To look at the Mid-Atlantic that is at issue here... based on our ranking system, Penn's stats happened to have made them the 7th team earning an invitation after all of the automatics had been awarded. Maryland placed 12 spots lower (statistically, following our method), despite being one game better in the won-loss column. (Did either team ever play a game that could have been decided by single tossup going the other way? Then that was the only difference between the teams in wins and losses; Penn however scored 1015 more points over the course of the full 13 game schedule.) Now if we had a rule saying that invitations must follow order of finish, and therefore Maryland must be invited before Penn, what do we do vis a vis the eleven teams from other tournaments ranking between the two in our statistical rankings? You agree that statistical comparisons are necessary to compare across regions. Should we bump up Maryland ahead of eleven teams that rank higher by statistical comparisons to place them immediately ahead of Penn, which _does_ rank higher by statistical comparisons than those eleven teams? Or do we drop Penn 13 spots in the rankings so that they remain behind Maryland, the all-important consideration? You see the problem. I know teams get angry when a team finishing below them is ranked higher, but I hope you can honestly see the reasons that that happens under the comparative system we have chosen. We do not do the complex statistical comparisons and careful adjustments for strength of schedule out of a love of complexity, but because we are trying to be absolutely as fair as possible in comparing all teams SCT performances, and in reducing the effects of teams playing in tournaments of differing field strength. This is nothing new; teams know going into NAQT SCTs that--unless they are title winners--what is going to be crucial to rankings for ICT invitation purposes is statistical comparisons of teams' point-scoring across entire tournaments. Eric H., for NAQT
This archive was generated by hypermail 2.4.0: Sat 12 Feb 2022 12:30:43 AM EST EST