MASN: A Primer on Pitch Framing

[Larry18]

Right, he says there is work to be done but doesn't explain how they are doing it. I have not seen anyone of the current generation explaining how they come up with their numbers.

Sorry I need the curtain to be pulled back before I buy into a system.

Why not contact Shepard through the Camden Depot website?

[Kalkan]

You weren't the wrong person to get snarky. Were is this data base of events coming from. With DRS e.g, I know exactly how the data base is determined. With pitch framing I don't. Before getting hung up on what it is worth, how about deciding what it is. It seems pretty stupid to me to spend millions on the probability of someone framing more pitches than another. How do you determine the likely hood of a framed pitch when you can't determine what it is to begin with? It's baseball not particle physics.

Pitch f/x, trackman, some other system is how the database is created.

[Number5]

There was a pitch to Cruz yesterday that was at his armpits a good 4 or 5 inches inside that was called a strike. Miraculous framing job by McCann.

Somehow, the catchers for certain teams will always score high in pitch "framing". If Wieters signs as a free agent after next season with the Yankees, Red Sox, Braves, or Cardinals, he will suddenly learn to be among the best pitch framers in baseball. Must be the coaching. :laughlol:

Pitch-framing is a catching skill. The higher the level of play, and the more experienced the umpires, the less impact it has. At the major league level, some catchers may be particularly adept at it and get a few extra borderline calls, but it isn't nearly the difference-maker in the majors that some would have you believe. The ball an inch off the plate and an inch below the knee may well be called a strike or a ball by that particular umpire regardless of who the catcher is. A catcher that has a pitching staff consisting of guys with excellent control and command who are always painting the corners is going to score well in "framing".

[Stotle]

He sure did. He then said he had trouble getting picked up by anyone else because they wanted him to have another big find.

That guy who came up with pitch framing never released even the nuts and bolts of his college pitch handling system because that is worth a lot to a team.

The guys who do projection models keep it proprietary. These things have value and it makes sense to hold onto that value if you created it.

This is it. I don't have nearly the level of valuable information that others do and I still only publish around 20% of my draft stuff.

[El Gordo]

Pitch f/x, trackman, some other system is how the database is created.

I don't think you understand the problem. If I look at pitch fx it tells me a ball was called a strike. It doesn't tell me why. I can see when a catcher moves his glove in such a way as to appear to deceive an umpire. But from my point of view watching on TV I can't tell whether the pitch would have been a ball or a strike anyway. It ls all subjective guesswork IMO. If I am referring to a data base of say BABIP, I know how that data base is established. It's pretty obvious. With pitch framing not so much.

[Kalkan]

There was a pitch to Cruz yesterday that was at his armpits a good 4 or 5 inches inside that was called a strike. Miraculous framing job by McCann.

Somehow, the catchers for certain teams will always score high in pitch "framing". If Wieters signs as a free agent after next season with the Yankees, Red Sox, Braves, or Cardinals, he will suddenly learn to be among the best pitch framers in baseball. Must be the coaching. :laughlol:

Pitch-framing is a catching skill. The higher the level of play, and the more experienced the umpires, the less impact it has. At the major league level, some catchers may be particularly adept at it and get a few extra borderline calls, but it isn't nearly the difference-maker in the majors that some would have you believe. The ball an inch off the plate and an inch below the knee may well be called a strike or a ball by that particular umpire regardless of who the catcher is. A catcher that has a pitching staff consisting of guys with excellent control and command who are always painting the corners is going to score well in "framing".

Those Brewers teams sure get all the breaks, eh?

Salty and Laird scored poorly last year.

If there was team bias, again this would pretty easily show up in the database. Same is true with pitching staffs. With several years of data, events should be pretty easy to isolate.

[SerenityNow]

There was a very interesting interview on MLB network with Andrew Zimbalist a few months ago. Toward the end of the interview he states that in his opinion the best way to judge a defender is to use your eyes. I'm not saying he is right or wrong but I found it interesting.

https://www.dropbox.com/s/knlsk5au34seqok/Clubhouse_Confidential_1-28-14.m4v

[El Gordo]

There was a pitch to Cruz yesterday that was at his armpits a good 4 or 5 inches inside that was called a strike. Miraculous framing job by McCann.

