Bauschard responds on MJP

Stefan's post to the NDCA:

I think that the expectation of high (1/2) prefs is just part of the
growing pains. I know from running many tab rooms that policy debate went
through the same issue.

When tabbing tournaments using MPJ (which is almost every tournament I
tab), I distribute prefs that following way --

R1, R2 (Presets) Even distribution across all teams (I try to keep it to 4
or less (1, 2, 3). Sometimes a mutual 5 slips in.
R3 - priority -- down 1, down 0, down 2
R4 - priority -- down 2, down 1, down 0, down 3
R5 -- priority -- down 2, down 1, down 0, down 3, down 4
R6 -- priority -- down 2, down 1, down 0, down 3, down 4, down 5

Once teams are down 3, I don't pay much attention to prefs. I do try to
keep it at 7 or below (on a 9 point scale). I actually favor ordinals, but
not because ordinals can enhance preference but because ordinals can
increase mutuality, if used properly. I my ideal world, debaters would
debate for 50% of the pool, but on mutual basis.

I do always use 9 categories because it enhances mutuality (though it does
make it appear that people are getting lower numbers).

I do think that MPJ is the best system that we have come up with. I'm
opposed to random judging because

(a) Empirically, it is not implemented "randomly"
(b) Some judges are better at judging than others (as people vary in the
skill of anything that we do) and it is a competitive event


I do think that random is better than "tab room decides who the best judges
are."


And my response:

Responding to Stefan:

That the rising tide of people storming the tab room wondering where all these 3s came from is a part of the growing pains is no doubt correct. As MJP settles in and becomes both the norm and virtually universal, people will learn that it does not always automatically mean 1-1 matches. But at the same time, a lot of that storming was accompanied by the request to put in a 2-1 if possible, even if the team making the request was “disadvantaged” by it. As I said originally, at the point where we allow this to happen, aren’t we effecting striking about 60% of the pool? This doesn’t seem right to me.

(I put disadvantaged in quotes because, at some level, I find it hard to believe that all the time and effort put into preffing is really that productive, but that’s a discussion for later down the road. My point it that, beside simply listing the judges in the order of preference, I know there are teams that have complex ranking systems that they believe are to their advantage. There’s a certain whatever-ness to this in my mind, in that any mathematical system will result in people trying to apply elaborate solutions that may or may not work, so an argument that people can game the system—which I’ve heard as a knock against MJP in general—is, in the end, non-unique. More to the point, I question how big a disadvantage a one-off pairing actually is, in a universe where the debaters can see the name of the judge, go read a paradigm, and prep accordingly. It wasn’t that long ago that teams got whatever judge was thrown at them; isn’t MJP just a better way of throwing judges at them?)

Question to Stefan: Your round 3 order is down 2, down 0, down 1? Always assuming down 2 is the bubble, why this way? I’ve always gone down 2, down 1, down 0 based on the premise of trying to keep people away from the bubble. Of course, the first 3 rounds of any tournament are hard, because realistically speaking no one is out of competition yet, so for all practical purposes you’re trying to preserve everyone’s chances.

Second question: 9 rounds doesn’t work for flighted debates in TRPC. But I think that if it did (we can simply assume that it’s a technical issue, and TRPC is only one tabbing system, albeit a fairly common one), it does not solve the problem of ghettoizing a substantial portion of the pool. If you ask me, it would just make it harder to find matches. And more importantly, you claim that in your ideal world debaters would debate for 50% of the pool. In my ideal world, debaters would debate in front of most of the pool. Why would I want to render half the judges useless? Granted that any pool has a handful of judges who have, indeed, been dropped on their heads as infants a few times too many, for the most part the pool ought to be used on a scale of proficiency for competitive needs that somehow includes the vast majority of them. Otherwise why are we asking them to show up?

Third question: I think the underlying issue that may illuminate a lot of things, is that, with your preference toward 9 tiers, why 9 versus 6, since you could presumably do either one? I would like to understand the thought processes behind the preference. My intuitive thoughts are toward fewer tiers because the fewest number means the easiest divvying up (but I think I may wrong about that). For argument’s sake, I think we should envision two different kinds of tournament, one with over a hundred entrants and a lot of judges, and one with, say, 60 or so entrants and a fairly tight number of judges. I can easily imagine that in a big tournament, 6 makes sense, while in a smaller tournament, maybe only 3 or 4 tiers make sense. In any case, the rationale for the difference between 6 and 9 should apply to some extent.

And, of course, thanks for responding in the first place. I really think that tournament procedures need to be clear and open, and I can’t imagine a better place to discuss them than here.