I'll second this with one minor caveat. They're not necessarily "downmodding offenders" until proven as such by the data you're requesting. They're just the top 50 downvoters.
I too am curious to see the downmod/upmode ratio (anonymously of course). I think mine is probably 10-20 upmods to 1 downmods, and I mod sparsely (maybe 1 comment in 20? depends on the thread). I downmod "noise", not signal I disagree with.
While we're geeking out, I'd be more interested in some kind of karma/comment metric than absolute karma.
Maybe. But some people tend to see the good in things, and some tend to focus on the bad. I think it could be reasonably hypothesized that persons who downvote a lot focus more on the negatives.
It's hard to read contextualization or to infer what we refer to as vocal inflections through writing; I am as guilty as anyone of the occasional "posting while drinking" (which I've noticed does tend to make my writing seem a little bit more hostile than intended); however, some attempts at humor here just end up falling pretty flat.
Down-modding is not what brings the interesting conversations out of the woodwork; up-modding does. So they are "downmodding offenders", especially if the comment was previously up-modded from its original default score.