When Joe Biden announced that he was withdrawing from the Presidential race, 538 suspended their prediction model, replacing it with a red sign saying "Forecast suspended":
https://projects.fivethirtyeight.com/2024-election-forecast/
There is a lot of interest on Manifold as to what the updated version of the model will show once it comes out:
/ManifoldPolitics/when-538s-next-model-comes-out-what
When will it come back online, showing us a percentage chance for each of Harris and Trump?
Market notes:
This resolves to the first day on which 538 is showing a model at this page or a similar page. A sneak-preview on Twitter or any other site will not count.
The market resolves based on Washington DC time (any time after midnight counts as the next day)
If it looks like it might be longer than a week, I will add more days and anyone better on "other" will receive an equal number of shares in those options. If you don't think that they will ever publish a Harris vs Trump model, you can still bet on Other.
1,000
3.00It took a strangely long time, but 538 have finally put their forecast back online and it's showing Harris with a 58% chance of winning the Presidency:
https://projects.fivethirtyeight.com/2024-election-forecast/
This resolves to Friday 23rd August.
Will Trump catch up with harris in either the 538 forecast or Nate Silver's forecast?
/SimonGrayson/when-will-trump-overtake-harris-in-tq3ucuxus2
/SimonGrayson/when-will-trump-overtake-harris-in
And will RFK Jr upend the race by withdrawing and endorsing Trump today?
I've added dates up until the end of August. Could it really be delayed even further than that?
If you just want the binary market @Simon74fe has created one here:
It says "presumptive" nominee, which Harris has been for quite a while (at least since the DNC announced she'd gotten enough of the virtual roll vote call to win the nomination, which was back on August 2nd).
So, why hasn't 538 published a new forecast yet? Or even updated about when the new forecast might be seen? A common speculation is that there's something seriously wrong with the new-this-year 538 model, something that the Harris switch made impossible to ignore, which they're scrambling to somehow fix.
This doesn't come out of nowhere; something being wrong with the new 538 model was speculated about before Joe Biden dropped out. The new 538 model was showing Biden's probability of winning running way ahead of his polling, and there was an anomaly where the state-level breakdown in some states was predicting Biden doing better than both the polls and the model's "fundamentals" analysis. The guy who designed the new-this-year model (G. Elliott Morris) said that was actually working as intended (something about inter-state correlations), but, well, there was a pretty skeptical reaction to that.
@traders Other jumped from 40% to 90% yesterday so I've added another week's worth of options.
If you had bets on Other, you now hold shares in those new options as well!
deleted