nfl power rankings: perception vs. statistics...

dr. zzzzzz.  http://sportsillustrated....18/ranks.part1/index.html
peter sofa king  http://sportsillustrated....er_king/10/16/mmqb/1.html
espn conglom  http://sports.espn.go.com...ng?season=2006&week=7
pete crisco  http://www.sportsline.com/nfl/powerrankings
peter schlonger  http://msn.foxsports.com/nfl/powerRankings


football outsiders http://www.footballoutsid...lings/dvoa-rankings/4409/


interesting.

i think there's SOME value in both.

but i think #'s don't really lie... it's just how accurately they are calculated, and whether or not we INTERPRET them correctly.


i'm not gonna argue, for example the steelers are #1.  but I do think their defense is certainly top 10 if not top 5.   you wouldn't expect the #17/18 team in a power rankings list to have a top 5 defense, for example.


as for the colts,  they're 5-0, and rightly so. but they have yet to play a team with a great run offense, and a decent passing defense.   the #'s say they will lose when they do so.


the bears lost about 20% of their dvoa after the cards game... and while still a good team, the #'s brought them back somewhat to the back, and indicate sd  could well beat them... which one would certainly think if having watched both play.


the #'s for DVOA are most interesting in the transition to the 0% mark (and below)...   notice the cluster of teams from 10-0%,  but none from 0- -10 %.   the gap between the middle tier and bottom tier teams is huge... as is the gap from the tier to top 3 or 4  to the middle... which show  GREAT parity.... as one might expect.