How Bad Are the Umps?

[Kimmi]

How is it not the same thing? The auto industry went from humans on an assembly line to robots monitored by humans. Accountants went from writing out all their work and doing the calculations themselves to data entry in a spreadsheet.

 

Well, one of the differences is that I got absolutely no enjoyment out of watching auto assembly lines or accountants doing calculations.

[Kimmi]

Now this is interesting. Your previous defense as a jobs program to give Angel Hernandez something to do I am pretty cool with. The Dickensian nightmare of unemployed umps bums me out. But the principle here - that a person should do this job which can be done with technology better, in a manner that would be easy to implement, just because - is harder to go with. Now this is not something like kiosks replacing workers at a restaurant - we are talking about roughly 12% of an umpires workload over a season - it's simplify a human task. Since the home plate umpire will be there, and be solidly busy - I am not sure how idiot-proofing a part of the job violates the spirit of the endeavor.

 

If the umpires were to become unemployed, then that would be a concern too.

 

What you state as 'just because' might seem like it to you, but it's not to me. I would find the game less enjoyable not having a human umpire. The idea of replacing a human with technology in a game that was designed to be played and officiated by humans is, to me, fundamentally wrong.

[Bellhorn04]

I enjoy the umps but only when they're making the right calls. When they blow calls I get angry at them.

[Bellhorn04]

What you state as 'just because' might seem like it to you, but it's not to me. I would find the game less enjoyable not having a human umpire. The idea of replacing a human with technology in a game that was designed to be played and officiated by humans is, to me, fundamentally wrong.

 

This argument doesn't really work because the game was designed so long ago, in a whole different world. No one knows what Abner Doubleday would have thought of the new technology. Maybe he would have liked it.

[Kimmi]

I actually get where the discrepancy lies: You pulled that up from the "About Gameday" part of the mlb.com site, correct?

 

Well what they define as the "ball zone" is actually a zone right outside the 3x3 grid where pitches should be called balls but can realistically be called strikes.

 

In this image:

 

http://mlb.mlb.com/shared/components/gameday/v4/images/about/3d2.jpg

 

You can clearly see the implementation of a "greyed out" area to illustrate this very point.

 

Remember that this discussion started because of the measurement of of umpire accuracy:

https://www.umpirebible.com/blog/?p=542

 

This is an interesting read on the methodology used by people who use pitch/fx to gather strike zone data ( As I have stated before, there is a margin for error applied to the umpire, they call them "true strikes"). Even better, and this is going to be right up your wheelhouse (and MVP's) there's a large-scale discussion on clear umpire biases here. Recommended reading.

 

I have read the studies. I am the Queen of Studies. I know all about the biases. I was arguing years ago about how Mariano received 13% more strike calls than the average pitcher. Biases were not the point of my argument. The point of my argument was whether umpires were given leeway on their error percentage on calls made in a certain "gray zone", and whether most errors were borderline calls or not.

 

And while on the topic of biases, this is a most excellent to time to point out again to the "I don't need stats, I know what my eyes see" crowd that this is precisely why you need stats. But I digress.

[User Name]

If you've read the studies, then you know that "true strike" percentage is taken into account in a lot of the more nuanced studies. In fact, accounting for true strikes, ump's error rate is still significantly close to 14%, reinforcing the notion of the earlier numbers produced. The fact that a significant portion of those judgement errors come from The Matthew effect reinforced the notion that something needs to be done to improve strike zone efficiency.

[Kimmi]

I actually get where the discrepancy lies: You pulled that up from the "About Gameday" part of the mlb.com site, correct?

 

Well what they define as the "ball zone" is actually a zone right outside the 3x3 grid where pitches should be called balls but can realistically be called strikes.

 

In this image:

 

http://mlb.mlb.com/shared/components/gameday/v4/images/about/3d2.jpg

 

You can clearly see the implementation of a "greyed out" area to illustrate this very point.

 

Remember that this discussion started because of the measurement of of umpire accuracy:

https://www.umpirebible.com/blog/?p=542

 

This is an interesting read on the methodology used by people who use pitch/fx to gather strike zone data ( As I have stated before, there is a margin for error applied to the umpire, they call them "true strikes"). Even better, and this is going to be right up your wheelhouse (and MVP's) there's a large-scale discussion on clear umpire biases here. Recommended reading.

 

Back to the topic of the 'gray zone'. There is nothing in this blog that says that umpires are not penalized (ie are given leeway) for incorrect calls made in a gray zone. The study in fact states that most of the errors are borderline, and that the decline in error rate as you move away from the plate is "rapid".

