The Colley and Massey methods are two mathematical algorithms that use linear algebra to rank sports teams. Both use the network of interconnections formed by the match-ups between teams that occur over the course of a season to produce a full ranking of all the teams, even though each team plays only a subset of the other teams in the nation. In essence, the Colley method uses strength of schedule to produce its ranking; the Massey method employs strength of schedule together with margin of victory. All 8050 games between Division I opponents occurring in the 2018 season, from February 16 through June 4, including the conference tournaments and the regionals, were used to produce these rankings. We use uniform weighting: each game is treated equally.
The Colley ranking was one of the six objective methods used to help rank NCAA FBS college football teams between 1998 and 2013 under the BCS ranking system. Another one of these methods was also designed by Kenneth Massey, who invented the eponymous method used here, but the Massey algorithm employed in college football is not the same as the method we use.
You can experiment with these ranking methods applied to international soccer at a companion website. On that site you may specify the importance of a number factors for soccer matches, and we produce a customized ranking of the teams playing in the FIFA World Cup.
You can contrast these rankings with a number of NCAA Division I Baseball Rankings available from the NCAA, including RPI and poll-based
methods, and with the Massey Ratings for NCAA Division I Baseball.
Rankings for Teams in the Super Regionals
Colley Ranking |
Massey Ranking |
|
|
Rankings for All Division I Teams
Colley Ranking |
Massey Ranking |
|
|
References
- Boyd’s World, database for scores of college baseball games.
- Who’s #1?: The Science of Rating and Ranking, Amy N. Langville and Carl D. Meyer, Princeton University Press, 2012.