Unfortunately, I know of no way to balance for geography short of finding the distance between every two sets of schools and developing some factor from that. Some geographic regions are obvious, others are not so immediately apparent. For example, is Missouri-Rolla a Southwest or Midwest team? Is Wichita State Southwest or Midwest? Washington-St. Louis? Is Arizona State Southwest or Pacific? I don't even won't to think about the nightmare of classifying the east coast, where you can throw a rock from one school's campus and land it in another's. I am open for suggestions as to how to quantify a geographic factor. A question arose about a school being adjusted down for entering many teams in a tournament. This is not the case. An example should suffice. A tournament has 20 teams and BSU A finishs 3rd. They receive 6.667 power points. If there is a BSU B and they finish dead last, BSU's adjusted rank is 1/(1/3+1/20) = 2.609. They then would receive 7.667 power points instead. If a BSU C finishes 19th. BSU's adjusted rank is then 1/(1/3+1/19+1/20) = 2.294, and BSU receives 8.719 power points. Regardless of where the school's team finishes, the more teams that enter a tournament, the more power points everyone in the tournament receives. This issue hasn't been raised yet, so I'll raise. Why were those tournaments the only ones chosen, and not Penn Bowl, etc.? These tournaments were selected simply because for those tournaments the question sets were standardized, not necessarily in difficulty but in the fact that the participant teams heard the same questions despite not being in the same location (except the Nationals, where they were at the same place). This eliminates the need for a factor for question difficulty. Another issue is imcomplete reporting of results. The ACF Regional held in Chicago posted only the top four finishes. Teams who finished fifth or below received nothing in the power rankings, and those who finished in the top 4 received 4/(# of teams there) of the points they could have earned. Similarly, those who finished fifth or below at all CB regionals received nothing for their efforts for the same reason. ATTENTION: This is NOT a rant about incomplete results. This is strictly to point out a limitation of this Power Ranking Scheme. Most responses thus far have been quite positive, so I'm guessing it is somewhat close to the real situation. I'll publish a new Top 40 after ACF Nationals and a final one after CB Nationals. Next year, if I have time, I intend to use tournaments with two or more mirrors in the ranking scheme because of the question standardization. I do not plan to carry over results from one year to the next. There will be no pre-season rankings. The first one for Fall 2000 will be after either NAQT's fall tournament or after the first doubly-mirrored tournament. Dan Beshear
This archive was generated by hypermail 2.4.0: Sat 12 Feb 2022 12:30:42 AM EST EST