Prize Market
The election has been announced for July 4th, and the Labour party is expected to take power from the Conservatives. But how much will they win by?
You can find more markets about the election on the dashboard.
Related questions
Come bet on the Reform UK party results
https://manifold.markets/MichaelDarmousseh/how-many-seats-will-the-reform-part
425+ is now the clear favourite, but Labour is only favourite in 425 of the individual seat markets, so there are some corrections to be made here if anyone has the time or the mana!
That must've taken a while to count up! FWIW, it would be possible for them to be favourites in <425 seats while still being expected to win 425+ in total (as a result of being high second-favourites across many seats), though it does seem to be the result of a discrepancy.
I went through the Lincolnshire constituencies (for no good reason other than I looked at them earlier) and flipped a few of the markets to make them in favour of Labour:
I also made Labour the favourite for Dumfries and Galloway (though not >50%). It's a lot of work going through them all though — would be nice if the bar colours matched the political parties when scrolling through the previews!
Labour are on track for a 262-seat majority, with the party currently leading in 456 seats
Survation’s most recent MRP poll
This is post-Farage but some of the polling is pre-D-Day so this really feels like it’s not going to get better for the Conservatives.
I love to see the trading activity here! I'm adding a subsidy to incentivize price discovery. Enjoy the free mana if you're right, @traders !
Also, check out @AustinPlatt 's new site summing up Manifold's forecast for every seat!
He's in the comments below talking about it and he says it's open source and accepting PRs!
@PontiMin Not really. The outcomes of each constituency are not very independent.
@thepurplebull someone could do it as a Monte Carlo simulation. E.g if a constituency has a probability of 60% Labour, 35% Tory, 5% Lib Dem, then use those probabilities in assigning that constituency to parties. Do 1000 runs and get the average number of seats won by each party. Would this provide an accurate forecast? I don't know but it would be interesting to find out.
@PontiMin I made the site here https://manifold-uk-election-2024.netlify.app/. I agree that just taking the most likely party from each constituency as I've done isn't a perfect idea idea. A monte carlo simulation seems like a better idea. I have all the requisite data for it, it'd just be a question of implementation. Hopefully I can do so during this long weekend. Otherwise, the code is open source and I would of course happily accept PRs.
But I also agree with @thepurplebull that the winner of each constituency isn't really independent of the others -- I'm not sure how you could model that in the simulation.
@AustinPlatt This is great work! I'm sending you 10k mana on the spot as a bounty, and I think there might be plenty of people who would liek to help on the manifold discord server.
I've just added the monte carlo simulation which should theoretically give more realistic results (and seat ranges for each party) https://manifold-uk-election-2024.netlify.app/
@AustinPlatt I think it would be cool if you could add the betting odds from bookies for a side by side comparison. It might fall against the purpose of the site, but would be interesting to see where the two differ the most.
@Riley12 I'm going to try and get the manifold forecast added to https://inglesp.github.io/apogee/. That seems to be developing as a central place to compare different predictions. Potentially we can get some bookies data on there too.