March 29, 2013
Spring Stats Without the Small Samples
At MLB.com—and also in the BP database—you can find complete spring training stats going back to 2006. Why do the stats start with 2006? I don’t know. Maybe no one at MLB thought they were worth tracking, until some smart employee realized that someone, years down the road, might want to do a blog post about the past eight seasons of spring training stats. If so, I dedicate this post to that person.
On Wednesday, I wrote about how Howie Kendrick has failed to win the batting title BP predicted he would, but has batted about .360 in spring training. That made me curious, so I went looking for other statistical standouts from the past eight exhibition seasons. I’m not talking about fluky single springs, like Jake Fox’s 10-homer 2011, or Micah Hoffpauir’s 2008, when he hit .413 and slugged .674. I’m talking about sustained spring excellence or awfulness. Mostly, good players are good and bad players are bad, even in exhibition games, but sometimes spring upsets the natural order of things, and not just over small samples. Combining several seasons of mostly meaningless spring training stats doesn’t make them that much more meaningful, so this is just for fun. Here are some of the more notable names, with minimums of 200 plate appearances for hitters and 50 innings for pitchers.
Todd Helton, .404/.470/.658, 378 PA
Granted, the Cactus League isn’t exactly Petco Park. Since 2011, the Rockies have played at Salt Rivera Fields, in Scottsdale; before that, they played at Hi Corbett Field, in Tucson. Scottsdale is 1257 feet above sea level; Tucson is over twice as high, and the air is dry in Arizona. And now we’re one step away from calculating Cactus League park factors, which is how sabermetricians get back in shape for the season.
Whatever Helton does to dominate spring training, he's still doing. PECOTA projects Helton to hit .276/.372/.408 in 2013. This spring, he’s hitting .407/.484/.741.
April, by the way, has historically been Helton’s second-worst month, so it’s not like he’s just at his best early on, before his back can start bothering them. He just hits like crazy all spring, then goes back to being his usual self. In his prime, Helton must have hit about .600 in March, but the furthest back we can check is 2002, thanks to the more rudimentary team-by-team spring training stats available at ESPN:
I don’t know what all of this means—most likely it doesn’t mean anything. But someday, when Helton is on the Hall of Fame ballot and his detractors say he didn’t hit enough away from Coors Field, you can remind them that he was the best spring training hitter of his era.
The worst, in case you were wondering, is Adam Everett, at .201/.258/.247 in 318 plate appearances. That makes more sense.
Dan Uggla, .231/.304/.403, 635 PA
Matt Kemp, .252/.291/.448, 561 PA
The success of Branyan, Ka'aihue, Fox, Jesus Guzman (1.031 OPS), Ryan Raburn (.973 OPS), and other so-called Quad-A players suggests that skill sets that work in Triple-A but don't translate to the majors do survive the transition to spring training.
Mariano Rivera, 57.3 IP, 0.47 ERA, 66 K, 6 BB, .342 OPS allowed
That said, Rivera has always had the stuff to get hitters out when the stakes are low: when ahead or behind by more than four runs, he’s allowed a career OPS nearly 100 points lower than he has when his opponent was within four runs. So it’s no surprise that he’s excelled in a low-pressure environment like the Grapefruit League.
To get serious for a second: Rivera’s 0.47 ERA is the lowest among the 438 pitchers with at least 50 spring innings pitched from 2006-13. The next-lowest is well over twice as high. Rivera is amazing, even when the outcomes don’t count.
Wandy Rodriguez, 120 IP, 7.13 ERA, 87 K, 63 BB, .917 OPS allowed
Cliff Lee, 131.7 IP, 6.29 ERA, 104 K, 26 BB, .910 OPS allowed
Thanks to Ryan Lind for research assistance.