> I think the whole premise that arbitrary priors and dreaming up probabilities and then washing them all together isn't applicable to a unicorn type event. No

> I think the whole premise that arbitrary priors and dreaming up probabilities and then washing them all together isn't applicable to a unicorn type event. Now I believe you said it. Maybe not exactly "made up priors" so pardon me for the misquote. But I'd be interested to see if you can make a distinction.