. The very last poll from YouGov has 19% of the SNP's constituency voters shifting to the Greens and 4% to the Alba Party on the lists. The same poll says that two-thirds of the Greens' list vote comes from SNP voters on the constituencies, the other third being mostly Labour voters with a tiny group of LibDems thrown in for good measure. But we all expect that Alex Salmond and Alba will be blamed for any lousy results here, don't we? Even more so if the Alba Party overperform and reach a higher vote share than the 2-3% predicted by the last batch of polls. If they do, expect Big Eck to be once more savaged for "endangering independence", even it the math proves conclusively that Alba nicked a handful of seats from the Unionist parties and not from the SNP. And of course I would not object if Alba happened to nick a couple of list seats from the Greens too. Then the Tories might also face a challenge from George Galloway's All For Unity, who have basically turned into the British National Party in Scotland. There's actually a high probability they would nick enough votes off the Tories to secure a seat for Catman himself, in the Borders or on the Border, once he has sorted out which is which.

There still are a couple of trick questions about the list votes. What would be the best choice in Highlands and Islands? Andy Wightman or Alba? What would be the best choice in Mid Scotland and Fife? Martin Keatings or Alba? I would definitely go for Alba in Highlands and Islands. Wightman may have earned some badge of honour when he left the Greens seconds before being kicked out for opposing their genderist obsession. But deep down his stance on "local powers" is definitely devolutionist, close to Labour's "Double Devo" plan. His earlier statement, when he still was a Green MSP for the Lothian region, seems clear enough. But maybe he's embarrassed by it now, in the middle of a campaign where he has to seduce a fair number of Yes voters to get a seat in his new region. Which is not a valid reason to ignore a legitimate question about this, when asked already three times and still failing to give a straight yes-or-no answer. Pun fully intended and free of charge.

The choice is trickier in Mid and Fife. Keatings has definitely done a great job on the Section 30 case, and there is no doubt he is fully committed to Independence. He has also made a number of strong points supporting his candidacy as an independent with no ties to any party. So I would be split if I lived there, even though I would probably go Alba in the end. Not that it actually matters now, just hours away from the polling places closing. But, before all this big circus comes to an end, I must say I really appreciated the exchanges with Martin on Twitter. He did not shy away from trick questions, and managed to make his case in a very civil way, you might even say in a professional way. And in the end quite convincingly, even if you disagreed with him at first. Quite different to Andy and, if I may, better value for votes.

I wouldn’t stay at a spit-and-sawdust place in the East End of Glasgow

Not that there’s anything wrong with that

(Kevin Bridges, 8 Out Of 10 Cats Does Countdown, 2017)

A recent survey revealed that 61% of Brits can’t speak a second language

And in Glasgow, many people can’t even speak their first

(Jimmy Carr, 8 Out Of 10 Cats Does Countdown, 2015)

There has been quite a frenzy of polling in the run-up to the election, with seven polls released between the last two debates. Five of them were released on Tuesday and Wednesday, and could provide an interesting snapshot if they agreed. But of course they don't and the tiny differences between one poll and the next mean there is still massive uncertainty about the SNP bagging an outright majority or not. My own projection, based on the weighted average of last five polls and factoring in the regional crosstabs for over and underperformances, happens to be quite close to what you get on uniform national swing. As usual the larger sample tends to iron out the differences you find in individual polls. Both predict the SNP will win and end up with a two-seat majority. 65-63 if the next Presiding Officer comes from one of the opposition parties. But they might also argue that the past rota of POs means that it's again "the SNP's turn", so we would have an unwelcome 64-64 tie. Unless the SNP pull a stunt and offer the Chair to one of the Green MSPs. Don't rule that out just yet, just think twice about it and you will easily find the arguments SNP HQ could make up to support this. And stranger things have happened already, haven't they? Or we could actually have the dreaded government coalition, and it would not really matter if the headcount was 77-51 or 76-52, so even more reasons to cuddle the Greenies with that Chair and all its perks.

