1,000
3.00>+80%
How did everyone manage to get this so wrong?
Is this just Manifold’s Dem bias or something else?
@elf the famously left wing polymarket had it at 28%, fwiw
+80% in an instant should happen about 1 in 5 times for well calibrated markets at 20% when the result hinges on an announcement (and basically nothing else). you can debate whether markets were too low on it but there's only one way to meaningfully do that (bet on the other side, on the highly liquid polymarket)
@Ziddletwix Great point re: Polymarket.
I didn’t have a strong opinion on the matter but I would have assumed this would be the sort of thing prediction markets would be much better at forecasting given there were a number of signals prior to the decision (eg trump winning, evidence in Hunter’s case etc)
@elf Markets seem to be biased towards what a politician says he'll do, regardless of his actual incentives.
I would have assumed this would be the sort of thing prediction markets would be much better at forecasting given there were a number of signals prior to the decision (eg trump winning, evidence in Hunter’s case etc)
the outcome of any single 80/20 market should by itself update you very little about how accurate prediction markets are at that topic.
it is perfectly fair to believe that this is an example of a broader systemic bias where prediction markets overrate the words of politicians over their incentives for personal gain. and if you believe that, going forward, you will certainly be able to find bets you can make based on that systemic bias. but "markets had this at 20% and it happened" by itself tells you very little in either direction. if this were a topic prediction markets were perfectly calibrated, 20%-ers would routinely resolve YES as well.
certainly, a good argument for if prediction markets are bad at this topic is if you feel the result was obvious, and thus 20% was a bad forecast (regardless of the result). and if people felt that outcome was obvious, there was a very profitable bet available on polymarket. (but "shouldn't this have been obvious" is very different from "this was obvious")
if this were a topic prediction markets were perfectly calibrated, 20%-ers would routinely resolve YES as well
I think you can make that argument but I've never collected the data. You made me curious.