Defensive Metrics Topic

Posted by bad_luck on 9/6/2013 1:40:00 PM (view original):
Posted by toddcommish on 9/6/2013 12:53:00 PM (view original):

And how do defensive metrics account for a ball-hogging CF who takes every single fly ball he can get to...?  Does that mean the RF/LF have lousy UZR numbers?

Or do they get credit for being close enough to make the play even if they're called off?

No, the RF would get zero credit (positive or negative) for a ball the CF caught.
Real life example: mid-to-late 70's Phillies.  Greg Luzinski in LF and Garry Maddox in CF.

Luzinski had so little range that he would stand half-facing the LF line to get a better jump on balls hit to his right, knowing that Maddox would cover LCF behind him.

So Luzinski would not be penalized by UZR?
9/6/2013 2:25 PM
HAHAHA!!  You pick fielding percentage(like I knew you would).   I initially put top 10 but narrowed it down some because, had I said "top 10", you'd have said "WHAT??  That's probably just above league average!!!"

Anyway, I wouldn't compare a batter with 300 PA to a batter with 800 PA and draw conclusions yet you're fine doing it in the field.    That's kind of dumb.
9/6/2013 2:27 PM
And what if that fielder only played 120 games? Are you still confident he had a great year based on his top 3 in F% at his position? What if he played OF on a groundball staff? What if he played third base like his feet were buried in cement? What if he was Miguel Cabrera and his fielding percentage was above average for his position in 2012? Would you still say he had a pretty good defensive year?
9/6/2013 2:28 PM
Please see above. 
9/6/2013 2:30 PM
Posted by MikeT23 on 9/6/2013 2:27:00 PM (view original):
HAHAHA!!  You pick fielding percentage(like I knew you would).   I initially put top 10 but narrowed it down some because, had I said "top 10", you'd have said "WHAT??  That's probably just above league average!!!"

Anyway, I wouldn't compare a batter with 300 PA to a batter with 800 PA and draw conclusions yet you're fine doing it in the field.    That's kind of dumb.
We do that all the time. Some hitters have 750 PA in a year. Some only have 500. We know most offensive stats aren't terribly reliable under 600 or 700 PA (900+ for batting average)

I've never seen you hedge your argument because the guy you're arguing for had less plate appearances.
9/6/2013 2:31 PM
There's a much larger difference between 300 and 800 as opposed to 750 and 500.   "Roughly" twice as much.

As for F%, same for most % stats, you have to look deeper.    If a 990% fielder had 400 chances and a 987% had 750 chances, you'd have to find out why. 
9/6/2013 2:35 PM
Posted by MikeT23 on 9/6/2013 2:35:00 PM (view original):
There's a much larger difference between 300 and 800 as opposed to 750 and 500.   "Roughly" twice as much.

As for F%, same for most % stats, you have to look deeper.    If a 990% fielder had 400 chances and a 987% had 750 chances, you'd have to find out why. 
OK. How exactly do you recommend we find out why?

I suggest we find a better way to evaluate defense.
9/6/2013 2:43 PM
Easy enough.  Games played and FB% for pitching staff.  

When you find a good way to evaluate defense, let me know.
9/6/2013 2:50 PM
UZR is pretty good. Miles and miles ahead of fielding percentage.
9/6/2013 3:01 PM
Yeah except for the whole "I can judge a RF on 300 chances and a SS on 800 chances" thing.

Who the **** would try to compare a batter with 200 PA to a batter with 575 PA?
9/6/2013 3:04 PM
Posted by MikeT23 on 9/6/2013 3:04:00 PM (view original):
Yeah except for the whole "I can judge a RF on 300 chances and a SS on 800 chances" thing.

Who the **** would try to compare a batter with 200 PA to a batter with 575 PA?
You're doing the same thing with F%.

9/6/2013 3:05 PM

Already addressed.  Read back. 

9/6/2013 3:08 PM
Posted by tecwrg on 9/6/2013 2:25:00 PM (view original):
Posted by bad_luck on 9/6/2013 1:40:00 PM (view original):
Posted by toddcommish on 9/6/2013 12:53:00 PM (view original):

And how do defensive metrics account for a ball-hogging CF who takes every single fly ball he can get to...?  Does that mean the RF/LF have lousy UZR numbers?

Or do they get credit for being close enough to make the play even if they're called off?

No, the RF would get zero credit (positive or negative) for a ball the CF caught.
Real life example: mid-to-late 70's Phillies.  Greg Luzinski in LF and Garry Maddox in CF.

Luzinski had so little range that he would stand half-facing the LF line to get a better jump on balls hit to his right, knowing that Maddox would cover LCF behind him.

So Luzinski would not be penalized by UZR?
"UZR is pretty good" according to BL.

Comments on the Luzinski example?
9/6/2013 3:09 PM
Posted by MikeT23 on 9/6/2013 3:08:00 PM (view original):

Already addressed.  Read back. 

It's perfectly ok (with any stat) to say "sometimes (often) we need to dig deeper." It's another to say "that stat doesn't address every possible fluke or bad bounce or odd positional alignment so it's garbage and should be ignored in favor of a way shittier alternative (f%, eye test, etc)."
9/6/2013 3:23 PM (edited)
Posted by tecwrg on 9/6/2013 3:09:00 PM (view original):
Posted by tecwrg on 9/6/2013 2:25:00 PM (view original):
Posted by bad_luck on 9/6/2013 1:40:00 PM (view original):
Posted by toddcommish on 9/6/2013 12:53:00 PM (view original):

And how do defensive metrics account for a ball-hogging CF who takes every single fly ball he can get to...?  Does that mean the RF/LF have lousy UZR numbers?

Or do they get credit for being close enough to make the play even if they're called off?

No, the RF would get zero credit (positive or negative) for a ball the CF caught.
Real life example: mid-to-late 70's Phillies.  Greg Luzinski in LF and Garry Maddox in CF.

Luzinski had so little range that he would stand half-facing the LF line to get a better jump on balls hit to his right, knowing that Maddox would cover LCF behind him.

So Luzinski would not be penalized by UZR?
"UZR is pretty good" according to BL.

Comments on the Luzinski example?
Probably not for that specifically. But if his range really was so bad, UZR would penalize him for not getting to high percentage balls, assuming he wasn't able to get to them.
9/6/2013 3:22 PM
◂ Prev 1...16|17|18|19|20...23 Next ▸
Defensive Metrics Topic

Search Criteria

Terms of Use Customer Support Privacy Statement

© 1999-2024 WhatIfSports.com, Inc. All rights reserved. WhatIfSports is a trademark of WhatIfSports.com, Inc. SimLeague, SimMatchup and iSimNow are trademarks or registered trademarks of Electronic Arts, Inc. Used under license. The names of actual companies and products mentioned herein may be the trademarks of their respective owners.