ZiPS Projects The Red Sox

Rick Osentoski-US PRESSWIRE

ZiPS has released it's projections for the 2013 Red Sox. Why is this team a bad match for the computer-based system?

While Opening Day is still a ways away, projection season is in full swing right now. One of the contenders is Dan Szymborski's ZiPS system over at Fangraphs. Last night Carson Cistulli tweeted this summary image, but today they've released the complete set, from plate appearances to WAR.

The quick math on the numbers in that image shows that Szymborski puts the Red Sox at about 38 WAR for their 20 most important players. Unfortunately, that's not far off the pace they set last year. On the other hand, the Red Sox underperformed before they tossed in the towel at the end of the season, and 38 WAR would put them in decent company, if short of the elites.

There's a fair few predictions, however, that at least to me seem to miss the mark, probably due to the objective inflexible nature of the system.

For instance, Jonny Gomes in left is projected to hit .240/.332/.423 and provide 0.3 WAR. Daniel Nava is projected to hit .240/.329/.369 and provide just 0.4 WAR. Either projection might make sense with either player taking on the role of full-time starter (and, frankly, with Gomes getting surprisingly little bump from a transition to Fenway). As it stands, though, conventional wisdom would say we're in for a platoon situation--one which ZiPS certainly does not seem to have accounted for given Gomes' 391 plate appearances. Of course, platoons never work out as well in reality as we imagine them on paper, but still, between a healthy Nava and a well-utilized Gomes, one can certainly hope left field would provide at least a little more than 0.7 WAR.

A similar situation may explain David Ross getting only 163 plate appearances, though his low batting line is still a bit surprising, and ZiPS can't necessarily account for what he brings to the table defensively.

There are plenty of other oddities to be found thanks largely to players with little history, Boston's recent run of injuries, and how ZiPS projects playing time. Ryan Lavarnway, for instance is expected to rack up nearly 600 PA (I'd expect less than 100, personally), while David Ortiz barely breaks 400. Junichi Tazawa has an ERA over 4.00 while still picking up 80 innings, and John Lackey will still manage to pick up 21 starts while Allen Webster and Rubby De La Rosa wait in the wings despite having a 5.24 ERA. Like that could ever happen.

On the whole, though, there are just a ton of average players, and that's not hard to explain. After all, these are computer predictions. If you or I were to make projections, we might say to ourselves "Will Middlebrooks could endure a serious sophomore slump, or he could be as good as he was once he got over that mid-season slump." The computer will take a point in the middle and return a pretty mediocre result. This year, the Sox are built on that sort of player who could end up being a big plus or a lost cause, with the hopes that replacements can be found in the minors or elsewhere for the ones that miss. Basically, they're hoping some of those guys that ZiPS counts as 1.0-2.0 WAR players end up more in the 4.0 range, and then the ones who are headed fast towards negative territory can simply be replaced.

This is not a criticism of ZiPS, mind. The system does what it says it does. It's just that the Red Sox are an unusual case this year, and between their lottery tickets and luxury platoon players, particularly ill-suited for this sort of projection. While some of the individual projections might be worth talking about, it just doesn't provide a great view of the team as a whole.

X
Log In Sign Up

forgot?
Log In Sign Up

Forgot password?

We'll email you a reset link.

If you signed up using a 3rd party account like Facebook or Twitter, please login with it instead.

Forgot password?

Try another email?

Almost done,

Join Over the Monster

You must be a member of Over the Monster to participate.

We have our own Community Guidelines at Over the Monster. You should read them.

Join Over the Monster

You must be a member of Over the Monster to participate.

We have our own Community Guidelines at Over the Monster. You should read them.

Spinner

Authenticating

Great!

Choose an available username to complete sign up.

In order to provide our users with a better overall experience, we ask for more information from Facebook when using it to login so that we can learn more about our audience and provide you with the best possible experience. We do not store specific user data and the sharing of it is not required to login with Facebook.

tracking_pixel_9351_tracker