Re: A Word about...

<<Thank you for the rhetorical statement -
the main purpose of my statistical analysis was that
there ARE certain methods which IMO are better
indicators than the formula NAQT used this
year.>>

<<While I use head-to-head matchups and Samer (and NAQT)
uses total points scored, each is correct in their
opinion. However, I believe my logic has a more correct
fundamental basis, and thus I continue to post and present
arguments in our defense - not for this year's results, but
for future teams who may be slotted in the same
position.>>

Just as you are entitled to your opinion, I am
entitled to mine. In a sense, it has a ring of holism v.
reductionism: the method you advocate looks at individual
parts, while the method NAQT chooses is more concerned
with the overall whole.

Personally, I think the
best approach would be a 'two-pronged' system. As an
example, let us assume that there is an agreed-upon and
reasonable method for comparing W-L performance among
regionals, as well as a method of comparing teams by points.
There is no reason you couldn't then alternate teams
off the lists: take the top team by record, then the
top remaining team by points; the highest remaining
team by record, and so on until all the spaces are
filled.


--STI

P.S. One method of doing this (using as the 'record'
method [W/L record] x [field strength]) gave the
following top 14 at-large bids: Michigan, Wisconsin,
Caltech, Pittsburgh, Virginia, Florida, Texas A&M, GWU,
Maryland, Penn State, Georgia, Yale, Penn, and Illinois.
The next four qualifiers would be, in order, FAU,
MIT, WUSTL, and Oklahoma. An added advantage of this
system is that it would make it significantly harder for
a team to 'leapfrog' over a much higher-ranked team
from the same sectional, unless its performance were
markedly better: Princeton B would be eighth on the
waitlist, behind every team ahead of it at the MidAtlantic
SCT. [In fact, the only teams that would 'leapfrog'
ahead of teams with better records would be Penn, Yale,
and MIT.]

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