[Husker] Michigan

[RJ Wessel]

If I am reading the page correctly- Michigan dropped from 21 to 47 a drop of 26 positions. 

On one other note - if you look at this really close - the PAC 10 currently has 10 of the top 11 positions on the SOS with Notre Dame sneaking in at number nine. 

Not sure how much weight I would put on this one at this time. Maybe, as stated, as the data fills in it will make more sense. 




---- Pat Gaule <pgaule at cox.net> wrote: 
> Bob Beach wrote:
> 
> >     I am absolutely shocked Michigan dropped as far as they did.  This poll ranks them 47th after ranking them 5th the week before.  I have never ever seen a team drop over 40 places no matter who they lost to.  I am not saying Michigan shouldn't have dropped big time but even dropping 15 spots is dropping big time.  This indeed was a big time lost but dropping out of the top 25 from number five?  Wow!
> >
> >
> >http://www.collegefootballpoll.com/current_congrove_rankings.html
> >
> >
> >  
> >
> I'm not surprised to see Michigan drop 40 places in a computer ranking.  
> Regardless of the formula, computer rankings are not very accurate early 
> in the season.  A few systems, such as Sagarin, get around this by 
> placing some "bias" into the rankings for the first few weeks.  These 
> rankings are really only designed for one thing: to be accurate at the 
> end of the season.  It doesn't mean much to look at them after only a 
> handful of games have been played.
> 
> This particular ranking is not included in the BCS, so it doesn't really 
> mean much.  However, Michigan is ranked 40th in the Sagarin 
> (http://www.usatoday.com/sports/sagarin/fbt07.htm).  I am sure this will 
> change as the season progresses and more games are played.  Computer 
> rankings are only as accurate as the data fed into them, and there isn't 
> much data right now.  These rankings will normalize over the season.
> 
> _______________________________________________
> husker site list
> husker at tssi.com
> http://romaine.tssi.com/mailman/listinfo/husker