Andrew McCutchen’s offensive numbers are down, but his approach is better than ever. (Photo Credit: David Hague)
It’s a three-word term you will hear often early in the season: small sample size.
Carlos Gomez leads the NL in batting average? Small sample size. Chris Davis is second in the majors in OPS? Small sample size. B.J. Upton is last in the NL in on-base percentage?
You get the point.
But not all early-season statistics are unreliable noise. If you’re looking in the right places, you can find stats that are indicative of a player’s true talent, and pretty good predictors about what he will do the rest of the year.
The best studies on the reliability of sample sizes have been done by Russell A. Carleton, first under the alias “Pizza Cutter” in 2007 and revised in 2012 under his own name for Baseball Prospectus.
At this point in the season, the rate stats that have a large enough sample for everyday players are those that measure contact, swings, strikeouts, walks and line driv...