Don’t Look Up

This week: why aren’t we talking about the looming Torypocalypse? Plus some graphs of city densities; and the many defenestrations of Prague.

Another week, another poll showing the looming electoral disaster facing the Tories. This one, from Ipsos, really was a corker, putting the party on just 20 points, an incredible 27 points behind Labour. (The gap at the ruinous 1997 election was 12.5.) It was described by the Evening Standard as the worst result for the Tories since 1978, which momentarily suggested a source of hope in the blue corner – remember what happened the very next year! – except it turned out that 1978 was merely the year Ipsos began polling, not a year which saw a worse poll.¹ This was, to all intents and purposes, the Tories’ worst poll ever. 

And it would, at a general election, produce a Labour majority of 424, and make the LibDems the official opposition. 

This is, of course, insane – but while this poll is an outlier, it is not an especially distant one. In a recent Medium post with the cheery title “The Death Clock”, Alastair Meeks noted that polling averages even before that poll had suggested Labour majorities of around 300, way above the 1997 result of 179. “To be clear,” he wrote, “you don’t have to regard this as the single most likely outcome… [But] it’s what the polls are telling us... It has to be within the realm of unsurprising possibilities.”

And yet, I can’t help but feel it isn’t: this week’s discourse has instead focused on how big a tax cut Jeremy Hunt might deliver in today’s Budget and what this might do to the polls. (Never mind that he tried this just four months ago and it did precisely squat.) The fact his party is facing apocalypse – that on current polling he will lose his own seat, an outcome he’s donated £100,000 of his own money to avoid – was barely mentioned. As Will Jennings tweeted on Sunday, “The government is now polling as poorly as at the low point of the Truss premiership – yet nobody is talking about it.”

The obvious question is why aren’t we talking about it – not merely the government and its MPs, but any of us. Every bit of data we have suggests the political landscape is about to change, more dramatically than at any election in living memory, and no one is even mentioning it. Why?

Well… there may be a lot of possible reasons. Some of them are about the parties themselves:

• Institutional: To motivate activists and donors, the Tories need to pretend they're still relevant; Labour need to pretend they’re not complacent. 

• Experiential: History suggests many reasons why expecting Tory over-performance and Labour under-performance is a good rule of thumb. (Stephen Bush once observed that a child whose favourite colour is blue would have better predictive power than most “experts”.) Why expect that to stop now? 

• Temperamental: Let’s be honest here, Labour is a party full of miserablists who expect the worst and are rarely disappointed. They’re just not used to winning. 

Others are about the people reporting on them:

• Partisan bias: Some reporters/editors/owners literally just want the Tories to win. This, via newspaper front pages and the Today programme, helps to set the agenda of the media as a whole. 

• Status quo bias: The government also gets to set the agenda through legislation, statements and set piece events like the Budget. It can be hard, when your job is to report on such things, to keep hold of the fact that none of this matters any more because these people are toast. 

• Novelty bias: Oh, another poll showing what the last 40 polls did, is it? Well, that’s not news. 

• Personal bias: Even members of the lobby who aren’t naturally pro-Tory may be worrying about what a change of government will do to the contact books and access they’ve built up over the past few years. That, too, might push people to ignore the big burn-y asteroid in the sky.

• Anti-boredom bias: Look, if your job is to write about the horse race aspect of politics, and there isn't a horse race because one of the horses is obviously dead, and this might be true every day for the next 10 months, then you’re probably going to pretend the lovely blue horsey is fine just to stop yourself going nuts, aren’t you? Add all these and a bunch of other stuff together and what you get is...

• Herding bias: Nobody wants to be the only one inside the machine predicting the thing everyone else thinks is mad. A high chance of being wrong in the same way as everyone else is somehow less scary than a low one of being wrong on your own.

And then, there are the cognitive biases that can affect every one of us: 

• Using the past as a guide to the future. No one alive in Britain has lived through the kind of realignment the polls currently suggest. Our range of plausible options is limited to 1997 and 1992.