Of course this is, as always, a projection and not a prediction. Democratic elections are like a series of accidents waiting to happen, and you never know for sure what will come out of them, even if you have excellent polls. Finally, the last updated version of my projection of seats by region. The usual caveats apply here, more than ever, as we can't possibly account for all local factors that would trigger some upsets. One seat in particular has unexpectedly made its way up my to-watch list: Banffshire and Buchan Coast, that the SNP held on 55% of the vote in 2016. But the partly overlapping Commons seat, Banff and Buchan, is now the safest Conservative seat in Scotland. It is also the only Scottish constituency that voted to leave the European Union, though it is within the Aberdeenshire Council area that voted to remain, but much less strongly than the national average. On top of this, the SNP incumbent Stewart Stevenson, who succeeded Alex Salmond as the area's MSP in 2001, has decided to retire. This goes beyond removing the proverbial incumbency bonus from the SNP, as their new candidate Karen Adam has ruffled many a feather with her unwavering support for the genderist clique within the party. So I don't rule out some voters' rebellion that would deliver an unexpectedly high vote share to Restore Scotland's candidate David McHutchon despite their Eurosceptic and socially conservative manifesto. Then it might be enough to swing some of the never-mentioned SNP-Leave voters, plus a sizeable part of the anti-genderist SNP voters. If just one seat switches to the Tories, my tenner is on this one. Count there is scheduled for tomorrow, so the suspense won't last long.

So I will use this over the next two days, to check if the polls clearly reflected the true state of mind of the Scottish electorate. Which of course is again a "mebbes aye, mebbes naw" nail-biter. The most important part is of course the constituency seats. And any different outcome in the constituencies will mechanically reshuffle the list seats, so the number of list seats is the result to check here, rather than the exact order of allocation. Unfortunately we will not have returns from all the key battlegrounds tomorrow, but don't let that lay waste to your supper and a good night's sleep. Let's just hope everything will proceed smoothly all over the country and we have the final good news and bad news around suppertime Saturday. And bear in mind that whatever happens is just the beginning. And that the next five years will be quite interesting, to say the least. Be seeing you next week with the post-mortem.

Hopeless battles can be won, but only if you fight them

Even if you believe you are going to lose, you have to carry on

(Norman Tebbit)

Unfinished business tends to be something more life-affirming

Than beating someone up, just not if you’re Scottish

(Alex Millar, Being Human: The War Child, 2012)

The most recent Indyref polls have been quite a rollercoaster again, and mostly disappointing for the Yes camp. I have no doubt the Scottish Parliament campaign has had some influence here. The SNP and most of all the Greens have been known to make a better case for Independence in the past. Then the Unionist parties insisting on a "recovery first" priority in the short term may have influenced some "soft Yes" voters. Getting a referendum within the next two years is definitely no longer at the top of the electorate's priorities. But there is still strong support for holding it within the next Holyrood term, which might in the end prove a blessing for the SNP. They would have more time to rebuild a strong case for an Independent Scotland having better opportunities to deal with a slow recovery during the post-pandemic period. It would also give the whole Yes movement enough time to restructure itself, as new fault lines have appeared during the Holyrood campaign. I'm not usually one for too much caution, but it might pay dividends in the current context. Better get it right than get it right now.

The very last polls are also the least favourable, unfortunately. If you isolate just the last three, that were released in May, you get 47.7% Yes to 52.3% No when the fence-sitters are expunged. Factor in a 3% margin of error, and you can easily see that an hypothetical Yes victory would not be just far-fetched, but also narrow indeed. I'm quite sure the Unionists would be satisfied and rub our nose in it even if No won by the weeest of rat's ass' hairs. And then tell us tae git tae fuck noo, it was a one-in-seven-generations, so be seeing you in 2235. But we Yessers know that we would need a much wider margin to get the ball rolling without the fear of a "Florida 2000" scenario with, ye ken, recounts and court actions and the Full Monty and his python. We don't need to just win it, we need to smash it. Or, as the denizens of the Former Colonies would say, score a fucking slam dunk. I do think that the "we need a steady 60% in the polls before we do it" message should be kept quiet for now, as it may sound discouraging for some Yes activists, but 60% is probably what we would need to be on the safe side and avoid rearguard actions.

They want independence for Scotland? Give it to them.

It’s just that the balance of advantages and disadvantages of certain constitutional arrangements can become so unfavourable to the major entity as opposed to the minor or peripheral ethnicity that a judicious integration of possible disembarkation of the relations should be pre-emptively instigated by the former rather than attempting the perpetuation of a relationship which is inherently and indeed incontrovertibly prejudicial both to our own political interests and economic prospects.

(Sir Humphrey Appleby, Yes, Prime Minister: Scot Free, 2013)

31/05/2021

Signal Lost In All The Noise

18/05/2021

At The End Of The Counts

06/05/2021

What's New On Election Day?

We Must Be Dreaming