General Election '24

[lpm]

https://x.com/nadhimzahawi/status/1806065206096691478

[shpalman]

I have always voted Labour says... Conservative counsellor Amanda Lloyd-Harris...

[Grumble]

https://x.com/nadhimzahawi/status/1806065206096691478

Is it worth pointing out that the rest of the expedition also died, despite the sacrifice being well intended?

[lpm]

https://x.com/nadhimzahawi/status/1806065206096691478

Is it worth pointing out that the rest of the expedition also died, despite the sacrifice being well intended?

THEY ALL DIED.

I'm taking it to mean Zahawi is predicting

LAB 525
LD 75
CON zero

[TimW]

...tactical voting will very much be more effective in this election than possibly any previous election. Estimates suggest somewhere in the region of double the 1997 vote, which was seen as a very tactical election.

Fair enough if there's going to be *more* tactical voting, as long as they vote for the right candidate!

[lpm]

Whilst there are some seats which are confusingly drawn up in terms of who to vote for to get the tories out, in most it's clear, and tactical voting will very much be more effective in this election than possibly any previous election. Estimates suggest somewhere in the region of double the 1997 vote, which was seen as a very tactical election.

Problem is we don't know how much of the tactical vote is already within the polling numbers.

"I'm voting Labour and I haven't thought about tactical voting yet" is different from "I'm voting Labour, I thought about voting LibDem as Labour can't win here, but I rejected tactical voting as I prefer to support my team".

But I guess all that matters is targeting undecideds with the tactical vote message. There's an awful lot of "Don't know, might not even vote, but there's no way in hell I'm voting Tory again" voters out there. The way to reach these is via Facebook and Nextdoor and face to face, but of course the twitterati think it's done by tedious patronising tweets.

[El Pollo Diablo]

Quite a lot of the "no way in hell I'm voting for the Tories again" cohort are in that camp because they're dead, as well. I saw a tweet some weeks ago that said that because of the way the vote splits by age so heavily these days, in each parliament there's something like a natural 2% swing against the Tories towards Labour as old people die and young people become eligible to vote.

Dylan Difford has been doing some good charts showing this:

[TimW]

Voting Tory is bad for your health.

[Stranger Mouse]

I have always voted Labour says... Conservative counsellor Amanda Lloyd-Harris...

Apparently she’s a different person. https://x.com/greghands/status/1806241786379051133?s=61

[El Pollo Diablo]

This is useful for people wanting to see who to vote for: someone has collated all the MRP outcomes into a single handy table by constituency.

https://t.co/Q4JeIqWKxu

[lpm]

This is useful for people wanting to see who to vote for: someone has collated all the MRP outcomes into a single handy table by constituency.

https://t.co/Q4JeIqWKxu

That is AWESOME.

I thought about trying it, then realised it was far too much work and I'm far too lazy.

[El Pollo Diablo]

Are there any seat calculators that aren't electoral calculus (where you can enter voting intention and see seat forecasts)? Because their seat outputs are now disturbingly narrow and I'd like to see a bit more variation

[discovolante]

This is useful for people wanting to see who to vote for: someone has collated all the MRP outcomes into a single handy table by constituency.

https://t.co/Q4JeIqWKxu

Absolutely nothing to do with the election, just reminded me:
https://www.theguardian.com/world/2022/ ... os-islands

[monkey]

Leader of a party about to get ~2/3 of the seats on ~2/5 of the vote is pro first past the post shocker - clicky

[IvanV]

Are there any seat calculators that aren't electoral calculus (where you can enter voting intention and see seat forecasts)? Because their seat outputs are now disturbingly narrow and I'd like to see a bit more variation

This is useful for people wanting to see who to vote for: someone has collated all the MRP outcomes into a single handy table by constituency.

https://t.co/Q4JeIqWKxu

The FT has a model you can fiddle with, and shows a constituency by constituency outcome. It doesn't seem to need a login to use it, unlike a lot of FT content. But it's very simple, just change the vote shares.

