Why I Don’t Trust Defensive Metrics

From the latest Baseball Prospectus annual on Yuni Betancourt:

“That (fielding behind higher quality lefties) doesn’t explain the entire improvement, but the good thing about looking through a statistical “lens” is that the sample size is now very large and the image that comes into focus is that of an average defensive shortstop.”

I’ve now read plenty of pessimistic projections on the Brewers and I tried to dig into some data to find out why* (and before I get into it, those projections tend not to take into account a Braun suspension which would obviously make things even worse).  I understand on the surface that losing Prince Fielder looks like a huge deal, but I figured that upgrading the left side of the infield would mostly, if not entirely offset his departure.  Last night I decided to compare the four infield positions WARPs from last year with their PECOTA projected WARPs for 2012, just for kicks.  Last year according to BP the Brewer infield combined to post an 8.8 WARP (5.3 from Prince, -.9 from Casey, 3.1 from Weeks, and 1.3 from Yuni.).  If you’re anything like me that 1.3 from Yuni jumped out at you.  It jumped out at me even more because I know that Alex Gonzalez projects this year as a 1.5 WARP player which is barely an upgrade at all.  Looking more closely Yuni actually rated as a positive defender according to FRAA in both 2010 and 2011

I watched a lot of Yuni Betancourt last year, and I’ve seen a good chunk of Yuni as a Royal too by virtue of living in Chicago where the Royals face the White Sox on a regular basis, and the idea the Yuni Betancourt is an average to better-than-average shortstop is insulting to average shortstops everywhere.  The eye test is often very flawed and I’m certainly no scout, but I’m pretty confident on this one. It’s also worth noting that Fangraphs’ UZR hates Yuni’s defense almost as much as I do, and saw him as a -6.9 run fielder (and a .5 WAR player) in 2011. Image

I’ll go into more detail about my predictions for the Brewers season at some point in the future, but I’m much more optimistic than most of the projections I’ve seen, and a good part of that optimism is because of what I see as vastly improved defense (and offense) on the left side, and yes, that includes the much maligned Aramis Ramirez at the hot corner. 

Baseball Prospectus’s projection for the four starters on the Brewer infield in 2012 is a WARP of 6.8 (.2 for Gamel, 1.5 for Gonzalez, 2.2 for Ramirez, and 2.9 for Weeks), or 2 fewer than in 2011.  Make of that what you will. 

*Projections are more for fun than anything else, but one thing they are good for is helping to identify deficiencies.  If you expect to win the division by 4 games based on your projection and everything goes according to plan except you have a 5 win player who gets hurt and is replaced by Casey McGehee and you lose the division by 2 games, you know exactly where everything went wrong.  It’s not usually so simple, but you get the idea.  If a team is projected to decline as a whole, the individual projections should show you why. 

One thought on “Why I Don’t Trust Defensive Metrics

  1. Pingback: Hope Springs Eternal, Part 1 | Ron Roenicke Stole My Baseball

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s