Request for defensive catching metrics

[BIG17EASY]

Originally Posted by RchW

I've read some commentary that the ESZ makes some bizarre strike calls on pitches which are clearly outside the zone. Does the ump giving signals have the right to overrule?

Wouldn't the ump overruling the computer defeat the purpose of the computer?

You might be referring to this video that made the rounds a little while back. MLB should have better technology than the Atlantic League has, but I also think we'll have to adjust our idea of the strike zone. The stupid box that's shown on television broadcasts conditions us to view the strike zone as a flat, two-dimensional rectangle. In truth, the strike zone is a three-dimensional rectangle and the computer is going to pick up pitches that clip the very front or very back (and occasionally the sides) of that rectangle that we think are way out of the strike zone because of what we're used to seeing displayed on that awful 2-D box.

[BIG17EASY]

Originally Posted by Argonaut

I know historical mode isn't the biggest focus for OOTPD. But even if pitch framing disappears via an electronic strike zone (there's still gamecalling then, but I digress)... framing will still be a factor for any historical game in OOTP. It affects things like the Hall of Fame, which I'm interested in. My Jim Sundberg in my dynasty doesn't have eye-popping WAR totals so I'm sure he won't get in, but what if he's really added a lot more value than the game thinks, because he is a defensive catcher?

But pitch framing wasn't measured by any analytics until the last decade, and game calling is not quantified in any way by current analytics (although it obviously helps things like catcher ERA).

I get your point that OOTP is not rewarding catchers in the WAR statistic for their defensive contributions. I'm just not sure pitch framing and "game calling" are the right factors to look at. For instance, Sundberg put up 25.3 dWAR (per Baseball-Reference) while playing in an era that didn't track pitch framing. So what did he do to earn that 25.3 dWAR, and are those things currently used by OOTP to give defensive value to catchers?

[CBeisbol]

Originally Posted by BIG17EASY

I don't know what this is referring to.



There are no pitch locations in the game. The ones shown in the one game view are only cosmetic. We can't have a stat that tracks something that's not in the game.

Zone Rating is tracked by the game.
Zone Rating, in real life, uses batted ball locations

I don't think, but I'm not sure, that the game tracks batted balls.

So, if they can fake Zone Rating, by, likely, giving better defensive players a higher probability of a high Zone Rating, then they can fake a pitch framing metric in the same way.

The question, then, obviously, is how useful is it if it's faked.

[CBeisbol]

Originally Posted by BIG17EASY

I think we're talking about two different things. It sounds like you're talking about a pitch framing rating, which I don't have an argument against, other than I personally don't feel it's needed because it's already part of the catcher rating.

Yasmani Grandal
He's a very good pitch framer
But not nearly as good at preventing wild pitches and passed balls
Those are two separate skills
OOTP blends them

I'm specifically talking about a pitch framing stat, which is what Argonaut asked for in the post I replied to. There's no way to quantify pitch framing in the game in a statistical sense when there are no pitch locations in the game. That's the point I'm trying to make.

As Argonaut said, and, I believe, showed with a study, this absolutely has an effect on the results of the game, if it has an effect that can be measured, it can be reflected in a stat

[BIG17EASY]

Originally Posted by CBeisbol

Zone Rating is tracked by the game.
Zone Rating, in real life, uses batted ball locations

I don't think, but I'm not sure, that the game tracks batted balls.

So, if they can fake Zone Rating, by, likely, giving better defensive players a higher probability of a high Zone Rating, then they can fake a pitch framing metric in the same way.

The question, then, obviously, is how useful is it if it's faked.

Yes, there are ball locations in OOTP. So Zone Rating is not faked.

Originally Posted by CBeisbol

As Argonaut said, and, I believe, showed with a study, this absolutely has an effect on the results of the game, if it has an effect that can be measured, it can be reflected in a stat

You're misunderstanding what I'm saying. I don't deny that pitch framing is a big aspect of real baseball. But it measures an aspect of real baseball - pitch location - that currently doesn't exist in OOTP. You can't measure something that doesn't exist. And providing a fake measurement, as you allude to above, would be misleading and pointless.

