Search

There are no published data on the reliability of competitive strength performance, even though it has a pronounced effect on athletes' medal prospects and the ability of coaches and athletes to identify factors that affect competitive performance. The purpose of this investigation was to analyze the reliability of elite olympic weightlifters over a series of international weightlifting meets. We obtained official results of international competitions over an 18-month period from 1999 until the 2000 Olympic Games at the International Weightlifting Federation website. The measure of reliability was the typical within-athlete variation, derived as the coefficient of variation (CV) by 2-way analysis of variance of log-transformed weights lifted for the snatch, clean and jerk, and total. The results of this analysis were (a) within-athlete variations were as follows: snatch, 2.7% (95% likely range, 2.4-3.2%); clean and jerk, 2.3% (1.7-3.6%); total weight lifted, 2.5% (2.2-2.9%); (b) men and women: the variation was higher for the women compared with the men in both snatch (ratio of CV, 1.4) and clean and jerk (ratio of CV, 1.5), and for the total weight lifted (ratio of CV, 1.3); and (c) the top 5 athletes were almost always more reliable than the bottom-half athletes. Recent work has shown that the smallest worthwhile performance is about half the within-athlete variation, so coaches and sport scientists should focus on enhancements of as little as 1.2% in total weight lifted for elite olympic weightlifters. KEY WORDS. reliability, olympic weightlifting, coefficient of variation, elite athletes

Often sports announcers, particularly in baseball, provide the listener with exaggerated information concerning a player's performance. For example, we may be told that Dave Winfield, a popular baseball player., has hit safely in 8 of his last 17 chances (a batting average of .471). This is biased, or selected information, as the ' 17' was chosen to maximize the reported percentage. We model this as observing a maximum success rate of a Bernoulli process and show how to construct the likelihood function for a player's true batting ability. The likelihood function is a high-degree polynomial, but it can be computed exactly. Alternatively, the problem yields to solutions based on either the EM algorithm or Gibbs sampling. Using these techniques, we compute maximum likelihood estimators, Baves estimators, and associated measures of error. We also show how to approximate the likelihood using a Brownian motion calculation. We find that although constructing good estimators from selected information is difficult, we seem to be able to estimate better than expected, particularly when using prior information. The estimators are illustrated with data from the 1992 Major League Baseball season. KEY WORDS: Brownian motion: EM algorithm; Gibbs sampling; Selection bias., Baseball announcers often exaggerate when describing the performance of some players by selective reporting of batting average instead of giving current batting averages. Such practice can be explained in terms of a Bernoulli process model where a maximum success rate is observed. It is shown that better estimates are derived when using prior information than by making estimates based on selected information., 1. INTRODUCTION Sports announcers--in particular, baseball announcers--often use hyperbolical descriptions of a plaver's ability. For example, when Dave Winfield, a popular baseball player, is batting, rather than report his current [...]

A recent article in Sports Illustrated argues that the National Football League (NFL) preseason schedule should be eliminated because the games are meaningless, feature subpar players, and are of no interest to fans. In this article we challenge this view, using statistical techniques to analyze both the outcome of preseason games and their value as a predictor of regular season performance. Our analysis is based on the application of a rank-order tournament model to the NFL labor market. The model explains preseason results in terms of the competition for positions on teams. Two implications are that the outcome of preseason games should reflect the labor market characteristics of the participating teams and that a team's preseason record should be positively related to its regular season record. Both of these hypotheses are consistent with our analysis of NFL data from 1970 through 199 1. KEY WORDS: Logit; Rank-order tournament; Seemingly unrelated regression., The preseason schedule of the National Football League (NFL) has been criticized for having no useful purpose and featuring subpar games and player performances. Statistical methods are applied to challenge this view. A rank-order tournament model is applied to the NFL labor market. It is shown that preseason performance is conditioned by competition for team slots. In addition, preseason performance is conditioned by regular season record., 1. INTRODUCTION In a recent essay in the sports weekly Sports Illustrated, Rick Telander argued that the National Football League (NFL) should eliminate its preseason schedule because the games provide [...]