THE BOOK cover
The Unwritten Book is Finally Written!
An in-depth analysis of: The sacrifice bunt, batter/pitcher matchups, the intentional base on balls, optimizing a batting lineup, hot and cold streaks, clutch performance, platooning strategies, and much more.
Read Excerpts & Customer Reviews

Buy The Book from Amazon


SABR101 required reading if you enter this site. Check out the Sabermetric Wiki. And interesting baseball books.
MOST RECENT ARTICLES
MAIL : You ask | We say

Advanced


THE BOOK--Playing The Percentages In Baseball

<< Back to main

Monday, August 03, 2009

It is impossible for a player to have “consistency skills” any more than a computer would have…

By

What do I mean by that?  Pujols slumping got me thinking about this again.  I think about it many times when I hear or read about so-and-so being consistent or inconsistent, as if it were a skill (I’m not saying it isn’t).  Pujols is usually considered a “machine” in terms of his consistency. Now that he has “slumped” (I’m pretty sure he has slumped before and we just don’t remember it), he is taking more heat - or at least there is more surprise - because of that reputation.

Anyway, what I mean is:

Let’s say that there is a consistency skill (and I don’t really think there is much of one, but that is not the issue here).  The most that a player can EVER do, if he is the most consistent player in the Universe, skill-wise (as opposed to what just happened-wise), is to have the same true talent level every time he steps to the plate (or pitches or attempts to field a ball or runs the bases or whatever), not withstanding his opponent, the park, the weather, etc.

If that is the case, he is STILL subject to the random fluctuation associated with a certain fixed skill level.  There is NOTHING anyone can do about that. Nothing. If his true K rate is 1 per 10 PA and he is so consistent that that never changes, he is still subject to a binomial standard deviation around that “p.” Same for HR rate, hit rate, BB rate, etc.  There is nothing that he can do about it, no matter how consistent he may be skill-wise.  If a player actually has been incredibly consistent over the course of his career or some long time period, it HAS to be a fluke.  A player cannot be more consistent than a computer that has the exact same skill level in every PA, and that computer “player” will produce a performance that looks like a bell curve with a known mean if we look at the distribution of results for a series of any fixed time periods (one month stretches, one week stretched, one day stretches, etc.).

That being said, is there any way that we can really distinguish between players who may have a high “consistency” skill and those that have a low one, without looking REALLY closely at some long-term numbers?  I don’t think so.


(75) Comments • 2009/08/07
Page 1 of 1 pages

<< Back to main