Empirically, it seems that pundits who make mistakes rarely admit it. Therefore, admitting it is evidence that he's got more integrity than most. (Or, of course

Empirically, it seems that pundits who make mistakes rarely admit it. Therefore, admitting it is evidence that he's got more integrity than most. (Or, of course, different readers, or a different mental model of what his readers care about, or something, but whatever it is it seems to produce similar results to integrity.)