Wednesday, February 01, 2006

Another Way to Measure Range

In an earlier post about individual defensive performance, I discussed the complications of measuring range. Specifically, I introduced Range Factor and Zone Rating and listed the limitations of each statistic. In a more recent post, I discussed an alternative statistic – David Pinto’s Probabilistic Model of Range. In this post, I’ll discuss another range statistic created by David Gassko. Gassko introduces his range measure in The Hardball Times Annual, a book I recommend for all fans of sabermetric analysis.


To calculate his range measure, Gassko first estimates how many balls were in a fielder’s zone based on the number of ground balls, fly balls and line drives allowed and the number of balls in play against LH and RH batters when that player was in the field. Then he determines how many plays were made by the fielder and how he compares to the major league average. Finally, he translates this number into number of runs above or below average a player contributed defensively. He also prorates the result to 150 games played. The table below includes the results for all Tigers with 600 or more innings played at each position and how they ranked among other players at their position.


As you look at the results, please note that this is a new statistic with only one year of data. I like his concept and I think it is probably a better measure than range factor or zone rating but we’ll need more data before we can get a really good idea of its reliability. According to this measure, Brandon Inge and Nook Logan had the best range in baseball at their positions. Placido Polanco was about average and Carlos Guillen was a little better than average. Chris Shelton was the worst in baseball and Magglio Ordonez and Craig Monroe were among the worst.


I’ve done a lot with range this winter but I’m not done yet. I’m currently working on getting range factor, zone rating, Probabilistic Model for Range and Gassko’s Range into one database. I will then aggregate the four measures to get a final range rank for each player. If I can get a hold of other range statistics such as Ultimate Zone Rating, I’ll throw them into the mix as well. I’ll post the results of this exercise in coming weeks.


Table: David Gassko’s Range Statistic


Position

Player

Runs Above Average (150 games)

# of players at position

Rank

1B

Shelton

-29

30

30

2B

Polanco

1

29

14

3B

Inge

22

26

1

SS

Guillen

7

29

11

CF

Logan

36

27

1

RF

Monroe

-20

27

23

RF

Ordonez

-22

27

24

4 comments:

  1. Isn't this similar to the way Basebaprospectus and Win Shares calculates their fielding stats? I thought both used similar methods for estimating the number of plays a player should have made.

    ReplyDelete
  2. Yes, there are definitely a lot of similarities between the different methods of calculating range. Gassko's method is not totally new and might not be any better than the other methods. I'd like to get a hold of complete data on other methods so I can compare the results. It's still hard to get fielding data in a format that's easy to analyze though.

    ReplyDelete
  3. our outfield defense is in t-r-o-u-b-l-e

    ReplyDelete
  4. I think the Tigers know they have a problem with outfield defense. That is why there has been some talk about Logan and Granderson both playing in the outfield and Ordonez being the DH.
    The problem is that they'd probably lose more offensively than they'd gain defensively with that set up.

    ReplyDelete

Twitter

Blog Archive

Subscribe

My Sabermetrics Book

My Sabermetrics Book
One of Baseball America's top ten books of 2010

Other Sabermetrics Books

Stat Counter