White hot and ice coldby Sal Baxamusa
May 12, 2008
Remember when Chris Shelton hit 10 home runs a few Aprils ago? The slugger-in-waiting had missed a bunch of development time after being plucked away from the Pirates in the 2004 Rule 5 draft, and his white hot April 2006 (.326/.404/.783) was hailed as a sign that Shelton had finally arrived. At the time, nobody thought that he would keep up that kind of performance, but there was a lot of chatter that the up-and-coming Tigers had struck gold and found a slugging firstbaseman they could rely on for the next few years.
It didn't last, of course. We have a natural tendency to attach significance to hot starts, and rationalize them ex post facto: he's finally healthy, he had offseason Lasik, the change of scenery did him good, it's gotta be the shoes. Occasionally, these explanations hold water, but more often than not, hot or cold starts are just a small-sample illusions.
But that doesn't mean that we should ignore them altogether. A player's talent is a dynamic thing, and his performance over the last six weeks is a better indicator of that talent than what he did over a six-week stretch last year. Recently, I've shown how we can appropriately weigh recent performance in order to estimate how a player's true talent changes over time. Prolonged streaks or slumps definitely change what we should expect a player to do going forward, but the effect is usually small.
Let's check out a few hitters who are off to hot starts and see how their true talent has changed over the first six weeks of the season.
Before his age-24 season, folks were ready to write Casey Kotchman off as a bust, but Kotchman broke out in a big way last year. He didn't flash big-time power, but a high average, good strike zone discipline, and a sweet doubles stroke inspired Mark Grace comparisons. Kotchman is off to great start this year, smashing six homers already and leaving observers asking if he's finally developing the power you would expect from a big kid like him.
Kotchman true talent April 1: .288/.365/.452 Kotchman true talent today: .292/.367/.464
Kotchman's true talent has hardly changed at all: a few points of average and about 12 points of slugging. Over 600 at bats, that difference in slugging is about 7 extra total bases - one homer, one double, and one single.
What about Carlos Quentin? Quentin is near the top of the league in OPS, slugging, and home runs. Not bad for a guy who came into the year with a career .742 OPS. How has our estimate of his true talent changed in the last six weeks?
Quentin true talent April 1: .251/.333/.429 Quentin true talent today: .255/.348/.456
Forty points of OPS is nothing to sneeze at. He's pulled up his true talent OBP to an acceptable level for a corner outfielder, and his slugging has jumped as well. Quentin's not going to maintain a .950+ OPS for the year, but he's no slouch either. Nice pickup by Kenny Williams.
The dark days in Pittsburgh continue, but there are reasons to watch. Two of those reasons are Nate McLouth and Ryan Doumit, who have come out of - well, not nowhere, but not really anywhere either - to lead the Pirates to what has been, to date, the fourth most prolific NL offense.
Doumit true talent April 1: .264/.340/.436 Doumit true talent today: .272/.344/.452 McLouth true talent April 1: .263/.340/.446 McLouth true talent today: .270/.350/.466
McLouth and Doumit have apparently developed some real power. As up-the-middle defenders, they are turning into very valuable palyers as they enter their peak years. Will Neal Huntington and company right the ship quickly enough to take advantage? Simply playing McLouth at all has to be considered a win for Huntington and manager John Russell. Things are looking up in Steel Town.
Not everybody comes out of the gates quickly, however. The Dodgers signed Andruw Jones expecting him to bounce back from his awful 2007. Some commentators scoffed, saying that Andruw was cooked. He's sure played the part this year. His miserable 2007, combined with the pratfall-to-date that has been 2008 has changed our estimate of his true talent.
Jones true talent April 1: .250/.339/.472 Jones true talent today: .245/.335/.459
Given the way things have gone, I think I'd be surprised if he managed even that line for the remainder of the season.
Emboldened by the successes of Melky Cabrera, Chien-Ming Wang, and Robinson Cano, the Yankees decided to go with the youth movement this year. But Cano, billed as the second coming of Rod Carew, has been awful so far this year. How has he changed over the last six weeks?
Cano true talent April 1: .307/.348/.481 Cano true talent today: .294/.337/.462
It's pretty amazing, no? Cano is struggling to keep his OPS over .550, and he's received around 150 plate appearances this year. We're weighting those aweful 150 plate appearances the most when considering his true talent projection. And yet, he still projects as a true talent .799 OPS, which is definitely an asset at second base. If you play fantasy baseball, now might be a good time to buy low. (Disclaimer: you probably don't want to take fantasy advice from the guy who still hasn't replaced Troy Tulowitzki on his fantasy team. Derek Carty is your man.)
Alfonso Soriano's 2008 performance is reminiscent of Cano's, albeit in fewer plate apperances due to the hamstring he tweaked doing that little crow-hop of his. Don't worry Cub fans, he's not that different of a player today than he was at the start of the year.
Soriano true talent April 1: .275/.333/.504 Soriano true talent today: 0.272/.331/.495
What about the struggling Ryan Howard? Phillie partisans have resorted to latin phrases of scholastic philosophy to describe the depths of his slump.
Howard true talent April 1: .288/.396/.584 Howard true talent today: .278/.387/.564
Like Cano, Howard has seen his true talent OPS fall by about 30 points. Fortunately, Howard has a impressive track record of success, and mighty heights from which to fall. I think most fans, except maybe Mets fans, would be okay if he put up a .950 OPS for the remainder of the year. And the slump, while bad, hasn't totally collapsed our estimate of his true talent power or patience.
Using performance data to make up-to-the-minute true talent estimates is, of course, only one way to look at how slumps and streaks affect our view of players. The method is rigorous and its basis sound, but like all statistical evaluations we need a pretty large sample size before we start to see an effect. I probably should have warned you earlier that it would take one heck of a hot start to drastically change our opinion of a hitter's true talent in only six weeks. Regression to the mean is the statistical equivalent of a wet blanket.
Scouting perceptions of a player's weight, bat speed, or swing mechanics are invaluable at this point, since those things tell us about the process rather than the outcome. But those perceptions are not nearly as rigorous or reliable, or at least haven't been shown to be, so in the meantime accept this as a neutral evaluation of the players highlighted here. If we're interested in knowing whether these hitters will continue on the path they've started on, this is the best way to do it.
Sal Baxamusa is a graduate student in chemical engineering. He can be reached here.
<< Return to Article