Chris Duncan vs Adam Dunn in the field

[Banedon]

Both bad OF's. Card fan buddy mentioned he'd like Dunn, and while I completely understood, and so would I, I made the comment "If you thought that Duncan was an adventure in the OF....."

 

He thinks Dunn is twice the fielder that Duncan is. I need some stat nerdy types to tell me who the better fielder is, and how much better.

[Transmogrified Tiger]

UZR and PMR suggest Dunn is a run or two better defensively in LF.

[Banedon]

UZR and PMR suggest Dunn is a run or two better defensively in LF.

 

Is that significant?

[nilodnayr]

UZR and PMR suggest Dunn is a run or two better defensively in LF.

 

Is that significant?

 

No, not at all

[Banedon]

Thx guys.

[CubinNY]

Thx guys.

Those "stats" are horrible. They are especially horrible for OF.

[Banedon]

Thx guys.

Those "stats" are horrible. They are especially horrible for OF.

 

If you've got better ones, feel free to volunteer them.

[TruffleShuffle]

Thx guys.

Those "stats" are horrible. They are especially horrible for OF.

 

If you've got better ones, feel free to volunteer them.

 

*crickets*

[treebird]

Thx guys.

Those "stats" are horrible. They are especially horrible for OF.

 

If you've got better ones, feel free to volunteer them.

 

*crickets*

 

the fact that you don't have anything better than a magic rock to keep tigers away doesn't mean you should use the rock

 

 

(and yes, i realize i've completely changed the context of the reference; i have no idea why this scenario came to mind, but it did)

[nilodnayr]

Thx guys.

Those "stats" are horrible. They are especially horrible for OF.

 

If you've got better ones, feel free to volunteer them.

 

*crickets*

 

the fact that you don't have anything better than a magic rock to keep tigers away doesn't mean you should use the rock

 

 

(and yes, i realize i've completely changed the context of the reference; i have no idea why this scenario came to mind, but it did)

 

But along the spectrum of magic rock to measuring tape, UZR falls closer to measuring tape than magic rock. No one is saying its perfect, but horrible is a wee bit extreme.

[CubinNY]

Thx guys.

Those "stats" are horrible. They are especially horrible for OF.

 

If you've got better ones, feel free to volunteer them.

 

*crickets*

 

the fact that you don't have anything better than a magic rock to keep tigers away doesn't mean you should use the rock

 

 

(and yes, i realize i've completely changed the context of the reference; i have no idea why this scenario came to mind, but it did)

 

But along the spectrum of magic rock to measuring tape, UZR falls closer to measuring tape than magic rock. No one is saying its perfect, but horrible is a wee bit extreme.

Just look at fielding % (catching) or OF assists (throwing). Simply because something is new and shiny doesn't make it useful. It's not like either Duncan or Dunn are super good route runners that get errors on plays others guys wouldn't get to. In addition, UZR and the rest of the fielding stats that try to quantify defensive prowess aren't reliable enough to even cite at this point in time, using them as a judgment aid is about as useful as a magic rock.

[Transmogrified Tiger]

Can you even explain what UZR's methodology is?

[CubinNY]

Can you even explain what UZR's methodology is?

I don't need to when you can look it up.

[Transmogrified Tiger]

Then what is it? And why is it so awful?

[CubinNY]

Then what is it? And why is it so awful?

It's awful because it is not a reliable tool.

 

Why should a player's defensive ability fluctuate with whomever is on the mound, or in a different ballpark, or vary wildly from year to year?

 

I know you don't think I know anything about UZR, but I do. I'm just skeptical of judgment aids that are so unreliable.

[Transmogrified Tiger]

Then what is it? And why is it so awful?

It's awful because it is not a reliable tool.

 

Why should a player's defensive ability fluctuate with whomever is on the mound, or in a different ballpark, or vary wildly from year to year?

 

I know you don't think I know anything about UZR, but I do. I'm just skeptical of judgment aids that are so unreliable.

 

If you do know, then explain it.

[CubinNY]

Then what is it? And why is it so awful?

It's awful because it is not a reliable tool.

 

Why should a player's defensive ability fluctuate with whomever is on the mound, or in a different ballpark, or vary wildly from year to year?

 

I know you don't think I know anything about UZR, but I do. I'm just skeptical of judgment aids that are so unreliable.

 

If you do know, then explain it.

It's awful because it is not a reliable tool. See above.

 

Reliability and validity are the defining quality indicators of any measurement device. If you don't have one or the other the tool is not useful.

[Transmogrified Tiger]

For the 4th time, can you explain how UZR measures defensive performance? To answer your question, the reason for fluctuations based on those factors is because it's a measure of performance, not ability, as well as outliers coming from UZR's imperfections.

[CubinNY]

For the 4th time, can you explain how UZR measures defensive performance? To answer your question, the reason for fluctuations based on those factors is because it's a measure of performance, not ability, as well as outliers coming from UZR's imperfections.

I don't need to explain it to you. You already know.

 

I'm not interested in a pissing contest here.

 

UZR is not reliable and doesn't account for throwing and throwing is at least half the equation.

 

It's not a measure of performance when it can show that player X is better than league average when a sinkerballer is on the mound and worse than league average when a fly ball pitcher is on the mound. The simple fact that variables other than defense are going into the metric is enough to invalidate it. It's probably useful a gross measure but to use it to see who sucks worse seems a little overblown to me.

 

Explain these UZR numbers for Omar Vizquel:

-8.4 2002
2.8 2003
-.2 2004
10.1 2005
6.7 2006
19.9 2007
5.3 2008

Nobody plays that inconsistently from year to year. And that is just one example.

[Mark_R]

http://letmegooglethatforyou.com/?q=cubinny+knows+what+UZR+is

[Mephistopheles]

But along the spectrum of magic rock to measuring tape, UZR falls closer to measuring tape than magic rock. No one is saying its perfect, but horrible is a wee bit extreme.

 

 

That's not exactly true. Only 20% of a players UZR is based off of his UZR in previous years. That coupled with the fact that other defensive stats with similar methodology can often differ by 20 runs or more and you've got a grab-bag of stats that you shouldn't put hardly any stock into. Tree's right. It's much more magic rock than measuring tape.

[nilodnayr]

But along the spectrum of magic rock to measuring tape, UZR falls closer to measuring tape than magic rock. No one is saying its perfect, but horrible is a wee bit extreme.

 

 

That's not exactly true. Only 20% of a players UZR is based off of his UZR in previous years. That coupled with the fact that other defensive stats with similar methodology can often differ by 20 runs or more and you've got a grab-bag of stats that you shouldn't put hardly any stock into. Tree's right. It's much more magic rock than measuring tape.

 

 

But thats not the argument.

[Mephistopheles]

Sure it is.

[nilodnayr]

Sure it is.

 

nope

[Mephistopheles]

Sure it is.

 

Tree's Point: UZR and other fielding stats are uselss.

Meph's Argument: Two things cause defensive stats to suck:

 

1. Year to year they're unstable and worthless. Being good or bad seems to have very little effect on if you will be good or bad the next year. Meaning IF they're accurate, they're useless as an indicator of future defensive value. We're better off ballparking it through a combination of scouts eye and a couple things here and there. Trying to nail down the value with an "exact" number is futile at this stage.

2. Each defensive stat, while based on similar methodology often come to drastically different results quite often. It's fairly common for a player to be -10 in one, and +10 in another, or -20 in one and 0 in another. This is akin to OPS placing a hitter as an average average, say OPS around 750, and then another metric, say EqA putting him as one of the top ten hitters in the league, around .310.

 

The combination of #1 and #2 makes Tree right.