Properties of sports ranking methods

  title={Properties of sports ranking methods},
  author={Baback Vaziri and Shaunak S. Dabadghao and Yuehwern Yih and Thomas L. Morin},
  journal={Journal of the Operational Research Society},
Ideally, the ranking of sports teams should incorporate information (comprehensiveness) obtained from the outcome of a match, such as the strength of the opponent and schedule. In addition, the ranking method should be fair and not reward teams for poor performance or factors beyond their control, such as the sequence of the matches. We state properties such that if followed, the ranking methods will be fair and comprehensive. We evaluate five popular sports ranking methods and whether or not… 

Learning to Rank Sports Teams on a Graph

To improve the prediction ability of ranking models in sports, a generalized PageRank model is introduced that can achieve better prediction performance than the existing ranking models.

Methods of creating sports rankings based on selected examples

Sports rankings play a very important role in modern sport. They most often determine who will be considered the favorite of a specific match or fight. In other cases, they decide on the seeding of a

The predictive power of popular sports ranking methods in the NFL, NBA, and NHL

Ideally, a ranking method for a sports tournament will be not only fair and comprehensive, but it will also possess strong predictive power. In a recent article, the $$(1, \alpha )$$ ( 1 , α ) method

Some Impossibilities of Ranking in Generalized Tournaments

It is revealed that self-consistency (SC) and order preservation (OP) cannot be satisfied simultaneously on this universal domain.

Coronavirus and sports leagues: obtaining a fair ranking when the season cannot resume

  • L. Csató
  • Economics
    Ima Journal of Management Mathematics
  • 2021
Abstract Many sports leagues are played in a tightly scheduled round-robin format, leaving a limited time window to postpone matches. If the season cannot resume due to an external shock such as the

Coronavirus and sports leagues: how to obtain a fair ranking if the season cannot resume?

This work identifies a set of desired axioms for ranking in these incomplete tournaments and proves that the generalized row sum, a parametric family of scoring rules, satisfies all of them and maximizes the role of the opponents.

A Forward-Looking Approach to Compare Ranking Methods for Sports

In this paper, we provide a simple forward-looking approach to compare rating methods with respect to their stability over time. Given a rating vector of entities involved in the comparison and a

When neither team wants to win: A flaw of recent UEFA qualification rules

It is shown that the root of the problem resides in the incentive incompatibility of certain UEFA qualification rules, and the governing bodies of major sports should choose strategy-proof tournament designs because of this.

Simulation evidence on Herfindahl-Hirschman measures of competitive balance in professional sports leagues

Abstract Measurement of the degree of competitive balance, how evenly teams are matched, is central to the analysis of professional sports leagues. A common problem with within-season competitive



Offense-Defense Approach to Ranking Team Sports

This paper proposes a flexible, easily coded, fast, iterative approach, called the Offense-Defense Model (ODM), to generating team ratings, which is grounded in the theory of matrix balancing.

A Natural Generalization of the Win-Loss Rating System

How are teams in a tournament usually ranked? For most of the popular sports in the U.S., the percentage of wins is all-important, and the team with the highest winning percentage earns first place.

Robust Rankings for College Football

This study investigates the sensitivity of the Colley Matrix rankings to (hypothetical) changes in the outcomes of (actual) games and proposes a robust variant of the rankings based on solving a mixed-integer nonlinear program, which requires about a minute of computation time.

Ranking rankings: an empirical comparison of the predictive power of sports ranking methods

In this paper, we empirically evaluate the predictive power of eight sports ranking methods. For each ranking method, we implement two versions, one using only win-loss data and one utilizing

Ranking participants in generalized tournaments

The fair-bets ranking method proposed by Daniels and Moon and Pullman is characterized using six independent axioms to deal with the problem of ranking social alternatives based on the number of voters.

Paired comparisons analysis: an axiomatic approach to ranking methods

An axiomatic analysis of several ranking methods for general tournaments finds that the ranking method obtained by applying maximum likelihood to the (Zermelo-)Bradley-Terry model is one of the ranking methods that perform best with respect to the set of properties under consideration.

Who's #1? The science of rating and ranking

While, perhaps inevitably given the size of the domain, the book misses quite a large amount of relevant material, it does provide an excellent, accessible, and stimulating discussion of the material it does cover and makes a valuable addition to the canon of rating and ranking.

Additive and multiplicative properties of scoring methods for preference aggregation

It is shown that a natural additive property has strong links to independence of irrelevant matches, an axiom judged unfavourable when players have different opponents.

Sensitivity and Stability of Ranking Vectors

The Markov method is highly sensitive to small changes in the data and the PageRank rating vector is strongly nonuniformly spaced, while the Colley and Massey methods provide a uniformly spaced rating vector, which is more natural for a perfect season.

Ranking the Participants in a Tournament

A tournament is a finite set whose elements are called players, together with a binary relation called beating which is complete and asymmetric. A ranking of the players is an order-relation on the