What type of research and data are your conclusions based on?
I tried to set up an "objective" index. Some people base rankings on what teams have "earned" and others on "expectations". This index is based on a simple idea: how many games would Team A lose if it played all 346 other teams? The fewer the number of projected losses, the higher the ranking. Every game is simulated using kenpom's efficiency numbers (but I use a slightly different method of determining the "pace" between two teams for each game) and a score is calculated for every 346 possible games for team A. This is repeated for every team.
For example, as of right now, Texas is projected to lose to 0 other teams. Duke is projected to lose to only 1 team (Texas 78 Duke 75) and Kansas is projected to lose to 2 teams (Texas and Duke)...Kansas St. would lose to 13 teams, which is good enough to be in 15th place.
FYI, UNC would lose to 37 teams (which would put them at #41).
Of course as the year goes on this index will (hopefully) converge upon the most accurate rankings.
