Re: Truest competition format

<<we wanted to do
 it all in one
day>>

Unfortunately, there's a real physical limit to doing that.
Consider if you get eight teams, two trash-better, two
academic-better, and four about equal at both. For a single
round-robin (seven rounds, or eight with submission), there
is going to be at least one medium team that ends up
facing the trash-better teams on trash rounds and the
academic-better teams on academic rounds. You can't avoid this
without initial seeding; I haven't done the math, but I
strongly suspect you can't avoid it at all.


Consider, then, a double-round-robin with eight teams, in
which each team faces each other team on both an
academic round and on a trash round. This is fourteen to
sixteen rounds in one day -- a stretch on ACF-length
questions, but possible -- but where do you get the packets
from? I can't see teams submitting two packets apiece;
I suppose DUCKS could generate seven packets
internally, but I've never seen that work really well.


This is, of course, all scaled down to eight teams. If
you have ten, things get much much worse...18 rounds
in one day is probably beyond reasonable boundaries
if teams want to be fit to drive home.

Then
there's the question of exactly what you're testing. With
all due respect to those whom I am about to mention,
the Chicago and UIUC squads have placed so highly
nationally are not, with the same roster, going to place
equally well on trash rounds. Then, atop that, there's
the inherently different nature of squads in trash
and academic events; apart from explicitly masters'
events, academic tournaments are the domain of students,
while a trash tournament's is almost universally a
blend of undergrads, graduate students, and the
elderly.

I guess what I'm getting at is that there really
isn't a "truest" format. There are academic tournaments
that have trash in them; there are trash tournaments;
and while the rules of the two overlap more than the
DNA of a human overlaps with the DNA of a chimpanzee,
different quiz bowl tournament simply test for different
things. Comparing them isn't necessarily a good
idea.

Edmund

This archive was generated by hypermail 2.4.0: Sat 12 Feb 2022 12:30:43 AM EST EST