[sabrtoothtiger]

Really the only metric necessary would be a Catcher BB/K ratio. From my understanding this is what the catcher impacts the most.

[CBeisbol]

Originally Posted by BIG17EASY

Yes, there are ball locations in OOTP. So Zone Rating is not faked.

From the link

About Ball Locations
Ball locations are purely cosmetic.

Defenders aren't getting defensive credit for getting to a difficult ball.
Just balls have a higher probability of being an out with better defenders

Same with good defensive catchers
Plate appearances are more likely to end up as strike outs and less likely to end up as walks

It seems like a very similar mechanism to what is used for defensive ratings could be used for pitch framing.

You're misunderstanding what I'm saying. I don't deny that pitch framing is a big aspect of real baseball. But it measures an aspect of real baseball - pitch location - that currently doesn't exist in OOTP. You can't measure something that doesn't exist. And providing a fake measurement, as you allude to above, would be misleading and pointless.

Zone Rating appears to be fake
Though, maybe that's not the best word. I mean, the hits, outs, etc are all fake too...

Maybe better to say "inferred"

[CBeisbol]

Originally Posted by sabrtoothtiger

Really the only metric necessary would be a Catcher BB/K ratio. From my understanding this is what the catcher impacts the most.

Not exactly, because it doesn't account for the quality of the pitcher nor of the park factors (nor league settings if you're comparing across leagues)

[BIG17EASY]

Originally Posted by CBeisbol

From the link

Defenders aren't getting defensive credit for getting to a difficult ball.
Just balls have a higher probability of being an out with better defenders

Same with good defensive catchers
Plate appearances are more likely to end up as strike outs and less likely to end up as walks

It seems like a very similar mechanism to what is used for defensive ratings could be used for pitch framing.




Zone Rating appears to be fake
Though, maybe that's not the best word. I mean, the hits, outs, etc are all fake too...

Maybe better to say "inferred"

Not sure how else I can explain it. Happy posting!

[Matt Arnold]

Originally Posted by CBeisbol

From the link

Defenders aren't getting defensive credit for getting to a difficult ball.
Just balls have a higher probability of being an out with better defenders

Same with good defensive catchers
Plate appearances are more likely to end up as strike outs and less likely to end up as walks

It seems like a very similar mechanism to what is used for defensive ratings could be used for pitch framing.




Zone Rating appears to be fake
Though, maybe that's not the best word. I mean, the hits, outs, etc are all fake too...

Maybe better to say "inferred"

That page refers to setting ball locations on the park, not ball locations in general. They are used in-game.

We do also generate pitch locations, which aren't 100% cosmetic, but would agree they're not really suitable right now for a framing metric.

I wouldn't want to add a catcher framing metric where the catcher literally was given 5% (or whatever) credit for each strikeout, but also obviously I don't want to "lie" in a metric and tell you something that's not true. So yeah, it's a hard problem.

[BIG17EASY]

Originally Posted by Matt Arnold

We do also generate pitch locations, which aren't 100% cosmetic, but would agree they're not really suitable right now for a framing metric.

I didn't know that. I stand corrected. Did this change in the last few versions? Or has there always been pitch locations that mattered in some form?

[Argonaut]

Big Easy: Fangraphs already has stats for framing, and incorporates them into their WAR calculations. See here:

https://blogs.fangraphs.com/war-update-catcher-framing/
https://blogs.fangraphs.com/fangraphs-pitch-framing/

So I'm sure it can be incorporated into OOTP. All it would be is adding a quantifying stat to something that is already happening in the game. It doesn't matter that there's no pitch locations, there's still the affected K/BB rates. I'm envisioning a line occurring in the play-by-play for a called strikeout affected by a defensive catcher: "The catcher really had soft hands there!"

Anyway I'd think that properly assigning value/WAR to the right players is something we can all get behind.

EDIT: From Matt Arnold's above comment perhaps it isn't as simple as I'm thinking. I suspect Catcher Ability is thrown into the blender for a plate appearance outcome, and isn't necessarily the reason for any given strikeout or walk. It could push it over the top in some dice rolls I suppose, but that might be hard to record as a stat.

[RchW]

Originally Posted by BIG17EASY

Wouldn't the ump overruling the computer defeat the purpose of the computer?

