Faculty Advisor

Morales, Carlos J.


In this project we investigate the performance of a variation of the so-called Bradley-Terry (BT) model for paired comparisons in the context of the 2004 Athens Soccer Olympics and the English League. We explore the usefulness of the BT model both as a tool to rank sport teams and as a prediction tool. We find that for the English League, the BT performs adequately in terms of ranking, but it performs poorly for the Olympic since data is too sparse.


Worcester Polytechnic Institute

Date Accepted

January 2004

Project Type

Interactive Qualifying Project


Restricted-WPI community only

Advisor Department

Mathematical Sciences