Does this make sense based on what you've seen this year?

[Stotle]

Maybe, but I have my doubts. I do think it will increase stability/accuracy. The stability will never be near the level of offense though. We'll have more data available (assuming it's going to be disseminated).

In the end, Field FX is a means to collect more precise data and not an evaluation system itself. DRS has been using Field FX timed data for a couple years now. They can measure the speed and location of a ball much more accurately than UZR from my understanding. The UZR and DRS systems are still "basically" the same though with an assumed neutral positioning and defined zone aspect. DRS has more nuances. Maybe they'll adjust those basic systems when the full Field Fx packages come on line or come up with a new one, but I haven't heard anything about such an undertaking.

The positioning aspect (particularly the increased IF overshifts) are petty valid concerns imo. I think it's less of a concern for the OF as hitters generally hit better to all fields to the OF than the IF. Though the overshifts are "accounted" for as COC said, it has to cause some issues with the individual credit as they are becoming almost standard alignments nowadays. The accounting may have start to shift more toward the team and the pitchers than the individual fielder for credit. Which leads to a whole other issue of how pitching and fielding interrelate and how do you assign such credit and if this even possible to measure or extract. The Cardinals for example are one of the few teams that rarely overshift. They say they would rather not have their pitchers burdened by the overshift as it constrains their ability to cover the plate. Hard to argue with their success on the pitching front.

It's more advanced than the Field FX system that never came to pass. I love it -- it's going to show who takes the best routes, who gets the quickest jumps, who covers the most ground, who has the best arm, who has the most accurate arm, which stadiums provide the biggest challenges for outfielders, what are the "true" ranges for infielders, and so on. It'll also tell us who gets the best jumps on the bases, who takes the best lines, who's the quickest out of the box, which pitchers disrupt base runners most effectively, etc.

The sad thing is I doubt the full body of data is made public for some time. Teams that crunch numbers are going to have a field day with this, and the teams that are the best at isolating the most important variables in determining defensive production are going to have a pretty accurate list of the most valuable defenders (and defensive profiles) in the game.

Advance scouts are going to be reduced to "what are these players doing now" status in short order. Even then, the data is going to be able to tell most of the story.

[CA-ORIOLE]

It's more advanced than the Field FX system that never came to pass. I love it -- it's going to show who takes the best routes, who gets the quickest jumps, who covers the most ground, who has the best arm, who has the most accurate arm, which stadiums provide the biggest challenges for outfielders, what are the "true" ranges for infielders, and so on. It'll also tell us who gets the best jumps on the bases, who takes the best lines, who's the quickest out of the box, which pitchers disrupt base runners most effectively, etc.

The sad thing is I doubt the full body of data is made public for some time. Teams that crunch numbers are going to have a field day with this, and the teams that are the best at isolating the most important variables in determining defensive production are going to have a pretty accurate list of the most valuable defenders (and defensive profiles) in the game.

Advance scouts are going to be reduced to "what are these players doing now" status in short order. Even then, the data is going to be able to tell most of the story.

Oh yeah, I get it. Whatever they are calling it now, it's going to be an an abundance of great data. I just don't have a feel of how that data is going to be incorporated into a rating system and/or into the two current primary advanced rating systems like UZR and DRS. For example, how are they going to balance the current positioning centric based systems and incorporate the range/athletic ability aspects?

[Stotle]

Oh yeah, I get it. It's going to be an an abundance of great data. I just don't have a feel of how that data is going to be incorporated into a rating system and/or into the two current primary advanced rating systems like UZR and DRS. For example, how are they going to balance the current positioning centric based systems and incorporate the range/athletic ability aspects?

I think teams are going to create their own proprietary systems. Not sure what the public will do with it. There will be so much more info that even incorporating the data will require a fairly significant overhaul (I'd think).

Player X's jumps save .2 runs a year, his routes cost .7 runs a year, his speed saves 1.2 runs a year, his arm saves .5 runs a year, etc.

It will be cool to be able to say the perfect defender has Player V's reads off the bat/jumps, Player W's speed, Player X's routes, Player Y's arm strength, and Player Z's arm accuracy. It may also help us to figure out what exactly is causing certain players not to get to balls (player X has been rating below average as a defender because he adds 15% of distance, on average, to his routes due to bad lines).

[CA-ORIOLE]

I think teams are going to create their own proprietary systems. Not sure what the public will do with it. There will be so much more info that even incorporating the data will require a fairly significant overhaul (I'd think).

.

Right, which is why I said think all this info may not clarify anything at this end of the spectrum. It's probably going to make things more confusing.

Player X's jumps save .2 runs a year, his routes cost .7 runs a year, his speed saves 1.2 runs a year, his arm saves .5 runs a year, etc.

It will be cool to be able to say the perfect defender has Player V's reads off the bat/jumps, Player W's speed, Player X's routes, Player Y's arm strength, and Player Z's arm accuracy. It may also help us to figure out what exactly is causing certain players not to get to balls (player X has been rating below average as a defender because he adds 15% of distance, on average, to his routes due to bad lines)

Right, but that doesn't account for positioning. It disregards it. Maybe that's the answer. Maybe you combine both aspects somehow. Maybe positioning is assigned to the team box and neutralzied. I just don't know.

[El Gordo]

I think teams are going to create their own proprietary systems. Not sure what the public will do with it. There will be so much more info that even incorporating the data will require a fairly significant overhaul (I'd think).

Player X's jumps save .2 runs a year, his routes cost .7 runs a year, his speed saves 1.2 runs a year, his arm saves .5 runs a year, etc.

