iiinova — where your digital self keeps evolving

Catching up with the Labour Conference speeches and spotted this little zinger in @heidialexander.bsky.social’s speech. It has that nice quality of being new but still feeling like you’ve heard it before.

021

I'm sure that there's a lot more that you could do with these data, so I've made a CSV file of all of the estimates available on GitHub for you to download and work with: github.com/jackobailey/...

141

Likewise, you can click on the party names in the bottom left to see their vote share across Great Britain as a whole.

Clicking on different wards shows the corresponding MRP estimates. You can also search by name if you have a specific place in mind.

Jack Bailey @jack-bailey.co.uk · 27/09/2026

Sick of work this Monday lunchtime? Or perhaps hungover from one too many on the first night of Labour Conference? If so, distract yourself with ✨ 𝘄𝗮𝗿𝗱-𝗹𝗲𝘃𝗲𝗹 𝗠𝗥𝗣 𝗲𝘀𝘁𝗶𝗺𝗮𝘁𝗲𝘀 𝗼𝗳 𝘁𝗵𝗲 𝟮𝟬𝟮𝟰 𝗴𝗲𝗻𝗲𝗿𝗮𝗹 𝗲𝗹𝗲𝗰𝘁𝗶𝗼𝗻 ✨ Full interactive at: jack-bailey.co.uk/ward2024

Screenshot showing interactive ward level MRP estimates at the 2024 UK general election.

75927

🇬🇧 UK Poll of Polls: 27 September 2026 26% 🌹 Labour 23% ➡️ Reform 20% 🌳 Conservative 11% 🐥 Lib Dem 10% 🍃 Green 6% 👽 Other 3% ✖️ SNP

010

I would welcome more diversity in politics, but a focus on material issues rooted in this sort of sentimentality about who counts as working class is a death sentence for Labour. As @jamesdaustin.bsky.social says, you can't hold your own voters in contempt then begrudge them for not voting for you.

3666

This passage strikes me as misplaced conservatism. The *Labour* Labour Party of Harold Wilson legalised homosexuality by passing the Sexual Offences Act 1967. I'm sure many thought that a sort of "hyper-progressive drift" in the 60s, but few now would doubt that it was the right thing to do.

4372

Another acknowledgement that Labour hasn't worked out how to communicate. Yes, but this is downstream of the fact that it hasn't worked out what it thinks. It's policy platform is basically a greatest hits of Labour past: nationalise, regulate, tweak.

1200

On economics and immigration, we should worry that the present resembles the 1930s more than post-material politics. And the government needs to reckon with the fact that conditions are outside of its control. There's not much it can do if Putin invades Ukraine or Trump bombs Iran.

1181

For UK politics, the English language is a curse as it lets our politicos obsess over American politics. As such, the US looms large, despite its alien political system and culture. After the UK lurched towards multiparty politics, Labour strategists looked to a country with only two parties.

3274

This is a tacit admission that the media ecosystem changed but that Starmer's Labour didn't have a strategy to navigate it. There's nothing to stop "strong delivery" working as a message: Mandani does it. The difference is that he understands how to connect with people online.

1292

Expressive politics is not new. Gay rights were an issue in the 1980s -- 40 years ago! -- and no one in Labour would say that it was a mistake to pick a side in that fight. Likewise, second wave feminism in the two decades before. There was plenty of identity politics before the internet came along!

2340

The next paragraph highlights the persistent problem with Starmer's Labour: it lacked a clear philosophy. New Labour could be ruthlessly pragmatic because its leaders spent time working out a theory to explain the moment they faced, distilled it, then communicated it to the public. Starmer did not.

2222

But, though perhaps not intended to be, this concession is major: before Burnham took over, Labour were on *18%* in the polls and had been for months. If you reach that point and can't bounce back, you need to get out of the way.

1151

To be fair, Lockey does go on to concede that "it is hard to see how Keir Starmer could have achieved" a poll bounce like the one Labour saw when it replaced him with Burnham.

2120

In the first paragraph, you can feel the sense of frustration. Starmer's Labour had "a straightforwardly left-wing economic agenda", but still "the vibe was off" amongst progressive voters.

1160

Looks pretty similar to mine but lighter! What breed?

Jack Bailey @jack-bailey.co.uk · 23/09/2026

This is something that I'm interested in but don't have the time to work out. The variation around the Green's vote share is much larger than the Lib Dem's, who they're neck-and-neck with.

293

Jack Bailey @jack-bailey.co.uk · 21/09/2026

Not much to say other than I rediscovered this photo again and can't stop laughing about it

070

Jack Bailey @jack-bailey.co.uk · 20/09/2026

Is this a US English thing? My UK English Mac tills says "update", innit.

000

Jack Bailey @jack-bailey.co.uk · 20/09/2026

🇬🇧 UK Poll of Polls: 20 September 2026 26% 🌹 Labour 23% ➡️ Reform 20% 🌳 Conservative 11% 🐥 Lib Dem 10% 🍃 Green 6% 👽 Other 3% ✖️ SNP

020

Jack Bailey @jack-bailey.co.uk · 19/09/2026

It’s all fun and games buying grilled squid flavour crisps at the Chinese supermarket until you remember that at some point you actually have to eat them

Two packets of Chinese crisps. The left is grilled squid flavour; the right, fried crab.

