Even "62% true" is too weak: it summarizes the posterior by a certain integral. This drives home why we don't do either, of course: it's far more expensive.