• Keir Starmer is a bit rubbish. How on Earth could he win a bigger majority than Tony Blair? Come on.

• “Reversion to the mean” bias. As the election approaches the polls will inevitably narrow, because... well, they just will, won't they? 

Actually, they don’t always, and when they do it’s because the parties do something, not an inevitably result of gravity. Lastly: 

• The Tories are the natural party of government. 

It’s possible to feel this is true of Britain, even if you hate the party and all it stands for. So the idea it might genuinely be facing wipeout is unimaginable, like waking up one day to find – not that the sun has gone out, because that would be bad, but like a pain you’d been living with your entire life just wasn’t there any more. It’s unthinkable. 

Maybe it won’t happen. A 1997-style result feels highly likely at this stage, but I still, if pushed, wouldn’t put money on the Tories doing substantially worse than that, because... Well, I can’t really justify why. It just feels like it can’t happen, even though the data all points in that direction.

But that’s surely thanks to many reasons – ideological, institutional, cognitive – that many people are assuming the final result can’t be like the polls suggest. To put it more simply, there are a lot of people, not all on the right, who have an interest in over-stating the Tories. There are very few who have an interest in over-stating Labour.

Sometimes, as I’ve noted before, it’s the people preaching moderation who are wrong, and those hysterically waving their arms who are right. Nine years ago, after all, poll after poll showed another British party on course to lose almost all of its seats. Party elders were wheeled out to explain why the data couldn’t be trusted, why the methodology was flawed, how there was only one poll that mattered. And on election night 2015, the Liberal Democrats went from 56 seats to just seven.

That doesn’t mean this, or anything remotely like it, will happen to the Tories. But nonetheless, the data says what the data says. Sometimes, there really is a wolf.

This week’s act of shameless self-promotion

Somehow, presumably thanks to my brilliant publicist Federica, a “bookstagrammer” named Ella got hold of an early copy of my book. “I wasn’t sure if I was going to like it, simply because I am TERRIBLE at geography and usually hate it because of that,” she wrote. “But this book proved me wrong – I loved it and learnt a lot! Jonn Elledge is a brilliantly witty writer, who creates a compelling narrative whilst providing a lot of information.... I highly recommend picking this book up when it comes out on 25th April!” Well, you heard the woman.

I realised this week, however, that there is at least one factual error in the book, which it is tragically too late to correct. Luckily, though, I have a newsletter:

The many defenestrations of Prague

One of the most consequential events in all European history was the Defenestration of Prague, which was exactly what its name suggests: some blokes in the capital of what was then Bohemia and is now Czechia² being literally chucked out of a window. Put like that it sounds pretty funny, and from some perspectives it was – just not the one where it sparked one of the longest, most destructive wars in the history of the world.

One of the ways in which it is funny, however, is this: it’s not the only historically important incident involving someone getting chucked out of a window in Prague. This is not quite as unlikely as it sounds – thanks to the combination of the power of the mob and the importance of Biblical stories, especially the fate of Jezebel, the act of chucking people out of windows was a fairly common form of political punishment in early modern Europe. The weird thing about the defenestrations of Prague, though, is both how often they happened, and how they tell a story about the Reformation (through, let’s not forget, the unusual medium of chucking people out of windows). All in all, there’s an extremely strong case for staying away from window ledges in the Czech Republic.³ 

A contemporaneous leaflet showing the 1618 Defenestration of Prague. Image: public domain.

The first defenestration of Prague happened in 1419 and involved a group called the “Hussites”. Four years earlier, Jan Hus – priest, philosopher, a sort of proto-Martin Luther who’d spoken out against the excesses of the Catholic Church – had been tricked into attending a meeting with a promise of safe-conduct. He was promptly arrested and, when he refused to recant, burned at the stake. This had not served to make him any less popular. People kept becoming Hussites; the authorities kept locking them up.