In response to the original Power Ranking scheme, Jerry Hagen of U Illinois-UC raised these two issues (paraphrased, not direct quotes): #1. This system unfairly penalizes team for decisions not to play based on ideology or practicallity. Indiana did win CB Region 9. Since CB posted the results of only 4 teams per Region, that all I could assume had played, even though the records posted clearly indicated that more played. Indiana only received 4 power points for that win (4 team/1 rank). Same with all other CB Regionals winners. Since Wisconsin is in that group, that may also explain their lower than actual ranking. Illinois was not severely penalized from not playing CB. As for practicallity, I understand that one ALL TOO WELL. Oklahoma is nowhere near any Nationals this year, and we are penalized in this scheme for it. I know that we are not the only ones being penalized for it. I think our Div II squad would have placed in the middle of the field at NAQT ICT if we could have brought them. Supposedly NAQT Nationals and CB Nationals next year will be in different parts of the country, and this should alleviate some of this problem. Teams do get better with the number of tournaments in which they participate. This scheme is purposely designed to award those team who play often, because they will be the better team soon. Good teams that do not participate will not be good for very long. There is another factor at work. A team that can raise more money than its rivals and budget itself well will play in more tournaments and eventually be a better team. Both Texas and Texas A&M were rewarded for getting the money to bring two teams, where Oklahoma was not. #2. Teams are penalized for hosting CCTs and SCTs because of ranking by aggregate points rather than averages. A good team would be penalized for running a good tournament if the best players or all players were involved in the operation of the tournament. Samer Ismail of Penn and Ahmed Ismail of MIT both thought Yale's ranking was too low. Both indicated that Yale had defeated Penn several times this year in other tournaments and thus should be higher ranked. Quoting the original missive: >19.747002817 22 Penn >19.744444486 23 Georgia >19.731818202 24 Yale This was a problem. I adjusted the scheme to award the host an automatic 4th place for their efforts. I'll play with it further, maybe I can work this kink completely out of it. It did raise Yale from 24th (original) to 19th, and closed the gap between Princeton and Illinois from almost 5 point to less than 1. David Goodman of Michigan and Emil Thomas Chuck of Case Western both pointed out that Illinois, Chicago, and Michigan data were skewed because they always played each other and that the scheme does not account for team strength across regions. Goodman further points out that at NAQT ICT, "the Midwest 3 feasted on the rest of the field. Chicago only lost to Illinois and Michigan. Michigan was 4-1 against Harvard, UC Berkeley, and Princeton. Illinois lost to UC Berkeley but otherwise only lost once to Chicago." "Harvard's only games against any of the three this year were at NAQT Nats and they lost all of them." At the NAQT ICT, Harvard received 10.064 from their Div I and 26.000 from Div II, for a total of 36.064. Chicago received 20.500 from Div I and 4.333 from Div II, for a total of 24.833. Michigan received 14.991 from Div I and 2.364 from Div II, for a total of 17.355. Princeton received 5.857 from Div I and 8.667 from Div II, for a total of 14.524. Illinois received 41.000 from Div I (no Div II). Harvard's Div II win buoys them above Chicago and Michigan, but on the same note, Illinois received 80% of their points from their Div I win.
This archive was generated by hypermail 2.4.0: Sat 12 Feb 2022 12:30:42 AM EST EST