Research I have seen suggests we do construct Bayesian prices but we don’t apply them rationally. If people assess that something happens 60% of the time, then

Research I have seen suggests we do construct Bayesian prices but we don’t apply them rationally. If people assess that something happens 60% of the time, then in a series of 10 observations they will predict it happening only 6 times. That is to say, they will bet on the 40% probability outcome 4 times, even though they believe the 60% is more likely each time