150

Jack Bailey @jack-bailey.co.uk · 17/09/2026

Real politics fans know that this is the ultimate Bob Katter image

0150

In other words, this is what we're dealing with

Meme showing a character from Watchmen saying "Source?" and Dr Manhattan bathed in light responding "I made it up".

41914

Turns out, the author just ... made them up 🤷‍♂️. I can't get over this. It really doesn't seem ok to me for someone to draw some squiggles, claim that they're illustrative of something in the real world, then act as though it's akin to evidence. The vibes you feel aren't data!

No hard mathematical proof can be proffered for the shift from post- to anti- to hyperpolitics. Figure 1 is therefore less an exact representation of historical reality than an intuitive dia-gram, synthesizing fragments of existing graphs-of trade union density, incidence of political assassinations, size and frequency of demonstrations-offered by way of illustration.

1431829

As for the evidence, the author draws on all sorts of stuff in the book, mainly contemporary history and literature. But he introduces his thesis by referencing this figure, which purports to show politicisation and institutionalisation over time.

81209

As a result, it's hard to know what "hyperpolitics" really means. You have to chase it down. Of course, that makes it hard to argue against because the author can just say "you haven't understood it". E.g. here's the thesis statement, which does not define hyperpolitics at all.

This essay ventures a tentative definition, through the lens of what I term "hyperpolitics." The concept is meant to make sense of what comes after the mass politics of the short twentieth century (1914-1989), the postpolitics of the "very long" 1990s (1989-2008), and the antipolitics of the 2010s. Attention will also be given to a range of adjacent political forms, from populism to technocracy to millennial socialism. None of these morphologies is exhaustive of the period in which it arose-periodization is rarely so neat-while a vast corpus of empirical studies could be marshaled to evaluate each. Only a small selection will be discussed here. Hyperpolitics denotes a tendency rather than a totalizing style. Nevertheless, this book argues that it constitutes a gravitational pole in the force field of twenty-first century poli-tics. As such, it is indispensable for understanding the dynamics of mobilization and contestation that determine the contemporary scene. But these can only be grasped when situated in a longer history, linking past and present in an effort of patient distantiation.

2212710

Jack Bailey @jack-bailey.co.uk · 13/09/2026

Picked up a copy of "Hyperpolitics" on a whim after seeing it promoted all over the place and I can only conclude that Verso have an amazing marketing department because there's not much substance here and the evidence base is weak.

A copy of the book "Hyperpolitics" by Anton Jäger on a pub table next to a pint of lager.

1821443

🇬🇧 UK Poll of Polls: 13 September 2026 27% 🌹 Labour 23% ➡️ Reform 20% 🌳 Conservative 11% 🐥 Lib Dem 10% 🍃 Green 6% 👽 Other 3% ✖️ SNP

3146

Importantly, small changes can have outsized impacts on the chance a party wins. If fragmentation looks like GE2024, a vote share of 36% gives you a 50/50 chance of winning. But a vote share of 37.4% -- just 1.4 points higher -- shifts your chances from 50% to 75%.

In the appendix to the paper, I include some R functions to help compute effective thresholds. If the election looks like the previous election, the Greens would need ≈36% of the vote to stand an even chance of winning. If it looks like Makerfield, they'd need ≈46% instead.

But what if we use a different benchmark instead: the Makerfield by election. While not the same seat, this might give us a better idea of the dynamic, given that it included a party leader and was a by-election. Here there were only 2.35 effective parties instead.

Under first-past-the-post, districts only have one seat. So that just leaves the amount of fragmentation. Our best guide is the last election. If we square the vote shares at GE 2024, add them up, then divide 1 by that number, we find there were ≈ 3.35 effective parties.

There are two factors that move the effective threshold: the number of seats in the district and the amount of fragmentation, as measured using the effective number of parties. As seats and fragmentation increase, the effective threshold falls, making it easier for parties to win a seat.

But, in a new working paper, I show that we can do much better. So long as you have a good idea of how much fragmentation to expect, you can work out the vote share that you need to stand an even chance of winning -- a quantity known as the "effective threshold".

140

Jack Bailey @jack-bailey.co.uk · 06/09/2026

Before they start campaigning, parties often work out their "win number" -- the vote share they think they need to guarantee the seat. For example, here's some guidance from an old Labour Party document. Most of the time, this number is just an educated guess.

130

🇬🇧 UK Poll of Polls: 6 September 2026 27% 🌹 Labour 23% ➡️ Reform 20% 🌳 Conservative 11% 🍃 Green 10% 🐥 Lib Dem 6% 👽 Other 3% ✖️ SNP

040

Worth reading because this is one of a series of tables. The point is that it's true for one condition but not for another. Their point is that you should explain all the variance, but they don't unless they split out one of the conditions.

Gary King attests that it's true!

1220

Not sure whether to count that as denial or bargaining

1180