I do have a login for The Economist, and I'm logged in so I don't know if I'm looking at subscriber-only material. But you can't fiddle with their model, even as a subscriber. Their model has already moved markedly from the report at that link - and that is only yesterday - and very considerably from its earlier prediction that looked so implausible. But, as was said in a recent post, once they get below a vote share of around 25%, seat outcome becomes very sensitive. The Economist are apparently updating it daily using the latest polling data. They are now predicting (excl N Ireland) as follows, showing median with 95% confidence intervals:

Con 117 (32 to 214)
Lab 429 (331 to 519)
LibD 42 (15 to 86)
Reform 2 (0 to 82)
SNP 23 (0 to 55)
Other 2 (0 to 5)

The ranges, as with every sensible predictor, are still very wide indeed. And I think we agree that is realistic. It is very hard to predict. The model has no provision for tactical voting, nor any adjustment for seats with local curiosities.

Wasn't I thinking 120 might be a benchmark for whether the Tories were having a bad day or not, and thinking perhaps I was setting myself up to be more likely happy than unhappy. But according to that I'm set merely at the median of their latest prediction.

The electoral calculus model, I have no idea what those "tactical voting" parameters mean. I tried playing with them, and what I got didn't make much sense to me.

[TopBadger]

In any case - seems LD's aren't going to form the opposition, and Badenoch will keep her seat... shame.

One week to go!

[El Pollo Diablo]

The FT has a model you can fiddle with, and shows a constituency by constituency outcome. It doesn't seem to need a login to use it, unlike a lot of FT content. But it's very simple, just change the vote shares.

This is fabulous, thank you!

[Lew Dolby]

FT is weird. They don't seem to take account of recent by-elections, just the GEs.

Here in N Shropshire, they're the only ones giving a Lab win. A year or three ago, we had a by-election when the awful Owen Paterson was ousted: a LibDem win. The LibDems are only ones putting any effort into campaigning here. LD bunf almost daily: one Green bunf: that's all.

[All bunf is wasted in thsi house. We votes ages ago by post. Good job we had postals as we're both down with the CoViD at the mo.]

[El Pollo Diablo]

In any case - seems LD's aren't going to form the opposition, and Badenoch will keep her seat... shame.

One week to go!

It's not very likely, certainly.

That said, there are three polls out today - Lord Ashcroft, YouGov and BMG. Electoral Calculus shows the Lib Dems as the opposition if any of those three polls is correct, even without the weird tactical voting thing. The FT show the Libs as the opposition with the YouGov poll, and within 15-20 seats of it in the other two. So it's not likely, but I live in hope that the hilarious might become possible.

[IvanV]

FT is weird. They don't seem to take account of recent by-elections, just the GEs.

They clearly document that they take no account of constituency level data. So they also make the unlikely prediction of a Con victory for Chesham and Amersham, where LibDems massively won a by-election. Their model seems to be regionally disaggregated, but not down to constituency level.

Electoral Calculus and the Economist have have individual models for individual constituencies, but that's mainly through data on local social characteristics. But even then, they don't take account of local personality issues, specific local controversies, etc. In general constituency-level opinion polls are not available, so they don't usually provide for it.

[El Pollo Diablo]

I'm going to keep track of what electoral calculus and FT say for seat numbers for each poll that comes in from now until the end of Wednesday, and use that to form one view of how Thursday might go

[lpm]

So it's not likely, but I live in hope that the hilarious might become possible.

"Hope for the Hilarious" is going to be my mantra for the next week.

[noggins]

If the fight for second biggest part is really close, can a couple of parties officially ally and add their seats together ?

[lpm]

If the fight for second biggest part is really close, can a couple of parties officially ally and add their seats together ?

You mean merge, so that they become the official opposition? I guess it would have to be LibDem merging with SNP to become the SNPLD.

Not sure without a full merger. The leader of the opposition gets a government salary and the opposition gets special days to debate stuff in the Commons. I believe the Speaker gets to decide the timetables and rules - so for things like PM questions it is just convention that the leader of the opposition gets a series of questions. No reason why the Speaker couldn't change tradition to reflect a political reality.

[shpalman]

If the fight for second biggest part is really close, can a couple of parties officially ally and add their seats together ?

You mean merge, so that they become the official opposition? I guess it would have to be LibDem merging with SNP to become the SNPLD.

Not sure without a full merger. The leader of the opposition gets a government salary and the opposition gets special days to debate stuff in the Commons. I believe the Speaker gets to decide the timetables and rules - so for things like PM questions it is just convention that the leader of the opposition gets a series of questions. No reason why the Speaker couldn't change tradition to reflect a political reality.

Can you have a coalition of opposition?