Somehow, the catchers for certain teams will always score high in pitch "framing". If Wieters signs as a free agent after next season with the Yankees, Red Sox, Braves, or Cardinals, he will suddenly learn to be among the best pitch framers in baseball. Must be the coaching. :laughlol:

Pitch-framing is a catching skill. The higher the level of play, and the more experienced the umpires, the less impact it has. At the major league level, some catchers may be particularly adept at it and get a few extra borderline calls, but it isn't nearly the difference-maker in the majors that some would have you believe. The ball an inch off the plate and an inch below the knee may well be called a strike or a ball by that particular umpire regardless of who the catcher is. A catcher that has a pitching staff consisting of guys with excellent control and command who are always painting the corners is going to score well in "framing".

:agree: You sound like someone who actually watches baseball as much as counting it.

[Kalkan]

I don't think you understand the problem. If I look at pitch fx it tells me a ball was called a trike, It doesn't tell me why. I can see when a catcher moves his glove in such a way as to appear to deceive an umpire. But from my point of view watching on TV I can't tell whether the pitch would have been a ball or a strike anyway. It ls all subjective guesswork IMO. If I am referring to a data base of say BABIP, I know how that data base is established. It's pretty obvious. With pitch framing not so much.

I don't think the system should care why. It is not explaining why a catcher tends to get strikes or does not get strikes. Like nearly all statistics, it is not a learning tool. All that matters to it is where a ball is located and how the other variables are set up (e.g., batter, ump, pitcher, pitch type). There is no real guesswork here.

[Kalkan]

There was a very interesting interview on MLB network with Andrew Zimbalist a few months ago. Toward the end of the interview he states that in his opinion the best way to judge a defender is to use your eyes. I'm not saying he is right or wrong but I found it interesting.

https://www.dropbox.com/s/knlsk5au34seqok/Clubhouse_Confidential_1-28-14.m4v

I agree with that, but it is nice not to have to rely on one tool. I mean, if your eyes tell you one thing and the metrics all say the opposite. That is a situation where you really need to challenge yourself on or simply focus on someone else where there is more agreement.

[CA-ORIOLE]

:agree: You sound like someone who actually watches baseball as much as counting it.

Well you two are Birds of a feather. That's for sure.

[El Gordo]

I don't think the system should care why. It is not explaining why a catcher tends to get strikes or does not get strikes. Like nearly all statistics, it is not a learning tool. All that matters to it is where a ball is located and how the other variables are set up (e.g., batter, ump, pitcher, pitch type). There is no real guesswork here.

But that is the principle question. You can't have a formula based on an event, if you can't measure the event. Say you have an umpire who calls 6 % of CB outside, strikes. How do you know how much of that 6 % is due to his eyesight, how much is due the catcher's slight of hand or the pitcher's deception. None of these events occur in isolation.If you are using that number to filter out and isolate pitch framing you already have a questionable number. I think a bunch of enthusiastic stat boys are creating something out of almost nothing and trying to sell it.

[CA-ORIOLE]

But that is the principle question. You can't have a formula based on an event, if you can't measure the event. Say you have an umpire who calls 6 % of CB outside,. strikes. How do you know how much of that 6 % is due to his eyesight or how much is due the catcher's slight of hand. If you are using that number to filter out and isolate pitch framing you already have a questionable number. I think a bunch of enthusiastic stat boys are creating something out of almost nothing and trying to sell it.

Every team throws more more than 20,000 pitches over a year and umpires move around. The pitch/event is measured by pitch f/x and other tracking tools. The umpires have track records, the pitchers have track records and the catchers have track records. What is so difficult to comprehend?

[El Gordo]

Every team throws more more than 20,000 pitches over a year and umpires move around. The pitch/event is measured by pitch f/x and other tracking tools. The umpires have track records, the pitchers have track records and the catchers have track records. What is so difficult to comprehend?

How do you determine who gets credit for the call? Why is that so difficult to comprehend? Umpire calls x % OZ pitches strikes. Pitcher gets y % OZ pitches called strikes. Catcher gets z % OZ pitches called strikes. How do you know which of those is a framed pitch? Emperors new clothes IMO.

?

[Kalkan]

How do you determine who gets credit for the call? Why is that so difficult to comprehend? Umpire calls x % OZ pitches strikes. Pitcher gets y % OZ pitches called strikes. Catcher gets z % OZ pitches called strikes. How do you know which of those is a framed pitch? Emperors new clothes IMO.

?

You are looking for differences. That is the whole approach. If an Ump has skill, then his skill follows him. If a pitcher has skill, that skill follows him. If a catcher has skill, that skill follows him. If the ump is calling balls strikes in a lower left portion at a higher degree, that should pop up for all catchers.

You basically can account for variables and normalize for them.

There is a degree of uncertainty there of course, but the repeatability of the numbers reported illustrates that skill is being measured.