Fun with Range

by squirrel on Jan 31, 2005 8:55 PM PST Comment 5 comments

Over at Baseball Musings, David Pinto has been playing around with his probablistic model of range. Though this is by no account the end all of fielding measurments, there are two weird little things that should be of interest to Giants fans.

First:

Rich Aurilia 2070    243    242.28    0.117    0.117    [b]0.00035[/b]
Bobby Crosby    4132    557    557.61    0.135    0.135    -0.00015
Jose C Lopez    1533    164    165.00    0.107    0.108    -0.00066
Jimmy Rollins    4187    473    476.56    0.113    0.114    -0.00085
Alex Gonzalez    3996    482    485.71    0.121    0.122    -0.00093
Neifi Perez    1729    202    203.81    0.117    0.118    -0.00105

FYI, a 0.0 would mean a fielder got every ball he was predicted to get, and anything over 0.0 is gravy. Only five SS got more than zero. But whatever, I couldn't pass up an oportunity to throw Neifi to the wolves yet again.

Second:

Marquis Grissom    3799    342    342.66    0.090    0.090    -0.00017
Mike Cameron    3772    354    355.96    0.094    0.094    -0.00052
Torii Hunter    3346    312    313.81    0.093    0.094    -0.00054

This not only shows Grissom in the same league as Cameron and Hunter, but slightly better last year. This isn't meant as a support for Grissom's fielding, but more an odd little factoid. Pinto has said that his model isn't perfect, but still, its weird. According to Pinto's model, Grissom's ranked 15th among CF's who had at least 1000 balls in play, getting almost everything he was expected. Far from a disaster. But of course, when placing Bonds and Alou next to him, average isnt going to cut it, especially over in Pac Bell.

This FanPost is reader-generated, and it does not necessarily reflect the views of McCovey Chronicles. If the author uses filler to achieve the minimum word requirement, a moderator may edit the FanPost for his or her own amusement.

0 recs | Comment 5 comments |

Story-email Email Printer Print

Comments

Fun With Range

I've been watching Pinto post those defensive ratings, and they seem very, uh, squirrely. You acknowledge that, but I just felt it needed emphasis. I wouldn't necessarily trust Pinto on this one.

by leftymalo on Jan 31, 2005 9:41 PM PST reply actions

Hmmmm

Although I greatly respect your opinion lefty, I must disagree. Yes, the probalistic model of range possesses very real limitations in both its method, of which I am admittedly unclear, and interprative value (not park adjusted, etc); however, we should not discount the data it produces simply because it does not gel with expectation. Perhaps PMR does not allow us to safely conclude that Grissom rocks the casbah as a center fielder. But, I believe that I can conclude with relative confidence that he is not as sub-par as we fear.

by kenshin1 on Jan 31, 2005 10:10 PM PST up reply actions

Not just Grip

Point well taken. But even Pinto admits he's still playing around with the formulas, and they should be taken with a big grain of salt.

by leftymalo on Jan 31, 2005 11:07 PM PST reply actions

I guess...

I could have emphasized more that these numbers are by no means definitive. The main reason I posted these was just becuase they were very interesting to me. I know there are a lot of variables that could make these numbers completely meaningless, but considering where the discussion has gone recently, they seemed interesting. And the Neifi/Aurilla thing was just funny, no matter where it came from.

by squirrel on Feb 1, 2005 2:07 AM PST up reply actions

They are Park Adjusted

They are park adjusted but I think the way he did it didn't quite make sense. I think Tango Tiger posted a comment on Pinto's first post of the numbers asking something about this.

The one thing that strikes me about these numbers is that they are very close together in most cases. If you go just by a percent bases, or three decimal points like AVG or OBA, then most of the fielders are very close together. So I wouldn't look too much in the exact rankings except for the extremes.

In the numbers posted above Aurilia makes 100.08% of expected outs and Neifi makes 99.11%. Grissom makes 99.81%, Cameron 99.45%, and Hunter 99.42%.

Also, on his post today it looks like he includes flyballs and liners in the infielders calcualtions. I understand flyballs outside of the infield where a fielder has to run to get to and has a chance to fall, but it would be a lot better if he could somehow not include infield pop-ups. On most pop-ups there are multiple fielders that can get to it so whether you catch it or not does not reflect ability. This isn't little league, people in the majors almost never drop infield pop-ups. I'm a little unsure on the line drives as well.

This is the breakdown for Christian Guzman, who rated very well this year:

Type of Ball InPlay Actual Outs Predicted Outs DER Predicted DER Difference
Fly 1358 85 84.66 0.063 0.062 0.00025
Liner 718 37 26.16 0.052 0.036 0.01510
Grounder 1786 377 381.53 0.211 0.214 -0.00254
Bunt Grounder 80 0 0.00 0.000 0.000 0.00000

He's getting 141.44% of expected line drives. This seems fishy to me. How could he be that much better than the average 2B at line drives? I think it would probably be better to throw them out altogether.

Signatures can be wild and exciting. But this one isn't.

by Nick Schulte on Feb 1, 2005 9:55 AM PST reply actions