Vernon Wells - UZR's Gold Glove?

Thursday, September 30 2004 @ 02:50 AM EDT

Contributed by: TangoTiger

As of Monday, Vernon Wells has 307 putouts, with an estimate of 334 balls in his "zone of responsibility". Zone rating (ZR) is simply balls caught per opportunity.

The play-by-play scorers make a note of exactly where in the field every ball landed or was caught using this grid:

ESPN arbitrarily sets the "zone of responsibility". I believe that this zone is static.

What are the issues?
* No man's land: not every zone is assigned. So, while the team ZR will be .70, the individual players, when totalled, will have a ZR of .85 or so. UZR takes care of this.

* Every zone is equally weighted: there is no distinction between catching a ball on the fringes of the zone, compared to catching an easy popup. UZR takes care of this too, by determining how often balls are caught on the fringes, and rewarding a fielder proportionately for catching those balls.

* Static zones: it's rather obvious that LH and RH have different spray patterns. I believe that ZR zones are the same. UZR takes care of this.

* FB/GB tendencies of the pitching staff: different types of pitchers will allow easier/harder flyballs to be caught. A FB pitcher will have a higher proportion of his flyballs caught than a GB pitcher. UZR takes care of this.

* Speed of batted ball: yada yada yada. UZR yada yada yada. Note: there is a scorer bias here, with an impact of which is yet to be determined. Probably 1 SD = 2 runs.

* Trajectory of batted ball: see above.

* Park factors: trying to catch a ball at Fenway or Coors is not the same as elsewhere. UZR catches this too.

* Base/out situation: you know the drill

* Inning/score: UZR *should* take care of this, but I don't know if the new version does yet.

* Double-counting: I have no idea what the current state is with the numerator in ZR, whether the DP counts as 2 outs and not 1, etc. It's one of those darn silly things where people are trying to force a stat beyond what it should be doing. Same thing with balls caught outside the zone. The fielder gets a "1" in his numerator, but "0" in the denominator.
* Shared zones: this probably impacts OF much more than IF. And this impact is probably not alot either, on the order of a couple of runs.

As you can see, there's alot to like about UZR. A tremendous amount. What's not to like is the dependency on the scorers for the trajectory/location issues. This might be captured somewhat in the "park factor".


Until UZR is published, we will have to stick with ZR. The league average is to convert 87% of balls in zone into outs. Therefore, an average CF, given the number of balls that Wells faced would have made 290 outs. Wells got 307, or +17 outs (which translates to about 15 runs). Repeating this for all CF, here are the top 10 and bottom 10 CF in ZR plus/minus:

ZR... +/-... BIZgames...
0.918 17 116 Vernon Wells, Tor ***
0.919 15 106 Aaron Rowand, CWS
0.901 12 127 Mark Kotsay, Oak
0.971 11 37 Nook Logan, Det
0.891 9 150 Carlos Beltran, KC/Hou
0.897 9 120 Corey Patterson, ChC
0.901 9 89 Milton Bradley, LA
0.889 8 132 Randy Winn, Sea
0.891 7 115 Torii Hunter, Min
0.954 6 23 Hiram Bocachica, Sea
0.931 5 31 Reed Johnson, Tor ***
0.783 (8) 30 Choo Freeman, Col
0.849 (8) 143 Juan Pierre, Fla
0.847 (9) 133 Tike Redman, Pit
0.799 (10) 46 Jeromy Burnitz, Col
0.803 (10) 51 Preston Wilson, Col
0.796 (12) 58 Craig Biggio, Hou
0.838 (12) 135 Marquis Grissom, SF
0.819 (13) 89 Bernie Williams, NYY
0.833 (17) 155 Andruw Jones, Atl
0.783 (19) 77 Ken Griffey Jr., Cin

I prefer the Plus/Minus figure than the rate because it's alot easier to grasp. .803? Hard to get much meaning out of it.

BIZgames is simply Balls in Zone divided by 2.88 (there are 2.88 BIZ per game in CF).

UZR knew all about Aaron Rowand last year, putting him in the same class as Erstad and Cameron. Will be interesting to find out more about this Nook Logan character. Wells did not fare well with UZR in the beginning, and only started looking decent last year.