Of course you are assuming a strong form of Bayesianism here. Why do we have to accept that strong form?
More precisely, I see no reason why there need be no change in the confidence level. As long as the probability is greater than 50% in one direction or the other, I have an expectation of a certain outcome. So, if some evidence slightly moves the expectation in a particular direction, but does not push it across the 50% line from wherever it started, what is the big whoop?
Eliezer,
Of course you are assuming a strong form of Bayesianism here. Why do we have to accept that strong form?
More precisely, I see no reason why there need be no change in the confidence level. As long as the probability is greater than 50% in one direction or the other, I have an expectation of a certain outcome. So, if some evidence slightly moves the expectation in a particular direction, but does not push it across the 50% line from wherever it started, what is the big whoop?