Competitve Balance in the NCAA FBS

Yesterday, this article was posted positing that NCAA FBS is broken because the same top teams are in contention for the national championship game year after year.  Yet in terms of analysis the article is severely lacking. 

So, here is one take on this idea, and that is with regard to the overall level of competitive balance in all of the NCAA FBS.  I am using the Noll-Scully Competitive Balance measure.  The Noll-Scully Competitive Balance calculates the actual standard deviation of winning percent to an idealized standard deviation of winning percent, where the idealized standard deviation is measured as if wins and losses are randomly determined.  The closer the Noll-Scully Competitive Balance measure is to one, the greater the league's competitive balance.  For those that are interested in calculating this on their own, here is a step-by-step guide as to how to perform the Noll-Scully competitive balance calculation.

For a viewpoint of more than one season, I have looked at the data since 1996 (the season that overtime was instituted - and ties no longer take place).  This allows for the win/loss competitive balance measure to be used and compared from season to season.  As you can see in the chart below, competitive balance has remained rather stable since 1996, calling into question the hypothesis that college football is broke.