RIT study uses new mathematical models to create better ranking system for sports teams
Every year, there is endless debate in the press and in living rooms across the country concerning sports rankings, particularly when it comes to deciding who are the best teams in college football.
Mathematicians and students at Rochester Institute of Technology are developing new models aimed at providing better ranking methods with less controversy.
"Current sports rankings such as the Bowl Championship Series take into account many factors, but head to head competition is not emphasized enough," says Darren Narayan, assistant professor of mathematics at RIT and head of the project.
When one team beats another, and the loser is ranked ahead of the winner, it weakens the credibility of the ranking in the eyes of the public. This becomes further apparent when one team soundly defeats another in the final game of the regular season, but finds themselves ranked immediately below the team they just beat.
The most glaring example occurred in 2002 when the University of Colorado was ranked third in the final Bowl Championship Series poll even though they beat the number two team, the University of Nebraska, by the score of 62-26 in the final game of the season. As a result of the final rankings, Colorado was kept out of the national championship game. Ironically a similar scenario happened the year before when Florida State and Miami were ranked second and third respectively, despite the fact that Miami defeated Florida State during the regular season.
The model Narayan has developed incorporates results of head to head competition, considering both the score and at what point of the season the game took place. He is also examining various ranking techniques involving an arbitrarily large number of teams.
Source: Eurekalert & othersLast reviewed: By John M. Grohol, Psy.D. on 21 Feb 2009
Published on PsychCentral.com. All rights reserved.
People ask for criticism, but they only want praise.
-- W. Somerset Maugham