…my un-analytical theory that defensive metrics WAAAAY overvalue players (i.e. I don’t trust them enough to feel confident in assuming their “value” values are reliable) forces me to adjust this, and assuming Endy Chavez is “really good” at defense, I’ve given him a +10.
You have to think that run prevention is about 98% pitching and 2% defense. Just think about it for a second – you’re saying that the very best defensive outfielder in the game (or one of the top two or three, at least), only catches about 10 balls per 162 games more than the average of some the very worst defensive outfielders in the game.
Of course, we know there’s a much larger variance than that between how many balls each of them catch, so you’re saying that you believe that all of those extra fly balls that the good defenders catch and the bad defenders don’t are due to the pitcher. You’re essentially throwing DIPS theory out the window.
Just so you’re aware of the consequences of your skepticism. It’s one thing to not believe that defensive statistics do a great job of separating great players from good players. It’s another thing entirely to believe that defense just doesn’t matter very much, which is exactly what you’re stating that you believe.
And I thought the "+10" equated to run values, not play values.
Because I’m not confident enough in the metrics to be able to accept the translation of “catching 10 balls per 162 games more than average” (i.e. making 10 more plays) into “preventing 10 more runs than average.”
But that’s my personal opinion, however wrong it may be, and not my analysis. I leave the analysis to those more interested and capable…
It’s pretty obvious. We know how many balls in play each team has a chance to catch. We know how many of those balls are caught and how many are not. We know how many runs each type of play made (or not made) is worth. These aren’t things that are really questionable. These are just facts.
Factually, the gap between good and average teams is much, much larger than you’re suggesting. The only way for that to be true is for the gap to be heavily influenced by the pitchers, not the defenders.
It might not be the conclusion you want to be drawing, but it’s the only one that works. If you think a great defender is, at most, +10 runs above average compared to an LF, then you think most of the difference in balls in play being turned into outs are due to the pitchers.
It’s pretty obvious. We know how many balls in play each team has a chance to catch. We know how many of those balls are caught and how many are not. We know how many runs each type of play made (or not made) is worth. These aren’t things that are really questionable. These are just facts.
Okay – this data may not be questionable (although the run value assignment on these plays is theory NOT fact – a subtle but important difference). But WHY these balls were caught (and, conversely, why the other balls WERE NOT) is a lot more subject to debate, and definitely not reliably measurable quite yet. Is it because of the pitchers? the defenders? the coaches? the hitters? the weather? the ballpark? the time of day? the field conditions? the league? the “WFB Grit and Hustle Inverse Subjunctive Quotient ®”? Who knows. There’s been some great work done on this and I know it will continue.
I’m an idiot with regards to this stuff, though, and I will submit my lack of knowledge to your (and others’) overlordship…
He WAS otherworldly, but he’s turning 31 in a few days, and defensive peak comes earlier than offensive peak. He’ll be good, but I don’t think it’s reasonable to expect that he maintain his career average anymore.
A +10 defender at shortstop is not worth the same as a +10 defender at first
There are always exceptions (You couldn’t really expect Troy Glaus to be a neutral centerfielder, or for Grady Sizemore to play a decent shortstop while throwing lefthanded) but for the most part this is a great shorthand way to determine a players value when switching positions.
He hasn’t played every day in a long time and he’s getting older. I don’t want to speculate about what playing every day would do to his defense down the stretch, but last time he was a regular he was an average CF. So +10 it is.
Eh, since I'm here
I’ll go with 10 runs above average.
ThundaPC - January 28, 2009
Samesies
kentroyals5 - January 28, 2009
Wonder where this could possibly be going?
DCMariner - January 28, 2009
Wow, you people are incredibly high on his Defense.
Is he really that good at defense, because I was thinking around 5ish runs.
joof - January 28, 2009
In ~1650 IP at LF/RF, his UZR is about +20 runs per 150G
Matthew - January 28, 2009
Well I be damned.
Thats pretty damned good.
joof - January 28, 2009
That's what I voted, +20
I was wondering if I wasn’t crazy when I saw the results.
esoteric - January 28, 2009
So...
…my un-analytical theory that defensive metrics WAAAAY overvalue players (i.e. I don’t trust them enough to feel confident in assuming their “value” values are reliable) forces me to adjust this, and assuming Endy Chavez is “really good” at defense, I’ve given him a +10.
PositivePaul - January 29, 2009
That means he's an average centerfielder.
JI - January 29, 2009
To believe this is true
You have to think that run prevention is about 98% pitching and 2% defense. Just think about it for a second – you’re saying that the very best defensive outfielder in the game (or one of the top two or three, at least), only catches about 10 balls per 162 games more than the average of some the very worst defensive outfielders in the game.
Of course, we know there’s a much larger variance than that between how many balls each of them catch, so you’re saying that you believe that all of those extra fly balls that the good defenders catch and the bad defenders don’t are due to the pitcher. You’re essentially throwing DIPS theory out the window.
Just so you’re aware of the consequences of your skepticism. It’s one thing to not believe that defensive statistics do a great job of separating great players from good players. It’s another thing entirely to believe that defense just doesn’t matter very much, which is exactly what you’re stating that you believe.
davidcameron - January 29, 2009
That's not what I'm saying...
