DOI QR코드

DOI QR Code

Efficiency of pairwise winning percentage estimators in Korean professional baseball

한국프로야구에서 쌍별 승률추정량의 효율성

  • Received : 2017.02.10
  • Accepted : 2017.03.13
  • Published : 2017.03.31

Abstract

In baseball, estimation of winning percentage is critical and many studies for this topic have been actively performed. Pairwise winning percentage estimation using Pythagorean winning percentages of individual teams against other individual teams has the property that the sum of estimated winning percentage totals must be a constant. In this paper, we consider two types of pairwise estimation including linear formula and Pythagorean formula to the Korean baseball data of seasons from 2013 to 2016 under the criterions of RMSE and MAD. In conclusion, pairwise Pythagorean methods have the smaller RMSE and MAD than traditional Pythagorean methods. We suggest the optimal pairwise Pythagorean formula with a fixed exponent. Also we show that there are very little differences of RMSE and MAD between variation in exponent values.

야구에서 승률 추정은 매우 중요한 문제이며 현재 이 분야에 대한 연구가 활발하게 진행되고 있다. 쌍별 승률추정은 팀대 팀의 경기결과를 이용하여 전체 승률을 추정하는 방법으로써 각 팀들의 추정된 승률의 합이 상수가 된다는 타당성을 가진다. 본 연구에서는 한국프로야구에서 피타고라스 승률과 선형 승률에 쌍별 추정을 적용하고 효율성을 RMSE와 MAD를 이용하여 살펴보았다. 사용된 데이터는 2013년부터 2016년 사이의 모든 한국프로야구 팀대 팀 기록이며, 그 결과 쌍별 피타고라스 추정이 기존의 방법들보다 RMSE와 MAD 측면에서 바람직하다고 간주되었다. 또한 쌍별 피타고라스 추정에 사용되는 바람직한 지수 값의 결정에 대하여 설명하였으며 추정에 사용된 지수 값의 변화에 따른 RMSE와 MAD의 차이는 크지 않음을 알 수 있었다.

Keywords

References

  1. Cochran, J. J. (2008). The optimal value and potential alternatives of Bill James Pythagorean method of baseball. STAtOR, 2 , 2008.
  2. Davenport, C. and Woolner, K. (1999). Revisiting the Pythagorean theorem: Putting Bill James' Pythagorean theorem to the test. The Baseball Prospectus, http://www.baseballprospectus.com/article.php?articleid=342.
  3. Heumann, J. (2016). An improvement to the baseball statistic "Pythagorean Wins". Journal of Sports Analytics, 2, 49-59. https://doi.org/10.3233/JSA-160018
  4. James, B. (1980). The Bill James abstract, self-published, Lawrence, KS.
  5. Jones, M. and Tappin, L. (2005). The Pythagorean theorem of baseball and alternative models. The UMAP Journal, 26 .
  6. Kim, S. K. and Lee, Y. H. (2016). The estimation of winning rate in Korean professional baseball. Journal of the Korean Data & Information Science Society, 27, 653-661. https://doi.org/10.7465/jkdi.2016.27.3.653
  7. Lee, J. T. (2014). Estimation of exponent value for Pythagorean method in Korean pro-baseball.Journal of the Korean Data & Information Science Society, 25, 493-499. https://doi.org/10.7465/jkdi.2014.25.3.493
  8. Lee, J. T. (2015). Measuring the accuracy of the Pythagorean theorem in Korean pro-baseball. Journal of the Korean Data & Information Science Society, 26, 653-659. https://doi.org/10.7465/jkdi.2015.26.3.653
  9. Lee, J. T. (2016a). Convergence characteristics of Pythagorean winning percentage in baseball. Journal of the Korean Data & Information Science Society, 27, 1477-1485. https://doi.org/10.7465/jkdi.2016.27.6.1477
  10. Lee, J. T. (2016b). A comparison of formulas to predict a team's winning percentage in Korean pro-baseball. Journal of the Korean Data & Information Science Society, 27, 1585-1592. https://doi.org/10.7465/jkdi.2016.27.6.1585