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

Barrow, Daniel L.; Drayer, Ian; Elliott, Peter; Gaut, Garren; Osting, Braxton

doi:10.1515/jqas-2013-0013

Cited by 34 publications

(27 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To compare the predictive power of the methods in section 2, we take a similar approach to that of D. Barrow et al in their analysis of sports ranking methods [6]. First, we perform cross validation on each method for all competition categories.…”

Section: Analysis Of Ranking Methodsmentioning

confidence: 99%

“…We find that the older USAC methods satisfy the desirable independence of irrelevant alternatives property which all of the rank aggregation methods fail to satisfy. In our analysis of the predictive power of these methods when they are used to rank climbers, we follow the steps laid out in [6]. First, we perform leave-one-out cross validation on the methods.…”

Section: Path To the Top Start Holds Finish Holdmentioning

confidence: 99%

“…We applied a ranking method φ to the training set and then calculated the prediction error, E(t), for the test set t, which Barrow et al [6] defines to be…”

Section: Cross Validationmentioning

confidence: 99%

See 2 more Smart Citations

Rank and Score Aggregation Methods in Competitive Climbing

Parker¹

2018

SIURO

View full text Add to dashboard Cite

Abstract. Ranking methods are used in all aspects of life, from Google searches to sports tournaments. Because all ranking methods necessarily have advantages and disadvantages, USA Climbing, the organizer of national climbing competitions in the United States, changed their ranking method three times between 2009 and 2016. The combined rank method employed in 2015 marked a drastic step away from the previous two in that it failed to meet the independence of irrelevant alternatives (IIA) criterion and was almost impossible for spectators to use to calculate ranks on their own. We compare this more recent rank aggregation method with older USA Climbing score aggregation methods as well as other methods from the literature. Three particularly important methods we consider are (i) the combined rank method, (ii) a combination of the previous two USA Climbing score aggregation methods (the merged method), and (iii) a linear programming (LP)-based rank aggregation method from the literature. Using data from the 2016 Bouldering Youth National Championships, we perform leave-one-out cross validation and the Friedman hypothesis test to conclude that at the 99% confidence level, the LP-based rank aggregation method has significantly more predictive power than the other two methods, while there was insufficient evidence to distinguish between the predictive power of the combined rank method and the merged method. However, due to the desirable properties, such as the IIA criterion, satisfied by the merged method, we recommend this method for use in competitive climbing.

show abstract

Section: Analysis Of Ranking Methodsmentioning

confidence: 99%

Section: Path To the Top Start Holds Finish Holdmentioning

confidence: 99%

See 1 more Smart Citation

Rank and Score Aggregation Methods in Competitive Climbing

Parker¹

2018

SIURO

View full text Add to dashboard Cite

show abstract

“…Although wrestling awards several victory types, matches result in binary win/loss outcomes. Barrow et al (2013) expand upon previous work evaluating predictive accuracy in sports rankings, assessing eight approaches over 4 datasets (NBA, MLB, Division I men's basketball, Division I football) in their paper. While the authors do not find any "best" method, they conclude that systems incorporating point differentials are usually more predictive than those only using win-loss data, which informs our approach.…”

Section: Introductionmentioning

confidence: 95%

Ranking and prediction of collegiate wrestling

Bigsby

Ohlmann

2017

JSA

View full text Add to dashboard Cite

Abstract. College wrestling rankings have a major impact upon the sport, as they are one of the criteria used to determine advancement to postseason championships and seeding, thereby benchmarking individual, team, and program success. In addition to ordering athletes based upon past performance, rankings may function as predictors for future match outcomes. This research identifies potential biases in traditional ranking methodologies and offers alternatives based on the networkbased PageRank and Elo ratings. Using data from the 2013-2014 NCAA Division I wrestling season, we evaluate both existing and our new rankings on the basis of predictive accuracy. Our new methods significantly improve upon a baseline of 67% accuracy and several of the existing ranking methods. These results and follow-up analyses suggest that Elo presents an especially attractive alternative to current ranking systems for college wrestling, with potential extensions to the prediction of point differentials, ensemble methods, and generalizability to other combat sports.

show abstract

“…Next, we give a short description of the ranking methods we will use. Hundreds of ranking methods have been appeared in the long history of ranking in sports: for a more detailed introduction on ranking methods, we refer to [12] and [111]. The selection of the methods we used satisfy the following criterion: (1) each method is based on linear algebra, (2) each method has been proved to be successful in real applications, and (3) each method has a simple formulation with, in most cases, a closed solution.…”

Section: Some Linear Algebraic Rating Methodsmentioning

confidence: 99%

Complex network models, graph mining and information extraction from real-world systems

London

View full text Add to dashboard Cite

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

Cited by 34 publications

References 5 publications

Rank and Score Aggregation Methods in Competitive Climbing

Rank and Score Aggregation Methods in Competitive Climbing

Ranking and prediction of collegiate wrestling

Complex network models, graph mining and information extraction from real-world systems

Contact Info

Product

Resources

About