You might be referring to this video that made the rounds a little while back. MLB should have better technology than the Atlantic League has, but I also think we'll have to adjust our idea of the strike zone. The stupid box that's shown on television broadcasts conditions us to view the strike zone as a flat, two-dimensional rectangle. In truth, the strike zone is a three-dimensional rectangle and the computer is going to pick up pitches that clip the very front or very back (and occasionally the sides) of that rectangle that we think are way out of the strike zone because of what we're used to seeing displayed on that awful 2-D box.

I meant this but got it backwards and thought it wasn't really on topic.

FWIW though
https://www.closecallsports.com/sear...ic+strike+zone

[BIG17EASY]

Originally Posted by Argonaut

Big Easy: Fangraphs already has stats for framing, and incorporates them into their WAR calculations. See here:

https://blogs.fangraphs.com/war-update-catcher-framing/
https://blogs.fangraphs.com/fangraphs-pitch-framing/

So I'm sure it can be incorporated into OOTP. All it would be is adding a quantifying stat to something that is already happening in the game. It doesn't matter that there's no pitch locations, there's still the affected K/BB rates. I'm envisioning a line occurring in the play-by-play for a called strikeout affected by a defensive catcher: "The catcher really had soft hands there!"

Anyway I'd think that properly assigning value/WAR to the right players is something we can all get behind.

EDIT: From Matt Arnold's above comment perhaps it isn't as simple as I'm thinking. I suspect Catcher Ability is thrown into the blender for a plate appearance outcome, and isn't necessarily the reason for any given strikeout or walk. It could push it over the top in some dice rolls I suppose, but that might be hard to record as a stat.

I know what pitch framing is and how it's used in various metrics IN REAL LIFE. I have never once denied that it's an important metric for measuring a catcher's defensive value.

My point (which Matt clarified) has always been that there are no pitch locations in OOTP that are useful enough to make measuring pitch framing in OOTP a worthwhile endeavor. What you think is happening in the game actually isn't. I explained it every way I could, but apparently I'm not very good at explaining it. Hopefully Matt's post puts this to rest.

[CBeisbol]

Originally Posted by BIG17EASY

I know what pitch framing is and how it's used in various metrics IN REAL LIFE. I have never once denied that it's an important metric for measuring a catcher's defensive value.

My point (which Matt clarified) has always been that there are no pitch locations in OOTP that are useful enough to make measuring pitch framing in OOTP a worthwhile endeavor. What you think is happening in the game actually isn't. I explained it every way I could, but apparently I'm not very good at explaining it. Hopefully Matt's post puts this to rest.

What's happening in the game is that catcher skill rating affects things that are primarily considered part of pitching - walk and strikeout rates.

There's no way, I don't think, for the user to measure this impact in a real game, only through tests like Argonaut did.

The developers could certainly create some stat that reflects how catcher skill impacts run prevention in OOTP.

We can see how defense ratings for other positions impact run prevention through zone rating. But there's nothing comparable for catchers

[Dyzalot]

Originally Posted by BIG17EASY

Yeah, he's looking for a pitch framing stat to help quantify that. I get the point, but OOTP can't have a stat that measures something that's not in the game (pitches outside the strike zone that are called strikes).

I think unless the game engine undergoes a major change to include pitch locations that affect gameplay, we have to accept that pitch framing is part of the catcher rating. I don't have the math behind how much pitch framing affects a catcher's WAR, but I can't imagine it's skewing the number so significantly that we can't tell who the best defensive catchers are in OOTP.

Plus, if the electronic strike zone comes into MLB -- and signs are pointing in that direction -- pitch framing will immediately become a zero factor.

But it could approximate them through its internal pitch framing metric. In other words, if about 10% of a pitcher's war calculation in OOTP should really be going to the catcher then do something where a guy with average rating gets 10% of the credit, a poor catcher gets 5% and an excellent catcher takes 15%. Those are just examples but I think you get the idea. Yeah you won't know based upon any actual outcomes but you could at least parse out how much of the WAR calculation should go to the catcher on average and then extrapolate from there. Again, I was just trying to come up with some sort of compromise that might mostly accomplish what he was trying to achieve.