In-season prediction of batting averages: A field test of empirical Bayes and Bayes methodologies

Batting average is one of the principle performance measures for an individual baseball player. It is natural to statistically model this as a binomial-variable proportion, with a given (observed) number of qualifying attempts (called "at-bats"), an observed number of successes ("hits") distributed according to the binomial distribution, and with a true (but unknown) value of p i that represents the player's latent ability. This is a common data structure in many statistical applications: and so the methodological study here has implications for such a range of applications. We look at batting records for each Major League player over the course of a single season (2005). The primary focus is on using only the batting records from an earlier part of the season (e.g., the first 3 months) in order to estimate the batter's latent ability. P i , and consequently. also to predict their batting-average performance for the remainder of the season. Since we are using a season that has already concluded, we can then validate our estimation performance by comparing the estimated values to the actual values for the remainder of the season. The prediction methods to be investigated are motivated from empirical Bayes and hierarchical Bayes interpretations. A newly proposed nonparametric empirical Bayes procedure performs particularly well in the basic analysis of the full data set, though less well with analyses involving more homogeneous subsets of the data. In those more homogeneous situations better performance is obtained from appropriate versions of more familiar methods. In all situations the poorest performing choice is the naive predictor which directly uses the current average to predict the future average. One feature of all the statistical methodologies here is the preliminary use of a new form of variance stabilizing transformation in order to transform the binomial data problem into a somewhat more familiar structure involving (approximately) Normal random variables with known variances. This transformation technique is also used in the construction of a new empirical validation test of the binomial model assumption that is the conceptual basis for all our analyses.

[1]  M. S. Bartlett,et al.  The square root transformations in the analysis of variance , 1936 .

[2]  Bartlett Ms The use of transformations. , 1947 .

[3]  M. Bartlett,et al.  The use of transformations. , 1947, Biometrics.

[4]  F. J. Anscombe,et al.  THE TRANSFORMATION OF POISSON, BINOMIAL AND NEGATIVE-BINOMIAL DATA , 1948 .

[5]  J. Tukey,et al.  Transformations Related to the Angular and the Square Root , 1950 .

[6]  H. Robbins An Empirical Bayes Approach to Statistics , 1956 .

[7]  Frederick Mosteller,et al.  Tables of the Freeman-Tukey Transformations for the Binomial and Poisson Distributions , 1961 .

[8]  C. Stein Confidence Sets for the Mean of a Multivariate Normal Distribution , 1962 .

[9]  L. Brown Admissible Estimators, Recurrent Diffusions, and Insoluble Boundary Value Problems , 1971 .

[10]  W. Strawderman Proper Bayes Minimax Estimators of the Multivariate Normal Mean , 1971 .

[11]  William E. Strawderman,et al.  Proper Bayes Minimax Estimators of the Multivariate Normal Mean Vector for the Case of Common Unknown Variances , 1973 .

[12]  B. Efron,et al.  Data Analysis Using Stein's Estimator and its Generalizations , 1975 .

[13]  Lawrence D. Brown,et al.  Estimation with Incompletely Specified Loss Functions (the Case of Several Location Parameters) , 1975 .

[14]  B. Efron,et al.  Stein's Paradox in Statistics , 1977 .

[15]  C. Stein Estimation of the Mean of a Multivariate Normal Distribution , 1981 .

[16]  H. Robbins Asymptotically Subminimax Solutions of Compound Statistical Decision Problems , 1985 .

[17]  A. Tversky,et al.  The hot hand in basketball: On the misperception of random sequences , 1985, Cognitive Psychology.

[18]  J. Berger Statistical Decision Theory and Bayesian Analysis , 1988 .

[19]  C. Stein,et al.  Estimation with Quadratic Loss , 1992 .

[20]  S. Christian Albright,et al.  A Statistical Analysis of Hitting Streaks in Baseball , 1993 .

[21]  Y. Benjamini,et al.  Controlling the false discovery rate: a practical and powerful approach to multiple testing , 1995 .

[22]  Cun-Hui Zhang,et al.  EMPIRICAL BAYES AND COMPOUND ESTIMATION OF NORMAL MEANS , 1997 .

[23]  A new test for the Poisson distribution , 2001 .

[24]  C. A. Mead,et al.  Curve Ball: Baseball, Statistics, and the Role of Chance in the Game , 2001 .

[25]  Jay M. Bennett,et al.  Curve Ball: Baseball, Statistics, and the Role of Chance in the Game , 2001 .

[26]  B. Efron Robbins, Empirical Bayes, And Microarrays , 2001 .

[27]  M. Lewis Moneyball: The Art of Winning an Unfair Game , 2003 .

[28]  A Unified View of Regression, Shrinkage, Empirical Bayes, Hierarchical Bayes, and Random Effects , 2007 .

[29]  Is an .833 Hitter Better Than a .338 Hitter? , 2007 .

[30]  Jim Albert Streaky Hitting in Baseball , 2008 .

[31]  Lawrence D. Brown,et al.  NONPARAMETRIC EMPIRICAL BAYES AND COMPOUND DECISION APPROACHES TO ESTIMATION OF A HIGH-DIMENSIONAL VECTOR OF NORMAL MEANS , 2009, 0908.1712.

[32]  ESTIMATORS FOR GAUSSIAN MODELS HAVING A BLOCK-WISE STRUCTURE , 2009 .

[33]  Harrison H. Zhou,et al.  The root–unroot algorithm for density estimation as implemented via wavelet block thresholding , 2010 .