Why wouldn't defensive performance vary?Originally Posted by edabbs44
Why wouldn't defensive performance vary?
Last edited by jojo; 08-13-2013 at 11:59 AM.
"This isn’t stats vs scouts - this is stats and scouts working together, building an organization that blends the best of both worlds. This is the blueprint for how a baseball organization should be run. And, whether the baseball men of the 20th century like it or not, this is where baseball is going."---Dave Cameron, U.S.S. Mariner
My initial thought would be that it would vary some (like anything) but quite a bit less than hitting and pitching. There are so many fine adjustments required to hit a baseball, or throw a baseball past a hitter. One small adjustment and your whole mechanics could change.
Fielding on the other hand I would think is much more in control of the player. Speed and athletecism change marginally from year to year, and day to day unless there is a significant injury. These aren't skills that require constant adjustments to stay at the top of their game. If a SS knows how to position themselves, read balls off bats, I think they more or less do so as a consistent function of their skillset rather than possess the ability for only a certain percentage of the time.
Generally when a good fielder makes an error, there is a reason (ie. little time to act, ball lost in sun, ball take bad bounce). It's very rarely just a complete goof or something. Good fielders generally make the plays they have the ability too, whereas, good hitters often fail because of the complexity that foes into an at-bat.
There is alot more to playing great defense than speed and athleticism.
Also, it doesn't take a significant injury to impact an ability to play defense. One can tweak a hammy and still hit, even hit very well.
"This isn’t stats vs scouts - this is stats and scouts working together, building an organization that blends the best of both worlds. This is the blueprint for how a baseball organization should be run. And, whether the baseball men of the 20th century like it or not, this is where baseball is going."---Dave Cameron, U.S.S. Mariner
I could imagine that fielding could be significantly affected by a relatively minor injury. If a guy is nursing any sort of a leg or core injury, his range is likely to decrease quite a bit. An injury that may not affect the guy at the plate could affect him in the field -- e.g. turf toe on a batter's front. You also have the mental part of how aggressively an guy approaches balls hit at the edge of his range; one particularly good or bad experience (say a great diving catch or hitting the wall) could result in him changing his approach.
I also think there's more technique in fielding than many people are giving credit for, especially in terms of footwork. Getting a good first step is more than having quick reflexes. Going back on a ball hit over your head is more than simply having the speed to run in down. Technique matters.
You're right that better positioning can offset loss in range, but something similar could be said on offense in terms of plate discipline vs. reaction time.
Sure, there's less inherent variability in defenders' conversation of a given fielding chance vs. a given batter's plate appearance, but we should be careful in oversimplifying defensive performance (remember, we're talking performance, not ability) as simply a function of athleticism.
Do we know how much variability this all adds up to? Nope. And that's the point. We react when we see variability in fielding stats as if we had really good, really trustworthy intuition and I'm not clear on why we should have so much faith in our intuition to get the scale of variability right given what we know about cognitive biases and the limits of our observation powers.
I'm actually much more convinced by critiques of defensive performance that show different approaches leading to vastly different results than I am critiques of the measured variability over time.
Last edited by RedsManRick; 08-13-2013 at 12:52 PM.
Games are won on run differential -- scoring more than your opponent. Runs are runs, scored or prevented they all count the same. Worry about scoring more and allowing fewer, not which positions contribute to which side of the equation or how "consistent" you are at your current level of performance.
I wouldn't think that technique and such would change much from play to play. For example, guys like Brandon Phillips appear to bring a consistent, standard approach from play to play.
If UZR were more consistent would you guys be making this defensive variability argument? Seems like a classic case of making the facts fit the experiment.
Yes.
Actually assuming that defensive performance doesn't vary is counterintuitive. As a premise, it's a quetion that really shouldn't be begged.
I guess players are only robots when asserting such serves to dismiss a sabermetric stat? And yes the alliteration quotient is stellar even after adjusting for park effects.
"This isn’t stats vs scouts - this is stats and scouts working together, building an organization that blends the best of both worlds. This is the blueprint for how a baseball organization should be run. And, whether the baseball men of the 20th century like it or not, this is where baseball is going."---Dave Cameron, U.S.S. Mariner
Batting Triple Crown winners who DID NOT win their MVP award (since 1930)...
(1) 1934, Lou Gehrig
>> One of the biggest WTHs ever. Mickey Cochrane wins b/c he is the leader of the AL champion Tigers.
(2) & (3) 1942 and 1947, Ted Williams
>> Hated the writers and they hated him. Enough said.
You win the triple crown, then you should win the MVP. It has happened 17 times in 140+ years. About 1 time per decade on the average. 1 time in the last 45 years. Same goes for a .400 batting average (god forbid BA gets mentioned!).
Sorry. No stat, UZR or anything else will convince me otherwise. Sure there are always "other variables" (position in order, who surrounds you in lineup, etc), but this is an accomplishment that most of us have seen done one time in our lives. ONE TIME in your life (except old-timers LOL). Say that over-and-over (one time in my life, one time in my life). Not often something happens once in your life in MLB.
I don't think anyone is arguing that it doesn't vary whatsoever or that players are robots.
The argument is that it is more controlled and less variable than hitting and pitching. I have no idea why that question shouldn't be begged.
Brutus (08-13-2013),Red in Chicago (08-13-2013)
Because it's a premise that is central to the conclusion and its a premise that is neither intuitive nor consistent with every other aspect of player performance that can be measured?
