alt Hacker NewsGiven infinite outcomes, experts in a field without a theory of the world that can be used to calculate the future will always perform worse than chance.
Experts of engineering perform better than non experts. However the field of political behavior (or economy) is difficult. The only way to know what's going to happen is wait for it to happen.
Sometimes you know more or less what's going to happen but not the details or the exact outcome. That's enough to make plans.
Examples: at the beginning of 2024 we average persons knew that Putin would win the Russian elections no matter what. We average persons also knew that either Trump or Biden would win the American ones but we didn't know whom. We have to wait. Then surprise, it became either Trump or Harris.
Maybe there are people around the world or even the USA that wonder why Obama don't run for president instead of Harris. They are not experts of the rules of the competition.
So the question is, do the experts predictions are consistently worse than the predictions of any randomly picked person?
Experts who felt certain about their ideas did worse than chance, and worse than simplistic models.
Experts who quantified uncertainty and tried multiple theories did better than both chance or simplistic models.
Sadly, the experts who felt certain presented themselves with confidence and got higher paying jobs.
Normal people were not in the data set reported. He's since done more research on good predictions. You can read Superforecasting by the same author for more.