It will be cool to be able to say the perfect defender has Player V's reads off the bat/jumps, Player W's speed, Player X's routes, Player Y's arm strength, and Player Z's arm accuracy. It may also help us to figure out what exactly is causing certain players not to get to balls (player X has been rating below average as a defender because he adds 15% of distance, on average, to his routes due to bad lines).

In other words they develop a defensive profile similar to an offensive profile? E.g. First step 5 R, Arm 5 R, Range 10 R positioning 5 R. total 25 R defender?

[Hallas]

At least regarding last season, there is a bit of evidence in the new Inside Edge fielding report that either UZR is really skewed in the way it is counting Nick's missed plays, or there is something missing in Inside Edge that I am not getting.

Nick:

1-10% 0.0% (5)

11-40% 40.0% (5)

41-60% 100.0% (6)

61-80% 83.3% (18)

81-100% 99.7% (289)

League average:

1-10% 7.8 %

11-40% 31.9 %

41-60% 59.4 %

61-80% 83.4 %

81-100% 99.1 %

Nick is way short on very difficult plays, but in such a sample size that a single made play in that category would put him well above average. Either Nick is good at making difficult and low run-value plays, or there is something wrong with UZR, or the buckets are simply too small for us to glean any useful data even after a full season's worth of data.

[Hallas]

No. It is all tracked. They know the exact percentage of right fielders that make the play via math.

To give an example of what they have the technology to do...

http://mlb.mlb.com/news/article/mlb/mlbam-introduces-new-way-to-analyze-every-play?ymd=20140301&content_id=68514514

I wouldn't call it math. Physics? Computer engineering?

[CA-ORIOLE]

At least regarding last season, there is a bit of evidence in the new Inside Edge fielding report that either UZR is really skewed in the way it is counting Nick's missed plays, or there is something missing in Inside Edge that I am not getting.

Nick:

1-10% 0.0% (5)

11-40% 40.0% (5)

41-60% 100.0% (6)

61-80% 83.3% (18)

81-100% 99.7% (289)

League average:

1-10% 7.8 %

11-40% 31.9 %

41-60% 59.4 %

61-80% 83.4 %

81-100% 99.1 %

Nick is way short on very difficult plays, but in such a sample size that a single made play in that category would put him well above average. Either Nick is good at making difficult and low run-value plays, or there is something wrong with UZR, or the buckets are simply too small for us to glean any useful data even after a full season's worth of data.

That's where I'd look first. Based up the Fielding Bible data that I have seen in the past that would seem to be Jones' primary issue. It does strike me as odd with Nick because he usually plays fairly deep, but laterally he's fairly weak imo. Also UZR makes Field adjustments where DRS doesn't.

[CA-ORIOLE]

I wouldn't call it math. Physics? Computer engineering?

How about applied math?

[El Gordo]

I wouldn't call it math. Physics? Computer engineering?

Since I understand how they do it I would say pretty simple math.

[Stotle]

Right, which is why I said think all this info may not clarify anything at this end of the spectrum. It's probably going to make things more confusing.

Right, but that doesn't account for positioning. It disregards it. Maybe that's the answer. Maybe you combine both aspects somehow. Maybe positioning is assigned to the team box and neutralzied. I just don't know.

I don't think positioning matters if you have trajectory, speed, etc. of the player and the ball. The need for "zones" goes away and you're left with the question of whether an average defender would have been able to get to the ball based on the type of ball struck and the starting point of the player.

[Stotle]

In other words they develop a defensive profile similar to an offensive profile? E.g. First step 5 R, Arm 5 R, Range 10 R positioning 5 R. total 25 R defender?

Exactly. It will be nuanced, and I think teams likely will differ as to what they choose to assign value, and the total makeup of a player's profile, but ultimately I think the end product will be very useful.

To me, the tricky (and awesome) part is applying this to 1) in-game management, and 2) player valuation/acquisition. What skills can be taught? What deficiencies can be managed around (or on the flip side, exploited)? What aspects of a profile improve over the course of a career and which deteriorate?

I love the idea of evaluating players with the human eye and picking out information that can be used to help place a value on a player or plan for a game against a particular pitcher/team. But I think this stuff is going to make MLB-level scouting all but obsolete. It will still be necessary at the minor league level, and certainly at the amateur level. But if I'm a team I'd look into getting this stuff installed in all of my minor league facilities. Even if the cost is a couple million a year, the info is so incredibly useful I'd pay that amount in a second.

EDIT -- back to the original point, it may not break down as a run-per-trait matter -- will depend on if those traits can be singled out as to their impact on whether a play was made or not made, but I would think you could do it (e.g. he didn't make the play, the play cost .2 runs, he didn't make the play because he took a below average route and had a below average jump). How teams determine which portion of the value gets assigned to each trait will be interesting, and I'm sure a wide variety of approaches will be taken.

I'm really curious to see what kind of an impact this will have.

[CA-ORIOLE]

I don't think positioning matters if you have trajectory, speed, etc. of the player and the ball. The need for "zones" goes away and you're left with the question of whether an average defender would have been able to get to the ball based on the type of ball struck and the starting point of the player.

Maybe that's the answer then. I'm not sure I see the current systems moving that way. That said, it certainly helps to separate the positioning / pitching aspects from the fielders capabilities. Not sure Cal Ripken would have done nearly as well on that system.

[Stotle]

Maybe that's the answer then. I'm not sure I see the current systems moving that way. That said, it certainly helps to separate the positioning / pitching aspects from the fielders capabilities. Not sure Cal Ripken would have done nearly as well on that system.

That's a great point. Although I think Cal would have scored very well on accuracy of first step and shortest path to ball. Less so on total range (max ground covered) and speed.

[Hallas]

Since I understand how they do it I would say pretty simple math.

It's image recognition and vector calculations right? Some math/physics, some computer engineering, right?