The neural net updating algorithm calls for a nudge in the appropriate direction of “Beware of dismissing those who speak with what you think is too much confidence.”
No, it doesn’t, not out of thousands of predictions of which you selected one post hoc. If I may quote you, our minds do not run on floating point beliefs.
Yes. If you look through the threads or the PredictionBook entries, there are plenty of people blowing predictions. (A particularly good example was the Wizengamot trial: how would Harry rescue Hermione? The actual solution was 1 of the top 2 or 3 suggestions, but that still implies a lot of people favoriting wrong solutions.)
Even if 75ths predictions aren’t just luck, you don’t have enough information to meaningfully update across such a broad reference class. If it’s got to overcome the weight of everyone I think is speaking with too much confidence on the other end of the lever, it’s not going to move far enough to be noticeable.
No, it doesn’t, not out of thousands of predictions of which you selected one post hoc. If I may quote you, our minds do not run on floating point beliefs.
If 75th’s predictions were just luck, wouldn’t it be likely for there to be other people who got a smaller number of the predictions right?
Yes. If you look through the threads or the PredictionBook entries, there are plenty of people blowing predictions. (A particularly good example was the Wizengamot trial: how would Harry rescue Hermione? The actual solution was 1 of the top 2 or 3 suggestions, but that still implies a lot of people favoriting wrong solutions.)
Even if 75ths predictions aren’t just luck, you don’t have enough information to meaningfully update across such a broad reference class. If it’s got to overcome the weight of everyone I think is speaking with too much confidence on the other end of the lever, it’s not going to move far enough to be noticeable.