HomeHOWHow Does A Pitcher Win A Gold Glove

How Does A Pitcher Win A Gold Glove

Pitching data included all counting statistics for pitchers, including ERA and opponents’ batting average, though the latter two were not used for the same reasons that batting average and other offensive-rate statistics were not used. Offensive-rate mea­sures (batting average, slugging, etc.) were excluded for pitchers (as for other positions). We decided to exclude these rate vari­ables because they depended on the number of innings pitched and batters faced, information that was not always available; this led to uncertainty about the reliability of these variables due to the size of statistical error in the measurements of their effects.

All records had indicators of league, season (as calendar year), whether the player was an All-Star in that year, whether the player made the post-season in that year, and age in each season (as of July 1). We also calculated a number of variables, including career totals and average-per-game rates for each counting sta­tistic. Variables were also created to indicate cumulative All-Star appearances. cumulative post-season appearances, and cumula­tive Gold Gloves won.

Gold Glove Distribution

Table 2 shows the distribution of all Gold Glove Awards in 1957- 2004. There were only 251 original winners of the slightly more than 850 awards given in that interval. About half the winners have won one or two awards each. Among the half who have won three or more awards, most have won between three and five, though there are 48 players (18%) who have won six or more Gold Gloves. This is no doubt the source of many sportswriters’ sug­gestions that the winner in any given year is whoever won the year before. In the absence of any other information, this is not a bad bet, and should be born in mind as the results from position­ specific models are presented.

Refer to more articles:  How To Knit Fingerless Gloves With A Loom

(Click image to enlarge)

Statistical Analyses

We fit logistic regression models to the data, with the Gold Glove indicator as the outcome variable (yes/no). The logistic regres­sion model fits the log (natural logarithm) odds of success for a binary variable (in this case win of a Gold Glove, yes or no) to a linear function of explanatory variables. The resultant parameter estimates can be used to calculate the probability of an event occurring based on the values of the explanatory variables for a given observation. Logistic regression is a robust method and is used widely in the health sciences. Further details on the meth­odology are available in standard statistics texts such as that of Hosmer & Lemeshow.

RELATED ARTICLES

Most Popular

Recent Comments