As March Madness is upon us there's
a lot of talk in reference to power rankings that tend to be very
confusing. Ultimately the tournament
itself decides who becomes the Champions, but the question every year is which
teams deserve to be there?
There are always controversies
involved in this process especially with the seedings and when a
bubble team didn't quite make it in for whatever reasons the NCAA tournament
Due to its simplicity, the NCAA
tournament committee uses Rating Percentage Index (RPI) as the biggest decision
aid in deciding which teams to invite to "March Madness" and where to
seed them even though it doesn't account for actual scores of games.
Besides the RPI, other power
rankings exist; the Sagarin, Massey and Ken Pomeroy’s. All of these methods
including the RPI are based upon the outcomes of games, locations and the
quality of their opponents. Each one puts these components together in slightly
different ways and will arrive at slightly different results.
In recent years the ESPN analytics
department created a method known as College Basketball Power Index or BPI for
short. They believe that this method is a little more refined than any other
existing power ranking. This method has a number of small details involved in
their methodology to make it a more accurate reflection for an NCAA tournament
team but many readers will not be interested so I will stay away from the
technical aspects of it and stick with the simplicity of their method.
On top of the technical details
ESPN’s analytics department decided to incorporate a little more information
than other power ranking systems being used. In particular, they added a way of
accounting for missing players. If a team or its opponent is missing one of its
most important players (determined by minutes per game) for a contest, that
game is less important for ranking the teams compared to games in which both
teams are at full strength.
Dean Oliver, ESPN’s director of
production, strongly believes that the
College Basketball Power Index (BPI) gives them a tool for rating teams that is
more useful than existing tools. It explains a team's wins and losses in a more
detailed context and seemingly predicts future results as well as other tools,
if not better. It also incorporates information about injuries and missed games
that should be relevant for weighting the résumés of bubble teams on Selection
Sunday. He is confident that it will enhance coverage of college basketball for
years to come.
In my opinion I like the BPI method,
but please do not confuse that with me believing that this is a strong tool to
predict the winner of the NCAA title. I do believe strongly that if teams were
missing key players in their lineups during a loss then it should be factored
in with all rating systems for regular and post season action.
including myself believe the current RPI method lacks theoretical
justification from a statistical standpoint. Other ranking systems which
include the margin of victory of games played or other statistics in addition
to the win/loss results have been shown to be a better predictor of the
outcomes of future games. However, because the margin of victory has been
manipulated in the past by teams or individuals in the context of gambling, the
RPI lessens the chances of such manipulation.
The downside to the RPI method is
with so much emphasis placed on Strength of Schedule (SOS) this strongly favors
and gives an unfair advantage to teams from major conferences. Very rarely do
you have multiple teams coming out of the same mid-major conference.
capturing blowouts, but not overweighting them, BPI credits the ability
of good teams to easily beat poor teams without providing incentive to
win by 30 when 20 is a safe margin. By capturing both blowouts and close
games in this way, BPI summarizes a team's résumé for the NCAA
Please follow Herman on twitter @babethomherman