Cards ink home & home with old foe
NCAA Football Blog Headlines
As promised/threatened the first BCS standings come with a series on conference comparisons.
This episode will explore the criterion used by the BCS. Next will feature OOC record and OOC strength, adjusted to account for biasing related to the central limit theorem. Consideration of records vs BCS teams, teams with winning records and opponents currently in the BCS top 25 will be examined as well. Finally a ranking using breaks in a collection of computer rankings will be used to separate teams into grades and each conference will be judged by their performance against each grade.
This process will be repeated at after the bowls are played and include bowl information and stats related to bowl wins, 10 win teams and data spanning 4 years.
Two of the factors used by the BCS are listed with the BCS standings. The third is harder to pin down because it is vaguely described. Some measure of the top to bottom strength of the conferences using the BCS computers is used to judge them.
Each BCS computer generates a conference standing, though one uses a different set of data to do so. Each presents their data as a raw data score and do not directly relate to each other in any meaningful way. By finding the mean and standard deviation for each computer a z-score can be obtained for each conference relative to that computer.
The average of these 6 z-scores is presented here:
SEC: 1.138
Big 12: 1.127
ACC: 0.804
Big 10: 0.776
Big East: 0.328
PAC 10: 0.277
MWC: 0.122
WAC: -0.605
MAC: -1.127
C-USA: -1.235
Sun Belt: -1.605
At the end of last year similar results were obtained but the WAC was in the lower group and the MWC was in the gap, close to 0. The PAC 10 and C-USA down significantly from previous years. The Big 12, ACC and MWC are up from their typical position.
So while the SEC and Big 12 are neck and neck for the top conference in the land, and the Sun Belt is in no danger of their basement status, the MWC is doing well on a relative basis and the ACC is muddling along without a clear cut leader, but few liabilities either. I don’t expect to see anything different in the other data.
At this time I would like to plug my favorite, based on its theory and openness of methodology, of the BCS comptuer rankings. Unfortunately it also has my team the lowest of any of the computers.
© www.fanblogs.com
Comment on How do the conferences rate for 2008? – BCS Computers…
