The BCS

Without getting too deeply into how the BCS rankings do what they do, it’s enough to know that the system puts two major chunks of information into its numbers grinder.  The Harris Interactive poll and the USA Today poll form one chunk.  Six computer rankers -- Anderson & Hester, Richard Billingsley, Colley Matrix, Kenneth Massey, Jeff Sagarin, and Peter Wolfe – form the other.  After considerable dividing, multiplying, and averaging, all of that input produces a result.

Reasonable people can haggle over the formula -- this link has a pretty good explanation of how it's all supposed to work...

http://www.ehow.com/how-does_4564626_the-bcs-work.html 

...but the point is, both components, but especially the computer rankers, take into consideration, when they look at the universally sparkling won-lost records of the teams in contention, who they won over.  And, of course, it’s the answer to that question that tells us why the Gorillas of Podunk State College are not considered the best team in the country, despite their 12-0 record. 

It’s not just a little about strength of schedule.  It’s all about strength of schedule.  The people who respond to the two polls in the formula presumably at least glance at strength-of-schedule in the course of their ruminations.  But the computer rankers focus on it.  They do so because it matters greatly to the accuracy of the result, and they do so because they can.  That’s what computers bring to this undertaking. 

People who think it can be determined which among America’s college football teams is the best in any given year are on a fool’s errand, in part because there are just too many variables.  College football is not like, say, Major League Baseball where the teams are of more or less equal stature, making the win-loss record over 162 games a not unreasonable indicator of best-team status. Obviously, no such indicator applies in college football  -- there will be any number of undefeated or close to undefeated teams in any year – and even among those teams which, by general acclaim, are thought to be high on the list, it is impossible to say which of them is indisputably the “best.”

But that doesn’t stop some folks from trying.

But the BCS rankings come closest, and here’s why:  They do a pretty thorough job of dealing with the strength-of-schedule question.  And in this calculation, that question is not merely one piece of input.  In this calculation, that question is everything.