BA: Adam Jones 2nd best defensive OF in the AL

[CA-ORIOLE]

Are the FB's over his head due to positioning or poor reads? I have often seen him start in on a ball, when it is over his head. That has nothing to do with his psositioning. They have positioned him deeper in general to try an accomadate for this shortcoming. I just don't think he is all that good at judging balls off the bat.

Could be a combination of both. I edited my post some, but I was thinking his neg rating may be inflated by the balls over his head being weighted more heavily than balls in front of him. Since you have a Bill James account what are his short/deep splits this year?

[El Gordo]

But don't we have one of the worst pitching staffs in the game? Shouldn't that play into this evaulation? If he played back more to catch those balls, won't more balls fall in front of him?

So our SP gives up 20% more LD's at 101 mph to vector B, but 70% of all CF make a play on this ball, while Jones makes it 40% of the time. What difference does the SP make?

[CA-ORIOLE]

But don't we have one of the worst pitching staffs in the game? Shouldn't that play into this evaulation? If he played back more to catch those balls, won't more balls fall in front of him?

No, theoretically it shouldn't. More chances shouldn't make him less effective. More chances should make the numbers more stable.

[AgentOrange]

So our SP gives up 20% more LD's at 101 mph to vector B, but 70% of all CF make a play on this ball, while Jones makes it 40% of the time. What difference does the SP make?

Gotta know how far each CF is going to make the play. His problem could be positioning. That could be a coaching problem. It could be players problem.

[El Gordo]

Won't Field F/X be able to show exactly how Jones reacted once the ball was put into play? Do you think that will show more defensive lapses in players?

So what? What matters is if he gets to the ball or not. Why is this so hard for you to grasp. The only thing that ultimately matters in comparing one players defense with another's is his effectiveness. I think you are talking about innate talent and not defensive effectiveness. I don't care how talented a player is, I care about how well he does his job.

[AgentOrange]

So what? What matters is if he gets to the ball or not. Why is this so hard for you to grasp. The only thing that ultimately matters in comparing one players defense with another's is his effectiveness. I think you are talking about innate talent and not defensive effectiveness. I don't care how talented a player is, I care about how well he does his job.

You are being dense. You are making this a black and white concept when it isn't one.

[El Gordo]

Gotta know how far each CF is going to make the play. His problem could be positioning. That could be a coaching problem. It could be players problem.

This is a stretch. There may be some coaching issues, but I doubt they would have that much of an impact. Coaching doesn't vary all that much from team to team in this regard.

[SrMeowMeow]

I think that is false. Its not that black and white. I think it does matter how far a player runs to get to the ball. You can't call a players positioning wrong based on where the ball was hit. They might be in the position based on hitters tendancies or how the pitcher is performing that day, or even the pitch itself. It a pitcher throws a mistake pitch and allows a dead pull hitter to hit it into a gap in the D, the fielder shouldn't be penalized. Knowing how far the player had to travel will give the data collectors a better indication on how routine the play was.

This is the correct perspective.

[CA-ORIOLE]

You are being dense. You are making this a black and white concept when it isn't one.

I don't think he's being dense (not in this case anyways). Ultimately the goal should be to see how the player performed relative to other players at the same position. Without considering position neutral I don't know how you would do that. If field FX can identify poor positioning/coaching, that's great and I agree it provides us more insight. Take the data and use it to make the corrections.

[KAZ97]

Let's discuss a less controversial topic, like whether there is a God.

Ultimately what we care about is the number of runs the Orioles would allow if Adams Jones played CF compared to how many runs the Orioles would allow if someone else (perhaps contrary to fact) played CF. This is a classic case of the counterfactual problem.

We can get a (somewhat) accurate estimate of the number of runs the Orioles would allow with Jones in center because Jones does play center for the Orioles.

Who should we contrast Jones to? The "average" fulltime centerfielder? All centerfielders in the AL?

Once that is decided, the problem is estimating the amount of runs the Orioles would allow with the "other" centerfielder. The problem is that the "other" centerfielders don't play for the Orioles, so we can only make guesses as to how they would perform behind our pitchers, with our coaches etc.

I believe using data to help make those guesses is appropriate. The existence of God, I'm on the fence.

[CA-ORIOLE]

Ultimately what we care about is the number of runs the Orioles would allow if Adams Jones played CF compared to how many runs the Orioles would allow if someone else (perhaps contrary to fact) played CF. This is a classic case of the counterfactual problem.

We can get a (somewhat) accurate estimate of the number of runs the Orioles would allow with Jones in center because Jones does play center for the Orioles.

Who should we contrast Jones to? The "average" fulltime centerfielder? All centerfielders in the AL?

Once that is decided, the problem is estimating the amount of runs the Orioles would allow with the "other" centerfielder. The problem is that the "other" centerfielders don't play for the Orioles, so we can only make guesses as to how they would perform behind our pitchers, with our coaches etc.

I believe using data to help make those guesses is appropriate. The existence of God, I'm on the fence.

He's compared to all other center fielders in baseball on a relative scale. I can think of no better way. There certainly are unknowns and some questionable variables, but they are no where in the league of guesses.

[SrMeowMeow]

This is a stretch. There may be some coaching issues, but I doubt they would have that much of an impact. Coaching doesn't vary all that much from team to team in this regard.

The point is, you simply have more information with Field f/x. For example, positioning is to some degree luck. If Adam Jones plays a heavy shift and catches a screaming line drive without taking a step while Brett Gardner catches the exact same ball after a perfect fire step, long run and a diving catch, UZR gives them both the same credit for the play. Yes, they both made an out on an identical ball in play, but Gardner demonstrated more skill and should get more credit towards being a better fielder.

[ledzepp8]

Since it is not as accurate as offensive stats, one season numbers aren't as usefull as two or three seasons worth of info.

I've never understood this statement. Why are one season's numbers not useful, but two or three season's numbers are?

[El Gordo]

The point is, you simply have more information with Field f/x. For example, positioning is to some degree luck. If Adam Jones plays a heavy shift and catches a screaming line drive without taking a step while Brett Gardner catches the exact same ball after a perfect fire step, long run and a diving catch, UZR gives them both the same credit for the play. Yes, they both made an out on an identical ball in play, but Gardner demonstrated more skill and should get more credit towards being a better fielder.

Maybe, but as Drungo said, it's not about style points. On one play Gardner may demonstrate more skill, but on 1200 innings worth of plays the coaching/ positioning factor is not likely to have that much of an impact. If Gardner played for the O's he'd still be among the to OF's in the AL.

[SrMeowMeow]

Maybe, but as Drungo said, it's not about style points. On one play Gardner may demonstrate more skill, but on 1200 innings worth of plays the coaching/ positioning factor is not likely to have that much of an impact. If Gardner played for the O's he'd still be among the to OF's in the AL.

Sample size issues are the single biggest downside to UZR, so you're actually exactly wrong. UZR takes three full seasons of data to fully stabilize, which is almost 4000! innings. And one of the reasons it is so variable is because of the very points AgentOrange made and I highlighted.