According to this old answer, global choice implies its own version of the well-ordering theorem. Namely that
Every proper class can be put in bijection with the class of ordinals.
There are two things about this that are shocking to me. First, the statement itself, that there is only one cardinality for classes strikes me as a surprising fact, although I have no intuition to believe otherwise. Second and more shocking (to me at least) is that this class well-ordering theorem seems to make no reference to ordering whatsoever. So my question is two-fold: how do we prove this claim, and why is it parallel to the well-ordering theorem?