PECOTA and the Angels: Projecting the 2013 Season

facebooktwitterreddit

Feb. 12, 2013; Tempe, AZ, USA: Los Angeles Angels pitcher Jered Weaver throws during spring training at Tempe Diablo Stadium. Mandatory Credit: Mark J. Rebilas-USA TODAY Sports

Every year Baseball Prospectus releases an elaborate projection system called PECOTA that was originally conceived by Nate Silver, a man whose algorithmic-predictive ability gained him notoriety outside of the realm of baseball this past fall when he very accurately predicted the outcome of the 2012 Presidential race.

You have to be a subscriber of BP in order to view the full projections which are quite comprehensive—and especially useful for fantasy purposes. Even if you have no interest in PECOTA or fantasy baseball, you should get a BP subscription—trust me, you won’t be sorry.

Anyway, the PECOTA projection system behaves in much the same way as other, similar projection systems in that it uses an elaborate algorithm which takes into account player histories, traditional aging curves and models of regression to project the statistics of players for the coming season. These statistic combine to give each player a specific value in terms of runs added which are then combined with the other players on their team to give each team a projected run differential. These run differentials are then used to project a team’s record for the coming season.

So how did the Angels fair in the 2013 projections?

Before we answer that question, it’s important to note a few things about projection systems if you’re unfamiliar with them.

First, a projection is not a prediction. A prediction is a subjective and usually informed guess as to how well a player or team will do. A projection is a computer-generated outcome which uses pre-determined variables to spit out a series of numbers. This is an important distinction as PECOTA—or any other projection system—is not subjective. If you think the projection is high or low on a player or team, there is no one to blame or to call an idiot.

Secondly, it’s important to realize that projection systems are naturally conservative—and PECOTA is particularly so. Seasons like Mike Trout’s 2012, for instance, are extremely rare and are often outliers. Seasons like that are at a very unlikely percentile and so no projection system can predict them. In fact, it would be irresponsible for them to do so.

Taking that into account, you will rarely see a 100-win or 100-loss projection on a team level as those types of seasons are not just the result of on-the-field talent, but also of luck and random variation. Projection systems are not equipped to handle luck.

So now that I’ve wasted 400 words of your valuable time, let’s get to some of the Angels projections for 2013.

The first thing to note is that the Angels are projected to win 90 games—the most in the AL West. The Rangers are at 87, the A’s at 83, Seattle at 79, and the Astros at 63. Remember what I said about 100-loss seasons a couple paragraphs ago? That should give you some idea of how terrible the Astros actually are.

90 wins for a team with the Angels talent may seem on the low side, but like I said, PECOTA is conservative—only the Tigers and Yankees (both at 92) are projected to have more wins in the American League.

On an individual level, the two best players on the Angels according to PECOTA will be, unsurprisingly, Albert Pujols and Mike Trout; what may surprise you is the order. Pujols is projected to lead all of baseball with a 6.9 WARP (Baseball Prospectus’ Wins Above Replacement metric, which you can read about here), while Trout is projected for a 5.3 WARP season.

Again, Trout’s numbers may seem conservative once you consider that Baseball Prospectus rated him as contributing more than nine full wins above replacement in 2012, but there’s at least a decent chance that Trout’s breakout season will also end up being his best season. A 5.3 WARP still puts Trout among the very best players in baseball according to PECOTA. And let’s all keep in mind that he’s still just 21. 9-10 WAR seasons do not come along often unless you’re Barry Bonds.

Other notable projections include Mark Trumbo who PECOTA projects to hit 30 home runs with a typically low on-base percentage of .303, which is probably bullish considering how terrible he was last season after June.

Josh Hamilton projects to hit some regression in 2013 with a .334 on-base percentage and a significantly lowered .486 slugging percentage to go along with 27 home runs. If he stays healthy over a full season, one would have to think he’ll at least slug a little higher than that.

On the pitching side, Jered Weaver naturally projects to be the Angels’ best pitcher with a 3.03 ERA and a 4.0 WARP, while C.J. Wilson and Tommy Hanson project to be firmly average at around 2 WARP a piece. Joe Blanton projects to be below-replacement level which seems unlikely when you consider the move to a more pitcher-friendly environment at Angels Stadium and also considering he’s never posted such pitiful value before. With his flyball tendencies and a terrific outfield defense in Anaheim, it’s difficult to see Blanton being that bad.

Ryan Madson, Scott Downs and Sean Burnett all project for solid, but not spectacular, seasons out of the bullpen and PECOTA predicts a little negative regression for Ernesto Frieri.

Overall, the PECOTA system really seems to like the Angels and it’s hard to argue given the high-end talent on the roster. Entering in a little subjective opinion, I think the system is a little bullish on Pujols and perhaps the same can be said of Trumbo and Weaver, but I also think it’s a little low on C.J. Wilson, Joe Blanton and Jason Vargas (who projects to a 0.9 WARP) provided all three can stay healthy in 2013.

For a full breakdown (with spreadsheets and charts, y’all!) of the PECOTA projections for all 30 teams, head on over to Baseball Prospectus and get a subscription.