Well that works if you wait until you get an answer and then make adjustments. But I think most people (hopefully at least) manage to avoid that degree of bias. But there’s still the issue of how you determine the prior probabilities in the first place—after all it’s not like you can’t make a good estimate of what sort of numbers are likely to produce the result you want before you ever do any calculations at all.
Well that works if you wait until you get an answer and then make adjustments. But I think most people (hopefully at least) manage to avoid that degree of bias. But there’s still the issue of how you determine the prior probabilities in the first place—after all it’s not like you can’t make a good estimate of what sort of numbers are likely to produce the result you want before you ever do any calculations at all.