Ranking Methods for Olympic Sports: a Case Study by the U.s. Olympic Committee and the College of Charleston

This paper develops a method to determine if the United States’ Olympic sports program is improving on an individual sport basis. In addition, the probabilistic likelihood of a U.S. Olympic sports team reaching the medal podium was investigated. The research presented in this paper focuses on U.S. Men’s Ice Hockey (a head-to-head sport), with additional research done on Women’s Alpine Skiing (a multi-competitor sport). Three ranking methodologies were employed for men’s ice hockey: Massey, Elo, and Microsoft’s Trueskill. Within these methodologies, information accumulated from Olympic tournaments and world championships (2004 to 2013) were applied. The subsequent results allowed a definite determination in regards to improvement of the U.S. team in respect to other competitors. A Monte Carlo simulation was run for ice hockey in the format of the 2014 Olympic Games. Comparison of a known distribution (normal and uniform) to the generated ratings resulted in a predictive measure for the U.S. to medal. Overall, it was found that ranking methodologies employed throughout this research were considerably accurate when compared to actual results of the 2014 Games.