March 18, 2005
Spring Training Stats
One of the great joys of being a baseball fan is the thrill that runs through you when spring training starts. Other sports have training camps in one form or another, but none provide the twists and turns on a daily basis that baseball's annual spring fling does. Those twists and turns, however, provide a unique challenge to the fantasy owner. Everyone wants to get a leg up on the competition and accurately pinpoint the identity of the next Albert Pujols or Johan Santana, but how do you separate the useful information from the dross?
Well, to some extent, you can't. Spring training stats aren't so much signal mixed with noise as noise masquerading as signal. Just about every caveat you can think of--small sample size, variable talent level from game to game and inning to inning, extreme park factors, very small sample size--gets applied to spring-training results. If the BP Not-Quite-Ready-For-Prime-Time Players were staging a dinner theater version of Se7en, the killer would probably force Clay Davenport to try and produce accurate Cactus League translations until his head exploded.
That doesn't mean, however, that fantasy owners should disregard spring numbers altogether. For one thing, major league teams don't disregard them. Job battles can be and have been decided on a handful of spring appearances despite player track records that would lead an impartial observer to the opposite decision. For another, chances are there's one guy in your league who bases his sleeper picks on the Grapefruit and Cactus League leader boards. (That would have been the person who gloated after snagging Abraham Nunez last year.) Knowing in advance which players another owner likes is always useful information.
Beyond all that, if you squint just right, you can get a fuzzy glimpse of the future in those spring stats. Just like a tarot card reading, the more you know about the kinds of answers people are looking for, the more accurate your answers will be. And just like reading tea leaves, pattern recognition plays a big part. In other words, there's as much art as there is science involved in making sense of spring training performances.
There are three things you have to keep in mind when sifting through spring stats:
The most important question to ask before diving in is, "What are the positional battles on the team?" A player almost never "forces his way onto the team" with a big spring; big springs help players win open jobs, but they rarely put an incumbent out of work unless the incumbent was on shaky ground to begin with. Once you know what roster spots are up for grabs, you need to figure out what kind of player the manager and/or organization would prefer to see win the job. (Call it the Dusty Baker Rule.) Finally, to provide at least a rough context for the numbers you'll be evaluating, you need to know whether the team plays in Arizona or Florida. Like the National League locales of the same name, the two spring leagues are very different run-scoring environments. The Cactus League in Arizona inflates offensive numbers, while hurting pitchers' stats; Florida's Grapefruit League has no such radical effect.
Once you know who to look for, and what lens to view their stats through, it becomes fairly easy to spot potential sleepers. What you're looking for is exceptional performances, not just good ones. For hitters, this is especially true in the Cactus League--a .350/.420/.550 line looks great, but not when the average hitter on the team is hitting .300/.360/.450.
There are also only a few stats you need to concern yourself with. For hitters, BA/OBP/SLG and raw home runs should be the main focus. It's also handy to keep an eye on strikeout-to-walk ratio and the player's raw strikeout total, as a more "old school" manager might decide that too many strikeouts are a bad thing in a young player. Stolen bases you can ignore, even from a fantasy perspective; players run a bit willy-nilly in the spring, and off batteries who may not care much.
As an example, let's take a look at the Cincinnati infield situation. Joe Randa was signed to be the third baseman, but he's there as a stopgap until prospect Edwin Encarnacion is ready. Based on their respective springs so far (.167/.286/.333 for Randa, .370/.485/.481 for Encarnacion) it might seem like the Reds wasted their money, but chances are they're going to look at Encarnacion's six strikeouts in 27 at-bats (granted, against six walks) and justify the Randa signing by concluding that the kid needs a bit more seasoning in the minors. Encarnacion is a decent bet to make an impact in the second half, but he'll probably spend April and May in Louisville. The battle at shortstop is developing along similar lines: perennial semi-prospect Felipe Lopez is hitting .346/.433/.423 in 26 at bats, but his seven strikeouts open the door for retread Rich Aurilia and his solid .320/.296/.560 line.
Reading pitching stats requires a bit more finesse. Walks, strikeouts and home runs allowed should be the main focus, with ERA being almost meaningless due to the sample-size issues. Again, you're looking for outliers, not simply good performances. Vicente Padilla, for instance, posted a stunning 16/0 K/BB ratio in 18 Grapefruit League innings in 2002, heralding his breakout season. Teammate Robert Person, coming off a 15-win, 4.19 ERA 2001, posted a respectable 13/6 K/BB ratio in 13 innings, but also allowed an unreal seven home runs, sending out a huge warning signal that all was not right with his arm.
The toughest task is trying to spot potential closers, something just about every fantasy owner is interested in finding. Saves, and even games finished, can be downright misleading because most teams have pulled their regular hitters by the late innings, so managers who want to get a good read on their pitchers will have used them by the fifth or sixth. A pitcher who is consistently working the final inning of spring games is essentially getting mop-up duty, not being groomed to close.
Take a look at the current spring numbers from the Diamondbacks staff, for instance. Greg Aquino, who finished last year as the closer, is hardly a lock to keep it up given his health issues and shaky peripherals from 2004. Of the candidates to replace him, Brandon Lyon has been by far the best (7/0 K/BB in five innings), but Brian Bruney (7/2 K/BB in five innings) has been nearly as dominant. Jose Valverde, however (2/6 K/BB, two home runs allowed in three innings), has been awful, and despite his prospect pedigree can probably be scratched off your "closers-in-waiting" cheat sheet.
It'll be another week or so before the final prospects and warm bodies get shipped out to the minor-league camps, leaving each team's core 40-plus-man rosters, so you should wait as long as possible for trends to develop in the stats before drawing any conclusions from them. Using these guidelines though, you'll hopefully be able to do a bit of fortune telling on your own.
Erik Siegrist is a beat writer for RotoWire, covering the Marlins, Nationals and White Sox.