Defense does matter, and I do believe that. I’m just not confident enough in the measurements to say how much it matters.
PositivePaul - January 29, 2009
And I thought the "+10" equated to run values, not play values.
Because I’m not confident enough in the metrics to be able to accept the translation of “catching 10 balls per 162 games more than average” (i.e. making 10 more plays) into “preventing 10 more runs than average.”
But that’s my personal opinion, however wrong it may be, and not my analysis. I leave the analysis to those more interested and capable…
PositivePaul - January 29, 2009
On a team level
It’s pretty obvious. We know how many balls in play each team has a chance to catch. We know how many of those balls are caught and how many are not. We know how many runs each type of play made (or not made) is worth. These aren’t things that are really questionable. These are just facts.
Factually, the gap between good and average teams is much, much larger than you’re suggesting. The only way for that to be true is for the gap to be heavily influenced by the pitchers, not the defenders.
It might not be the conclusion you want to be drawing, but it’s the only one that works. If you think a great defender is, at most, +10 runs above average compared to an LF, then you think most of the difference in balls in play being turned into outs are due to the pitchers.
davidcameron - January 29, 2009
Wait.
Okay – this data may not be questionable (although the run value assignment on these plays is theory NOT fact – a subtle but important difference). But WHY these balls were caught (and, conversely, why the other balls WERE NOT) is a lot more subject to debate, and definitely not reliably measurable quite yet. Is it because of the pitchers? the defenders? the coaches? the hitters? the weather? the ballpark? the time of day? the field conditions? the league? the “WFB Grit and Hustle Inverse Subjunctive Quotient ®”? Who knows. There’s been some great work done on this and I know it will continue.
I’m an idiot with regards to this stuff, though, and I will submit my lack of knowledge to your (and others’) overlordship…
PositivePaul - January 29, 2009
The run values are facts
Not a theory. Calculating run values isn’t even that complicated. I can’t imagine a valid critique of them.
As for the rest of it, well, I’ll just say that the evidence isn’t in your favor and leave it at that.
davidcameron - January 29, 2009
Chavez wouldn't be worth anything is he was only +5 on a corner
JI - January 28, 2009
I voted +15 just to be conservative.
I think +20 is completely reasonable. Chavez is a beast.
Teej - January 28, 2009
+20
I’m combining the runs he’s going to save himself plus the runs he’ll allow Gutierrez to save by not having to worry as much about left-center field.
katal - January 28, 2009
Because he's always going to be worried about covering that slacker Ichiro's ass in right
seattlebruin - January 29, 2009
Now that we don't have gritty veterans like Bloomquist or Ibanez to push him.
Fin - January 29, 2009
Though to be fair, it's entirely possible that Ichiro starts to lose a step as he ages
seattlebruin - January 29, 2009
+10
He WAS otherworldly, but he’s turning 31 in a few days, and defensive peak comes earlier than offensive peak. He’ll be good, but I don’t think it’s reasonable to expect that he maintain his career average anymore.
marc w - January 28, 2009
I agree
+10 runs is an amazing defender. Predicting a guy to be above +10 runs should only be done for players like Adam Everett in his prime I think.
Edgar for Pres - January 28, 2009
I'm under the impression that +10 corner
would roughly convert to 0 CF, and I think he’s still plus a handful in centerfield.
DCMariner - January 29, 2009
Conversion table
C: 12.5
SS: 7.5
2/3/CF: 2.5
LF/RF: -7.5
1B: -12.5
DH: -22.5
So, yes you were close.
JI - January 29, 2009
I'm not so sure you can do that with the positional adjustments
by your logic a shortstop would make a -5 defensive catcher
Bearskin Rugburn - January 29, 2009
This isn't my system
Catcher is a different animal and it requires a different set of skills than the other seven positions.
JI - January 29, 2009
All the same
those values don’t reflect differences in defensive performance following a position switch.
Bearskin Rugburn - January 29, 2009
Well, no, defensive performance stays the same
A +10 defender at shortstop is not worth the same as a +10 defender at first
There are always exceptions (You couldn’t really expect Troy Glaus to be a neutral centerfielder, or for Grady Sizemore to play a decent shortstop while throwing lefthanded) but for the most part this is a great shorthand way to determine a players value when switching positions.
JI - January 29, 2009
He was average in center back when he played it in canada
Bearskin Rugburn - January 29, 2009
We have data on him
Historically he is a +12 LF/RF and +6 in CF. This includes data when he was a younger man. Knock him down a few and you get +10 runs.
I think he’s about average in CF (within a 5 run window). Being above average in CF is pretty hard with the competition so high.
Edgar for Pres - January 29, 2009
+25. Crap, I'm still not reading UZR ratings correctly.
Kermit. - January 29, 2009
I need a tutor.
royalcurve - January 29, 2009
+10
He hasn’t played every day in a long time and he’s getting older. I don’t want to speculate about what playing every day would do to his defense down the stretch, but last time he was a regular he was an average CF. So +10 it is.
Bearskin Rugburn - January 29, 2009
You must Login with your SB Nation account and be a member of Lookout Landing to post a comment.