The reaction to Kyle Kendrick’s stellar March raises an interesting question: Do spring training stats matter?
For fans, it's tough to look at Kendrick's career performance, see that he has surrendered only two earned runs on seven hits over 14 innings this spring and not be encouraged.
But would it matter if instead, Kendrick gave up eight runs on 15 hits during those same 14 innings?
The answer is a resounding "no."
Several articles, including Baseball Prospectus columns by Marc Normandin and Joe Sheehan, claim that spring training stats either rarely predict regular season performance or are completely unrelated.
There are a number of reasons why spring training stats are irrelevant. First and foremost, the sample size of at-bats or innings pitched is nowhere near that of the regular season. So while the effects of a hot streak or slump are mitigated over the course of the regular season, they can determine overall spring training numbers.
Second, teams aren’t always playing their best players during spring training games, so the competition is uneven and generally worse than the regular season.
Third, players may experiment with techniques in March that they wouldn't use during games that count.
"Usually in spring, if I get to three balls on a guy, I'll just throw it down the middle,'' Red Sox pitcher John Lackey said in a postgame interview last week.
While research shows that league-wide, spring training stats don’t matter, I thought it would be interesting to see if there were any trends that pertained to the Phillies' roster. I compiled a list of the likely starting pitchers and lineup, and then subtracted their regular season stats from their spring training stats for each of the past three years. For pitchers, I used ERA and WHIP. For batters, I used batting averaging and slugging percentage.
|ERA difference||WHIP difference|
|BA difference||SLG difference|
There are some trends. Joe Blanton’s regular season stats relative to his spring training stats have gotten worse each year, but the opposite has held true for J.A. Happ, Jamie Moyer, and Shane Victorino. These are most likely insignificant, and by and large there is no correlation between spring training and regular season performance.
But there might be one exception to the rule that spring training stats are irrelevant. Back in 2005, John Dewan from ACTA Sports found that batters who post significantly better slugging percentages in spring training than their career slugging percentages perform better offensively in the upcoming regular season.
Specifically, the batters must have 36 or more spring training at-bats and post a slugging percentage that is .200 or more better than their career slugging percentage (at least 100 previous regular season at-bats are required).
Is it really shocking that players who take monstrous leaps forward in spring training tend to play better in the regular season? Given that only 30 players out of the hundreds that play in spring training met the criteria in 2005, Dewan's .200+ rule only applies to a tiny minority of hitters. Furthermore, 30 players may be too small of a sample size to generalize any rule.
Setting aside these criticisms, are there any current Phillies players who fit Dewan's mold of improving their offensive performance this upcoming season?
Not right now. In other words, there is no Phillie currently on pace to have at least 36 at-bats this spring and a slugging percentage .200 or more higher than his career average.
That, of course, does not mean no Phillies will take a big step forward offensively this season. Nor does it mean that fresh faces like Phillippe Aumont and John Mayberry, Jr. have nothing to play for. Whether or not their spring performances are indicative of how well they’ll play in the majors, they will factor in to what their roles will be with the team going forward.
So sit back, relax, and enjoy the rest of spring training stress-free. As far as the Phillies’ regular season is concerned, it’s completely meaningless.
Ben Singer is a graduate of Brown University and a Philly.com Sports intern.