A DePauw player wrote: >As far as the ranking thing goes, I agree with the principle, but >think everything can be remedied (and may very well be, if NAQT >makes as much sense as I think they do): >In theory, NAQT adjusts its proverbial "S-Value" of teams based on >their competition. In order to do this, it seems to me they would >have to painstakingly go game by game, if they take the entirety of >all the sectionals as a group. >However, even if they take the groups sectional by sectional, then >the more times you play Emory or Florida, the better off you are, in >theory. To be totally frank, we here at DePauw are clinging to this >hope, in the absence of "wild card" bids. In theory, going 1-1 >against Iowa and losing one match to Arkansas by only 10 points >should work in our favor (I hope). I of course am not a member of NAQT and am not privy to their formula; nevertheless, the above appears to entirely misconstrue the way their ranking system works (as I understand it). NAQT would not have to go game by game at all--they simply would adjust one's S- value based on strength of schedule by using the S-value of the teams one played--think in terms of Sagarin rankings if that helps. To be more exact--probably the only thing actually considered would be the average S-value of teams one played and that would bear on certain variables within your own S-value. However, insofar as I can tell, the largest single component of a team's S-value is its bonus conversion (which is as it should be--that is the only statistic that can fairly be compared across regionals and across varying levels of competition--all other factors, even power tossups per tossup heard are variable depending upon the strength of one's competition). In other words, almost never would it matter if you played all crappy teams or all Michigan A's. DePauw is probably better off relying on the simple fact that there appeared to be a relative paucity of Div. I teams this year and there may well be spots for the asking in LA. my 3 cents, nathan freeburg
This archive was generated by hypermail 2.4.0: Sat 12 Feb 2022 12:30:46 AM EST EST