[Husker] I guess I have to respectfully disagree

[Andrew Smith]

gscratch1 at comcast.net wrote:

>>I guess it depends what rankings you go by.  I've always hated the whole 
>>"record vs. ranked opponents" stuff because most of those statistics 
>>only consider where the opponent was ranked at the time you beat them.  
>>The only rankings that have any meaning are those at the end of the 
>>season.
>>    
>>
>
>I guess I have to disagree.  It seems to me that the ranking of a team when you play them is more relevant to how your team is than their eventual ranking (or their pre-season ranking, for that matter).  This no-longer-used factor of adjusting your ranking based on your opponents' rankings at the end of the season seemed to ignore how that team might have changed (for better or worse) since you played them.
>  
>
It's a contradiction to advocate "ranking when you played them" over 
eventual ranking or pre-season ranking since (1) for the first game of 
the season, pre-season ranking = "ranking when you played them" and (2) 
"ranking when you played them" is often a mixture of pre-season and 
eventual ranking.

While it is true that teams can change for better or for worse during 
the season (such as due to a key injury), the ratings for teams change 
far more simply because voters and computers get more data.  Ask 
yourself, when do you have a higher confidence in your ability to judge 
a team's ability, after 1 game or after 11?

Finally, the Billingsley computer ranking is one that uses "ranking when 
you played them" and in some cases the results are almost comical due a 
team either utterly failing pre-season expectations or being Cinderella 
team.

>what if, for example, Rutgers had played and beaten Louisville their first game?  what would that have done to the BCS ranking of both?
>
>Glen in NJ
>
> 
>_______________________________________________
>husker site list
>husker at tssi.com
>http://romaine.tssi.com/mailman/listinfo/husker
>
>
>  
>