"This isn’t stats vs scouts - this is stats and scouts working together, building an organization that blends the best of both worlds. This is the blueprint for how a baseball organization should be run. And, whether the baseball men of the 20th century like it or not, this is where baseball is going."---Dave Cameron, U.S.S. Mariner
To me fielding doesn't slump to the extend that hitting will. Fielding for the most part stays consistent. The variables that make up a good fielder or a poor fielder are pretty much God given. The only things that I can see having a negative impact on fielding are injury, old age, or mental lapses. One of them is pretty evident, old age, one is readily reported in the press, injury, and one is a little murkier but can be on display.
Have you ever heard someone say a guy got great jumps last season but poor ones this season? Have we ever seen an outfielder who one season was great at tracking fly balls down but he next season wasn't? You can practice things like footwork and fielder but IMO the best fielders possess God given traits that can't be duplicated.
If someone sees a pretty significant jump in his defensive metrics there should be a couple of easy explanations. If the player was on the downside of his career and added another year maybe his first step reaction slowed a little bit. Or maybe he lost a step speed wise. If he is in his prime maybe it is a nagging injury, but how many times have we seen someone's performance drop off a cliff and they hide an ailment? If injury is a legitimate excuse for a drop in performance then most of the time it is reported. If in this case Mike Trout has turf toe, do you think he is going to go through the entire season of poor defense without disclosing turf toe?
I think defensive ratings have done a good job in moving the judgement away from errors to a more in depth look at fielder. But as debated a few years ago it was noted that you need 3 years of data to really get a feel for defensive metrics. As a fan and one more tilted towards the traditional stats side of things it irritates me to no end when some of the so called saber "experts" arrogantly force defensive metrics down you throat like they are written in stone truth.
Just to further the topic all off season we heard how both Bruce and Choo profiled in 2012 as below average RF's and one was going to play CF this season. I have no idea how Bruce rates this year in RF, but if it were below average I would be shocked.
It does vary. The KGJ at 36 wasn't the same KGJ at 26. But my intuition says that healthy players in their prime don't go "up" on defense one year, "down" the next, then "up" again. Those kinds of swings are probably better explained by the wild variability of the chances (round ball, round bat, etc...).
A football analogy. If a great wide receiver has his numbers tank because he's playing with Tim Tebow, where would you put the blame? The receiver likely didn't forget how to run his routes.
Lets not drift into hypothetical territory that suggests what your describing is an accurate characterization of systems like UZR. Lets talk specifics.
People who have watched Trout this year generally suggest he's not making plays that he made last year. He seemed to come into camp heavier than his rooky year. UZR is reflecting what the eyes seem to be seeing. Also his sample is spread across multiple positions meaining his sample is actually a composite of smaller ones so UZR is also likely influenced by randomness.
So if a player's defensive numbers are up one year and down the next, could it be him and not his set of chances? Sure. He's human and his defensive game is subject to the same slings and arrows as his bat. In fact just a little decrease in range can impact his defense even if the eyes aren't good enough to pick it up from one season to the next.
There really is no mystery here unless one is arguing that defensive ability really isn't subject to variation and it's not possible for a player with good defensive tools to have a mediocre defensive year. And as Bill James argues, why would one assume that?
Last edited by jojo; 08-13-2013 at 02:14 PM.
"This isn’t stats vs scouts - this is stats and scouts working together, building an organization that blends the best of both worlds. This is the blueprint for how a baseball organization should be run. And, whether the baseball men of the 20th century like it or not, this is where baseball is going."---Dave Cameron, U.S.S. Mariner
The quantitative evidence suggests variability exists. Some are asserting that defensive performance does not vary much, thus the data must be wrong. Their argument is based on what amounts to logic and personal observation. We (Jojo, me and others) are providing a similar argument based on logic and observation to make the case that defensive performance may vary on the scales suggested by UZR data. Frankly, neither camp has a strong enough logical case to be convincing. Normally, that's where we would bring in the available evidence.
Am I asserting with authority that I know defensive metrics to be very accurate? I hope not. What I am asserting is that simply coming up with a story of why defense doesn't vary much does not in any way convince me (nor should it convince anybody, in my view) that the data must be wrong. If there is no burden of proof required, one can easily come up with any number of plausible hypothesis regarding the amount of variability in defensive performance.
My only point real point is that if I have to move forward and either accept or reject the data, as one must do if he/she is to come up with a system of holistic performance measurement, then I think UZR provides us better information than simply ignoring defense or applying some arbitrary defensive performance number based on our gut intuition, ad hoc adjustment to UZR, or any other approach I've seen put forth.
Hopefully, over time, defensive data will get better. But I've yet to see anybody put forth an approach to measuring defense that actually produces a better estimate. I have no more confidence in the accuracy/reliability of a "WAR" measure that ignores defense than one that incorporates data that might not be as reliable as we would wish it to be.
Last edited by RedsManRick; 08-13-2013 at 03:56 PM.
Games are won on run differential -- scoring more than your opponent. Runs are runs, scored or prevented they all count the same. Worry about scoring more and allowing fewer, not which positions contribute to which side of the equation or how "consistent" you are at your current level of performance.
Posting Permissions
|
Board Moderators may, at their discretion and judgment, delete and/or edit any messages that violate any of the following guidelines: 1. Explicit references to alleged illegal or unlawful acts. 2. Graphic sexual descriptions. 3. Racial or ethnic slurs. 4. Use of edgy language (including masked profanity). 5. Direct personal attacks, flames, fights, trolling, baiting, name-calling, general nuisance, excessive player criticism or anything along those lines. 6. Posting spam. 7. Each person may have only one user account. It is fine to be critical here - that's what this board is for. But let's not beat a subject or a player to death, please. |
Reply With Quote