B-Ref's Defensive WAR

[Formerly Snayke]

Does anyone trust B-Ref's calculation of defensive WAR? Because right now Carlos Lee (that Carlos Lee) has been worth 1.7 wins on defense and leads the NL. That just doesn't seem right.

[NonProfitCow]

It just uses Total Zone, so no.

[SouthSideRyan]

UZR has him at +8. What I'm saying is defensive metrics at the micro level have a long way to go.

[Productive Outs]

http://www.fangraphs.com/statss.aspx?playerid=847&position=2B/OF

 

Check out his UZR for 2007. I'm not someone who thinks that Soriano sucks at defense, but that can't be right, can it?

[SouthSideRyan]

http://www.fangraphs.com/statss.aspx?playerid=847&position=2B/OF

 

Check out his UZR for 2007. I'm not someone who thinks that Soriano sucks at defense, but that can't be right, can it?

 

Those defensive #s manage to put him 9th overall in WAR for that season, and 6th on a per game basis.

 

I believe he was a very good defensive LF before injuries took their toll, but he wasn't that good.

[Ding Dong Johnson]

His arm really helped elevate those numbers.

[davearm2]

His arm really helped elevate those numbers.

It would be very interesting to see how sensitive those ratings are to OF assists.

 

Some of the best OFs have very few assists, because guys rarely test them. Meanwhile a guy with a crud arm might get a bunch of assists every year since guys are running at will.

[SouthSideRyan]

His arm really helped elevate those numbers.

It would be very interesting to see how sensitive those ratings are to OF assists.

 

Some of the best OFs have very few assists, because guys rarely test them. Meanwhile a guy with a crud arm might get a bunch of assists every year since guys are running at will.

 

It's not strictly assists they measure (the better ones at least.) It takes into account the total # of bases taken on an OF. It wont' matter if Juan Pierre has 10 assists, because he'll have 800 extra bases taken on him throughout the year. It's likely that Soriano's #s were inflated in '07 because word wasn't out on his arm quite yet. (i.e. in '08 that guy stays at 2nd vs getting gunned down at 3rd.)

[NonProfitCow]

I don't think you can completely discredit defensive stats because of weird stuff like that. Fluky crap happens. Uggla was hitting like .170 over a couple hundred PA, which is fluky enough, and now has a 30 game hitting streak. Casey Kotchman is hitting .340.

 

Lee has probably just been positioned perfectly for a large amount of his chances. No different than a lucky BABIP. Since all the metrics seem to be in agreement over it (TZ, UZR, DRS) that's gotta be what's going on.

[mul21]

Won't Lee have an advantage because of how short the Crawford Boxes are in Houston as well? I mean, there's less ground where a hit could fall in and a ball off the wall there might get caught by a Campana type guy at a different field.