Un projet des gvrnmts suédois & ukrainien, sur une plateforme de sondages prédictifs @Hypermind_com. Ouverte au monde entier, en 🇫🇷🇬🇧🇸🇪 L'intelligence collective comme arme de guerre contre le mal : https://t.co/ibrITFQiiW
@gernelle@mathieulaine@sebastienlefol@guillaumeerner
Final pre-election day forecast landscape. Skilled forecasters give an edge to Harris. Betting markets (except PredictIt) give an edge to Trump. Statistical models occupy the toss-up middle-ground.
Harris' chances continued to improve overnight.
Interestingly, the US-based mrkts give her better odds (51% avrg) than the foreign-based mrkts (46% avrg).
She's the favorite of the skilled forecasters panels (Hypermind, Good Judgment, Metaculus), with 55% probability avrg.
@Leucoium_vernum@robinhanson Very few random frenchmen have this kind of money. My guess is that he's a Russian national with a recent French passport, as in Pavel @durov
It looks like there is consensus on Harris trending back up in the final stretch. @Hypermind_com is still the most optimistic for her vs @polymarket the most pessimistic.
Herding behavior or crowd wisdom? There is cognitive diversity in the sense that all these platforms operate quite differently (geography, demographics, rewards, etc.). Yet, they may not think really independently of each other (except 538).
It is remarkable that our prediction market is now (Oct 26) the only crowd-forecasting or statistical model that still gives Kamala Harris slightly more chances of winning (54%) than losing. All others are now favoring Trump more or less boldly.
Join the FIGHT to make the most accurate forecasts about swing states and congressional control! New contest launched today with > $1,000 reward. https://t.co/QeP6ufkRUG
In today's Wapo, @IgnatiusPost mentions an intelligence estimate that looks very much like an OVERestimate compared to the prediction market's at the time. Just one data point, of course, but it's so rare to be able to compare these things...
The 2nd assassination attempt on #Trump has had no impact on his chances to win the election as estimated by our prediction market. This is in stark contrast to the uplift he got from the first attempt. #newnormal ?
@fadouce Les marchés prédictifs ont basculés en faveur de Harris depuis bien longtemps. @Hypermind_com depuis le 23 juillet, suivi de tous les autres une semaine plus tard. :)
Our prediction market thinks that @realDonaldTrump's chances of winning back the White House (33%) are lower than at the same point in time in 2020 (37%) and 2016 (40%). @KamalaHarris@VP is in better position to win than Clinton (pre-Access Hollywood tape) or Biden were.
Today, on our prediction market Trump's chances of winning back the White House reached a new low, about 33%. In fact, they are as low as Biden's chances on the day that he dropped out of the race in July.
@hendrycks I believe in the potential of AI forecasting, so much so that @Hypermind_com went public with its own a few days ago: https://t.co/XtxcQspMYk
However, prediction markets, crowd forecasting and AI forecasters do not serve the same customer needs at all. They will coexist nicely.