 

That said, I have seen some sites, not all, that will measure umpire accuracy according to what they term the 'called zone' rather than the true zone. In other words, a zone that has become 'accepted' by umpires, batters, and MLB. Is it okay that the interpretation of the zone has changed from the defined strike zone? Probably not, but MLB could fix that by easily enough by enforcing the actual strike zone.

[Bellhorn04]

I'd like to know the % of bad calls that are no-doubters, outside the gray zone, like the 2 against Bradley.

[User Name]

Back to the topic of the 'gray zone'. There is nothing in this blog that says that umpires are not penalized (ie are given leeway) for incorrect calls made in a gray zone. The study in fact states that most of the errors are borderline, and that the decline in error rate as you move away from the plate is "rapid".

 

That said, I have seen some sites, not all, that will measure umpire accuracy according to what they term the 'called zone' rather than the true zone. In other words, a zone that has become 'accepted' by umpires, batters, and MLB. Is it okay that the interpretation of the zone has changed from the defined strike zone? Probably not, but MLB could fix that by easily enough by enforcing the actual strike zone.

 

If it was as easy as you say, they would have done it by now. It's just impossible for the human eye to discern borderline calls with any sort of consistency.

 

As for the study, unless I'm mistaken, under-recognition errors apply "true-strike" (or recognized zone) judgement.

[Kimmi]

This argument doesn't really work because the game was designed so long ago, in a whole different world. No one knows what Abner Doubleday would have thought of the new technology. Maybe he would have liked it.

 

That's fair enough, but I am a traditionalist in that sense. I like the way the game is. Improving the strike/ball calls by replacing the human element with technology has no appeal to me. I suppose I have no real argument besides that. But as I've said many times, it's my preference.

[Kimmi]

If you've read the studies, then you know that "true strike" percentage is taken into account in a lot of the more nuanced studies. In fact, accounting for true strikes, ump's error rate is still significantly close to 14%, reinforcing the notion of the earlier numbers produced. The fact that a significant portion of those judgement errors come from The Matthew effect reinforced the notion that something needs to be done to improve strike zone efficiency.

 

IMO, the true strike percentage is not so much error as it is what has become accepted as the strike zone. MLB has accepted it, so umpires will continue to call it.

 

I am not arguing that bias in calls doesn't exist. If MLB enforced the actual zone and showed the umpires their data on missed calls, they would improve. They have improved since the advent of Pitch/FX.

[User Name]

IMO, the true strike percentage is not so much error as it is what has become accepted as the strike zone. MLB has accepted it, so umpires will continue to call it.

 

I am not arguing that bias in calls doesn't exist. If MLB enforced the actual zone and showed the umpires their data on missed calls, they would improve. They have improved since the advent of Pitch/FX.

 

Slightly improved.

[Kimmi]

If it was as easy as you say, they would have done it by now. It's just impossible for the human eye to discern borderline calls with any sort of consistency.

 

As for the study, unless I'm mistaken, under-recognition errors apply "true-strike" (or recognized zone) judgement.

 

They haven't done it because apparently it has become accepted. I'm not saying that making umpires stick to the true zone is going to improve borderline calls. But it will make the borderline calls surround the actual zone rather than their redefined zone, and umpires can be trained to know what the actual zone is.

 

From this blog:

 

But for our purposes the point is the obvious rounding of the zone at the four corners. That, as you can plainly see, is where a great many (probably most) of the under-recognition errors occur.

 

 

That statement says to me that they are counted as errors.

[Kimmi]

Slightly improved.

 

As I've said many times, I am okay with a 14% error rate if that's the best that humans can do. I think they can improve upon that holding umpires to a higher standard, but even if they can't, I'm good with it.

[Kimmi]

A couple of thoughts that are related:

 

1. Should fans not be allowed to sit so close to the HR lines where fan interference could affect the game? A fan last night may have affected the outcome of the game. Isn't that similar to the umpire affecting the game with an 'unfair' call?

 

Also, should we make all stadiums be dome stadiums so that, for instance, an outfielder does not 'unfairly' lose a ball in the sun? Is it fair that Wright's one bad outing this year was due to rain?

 

2. As far as the frustration with umpires goes, I think it would be extremely boring knowing that a ball/strike call is always going to be right. Part of the experience of being a fan is suffering the frustration of bad calls (I know there's plenty of frustration outside of umpire calls). The fan experience would become rather mundane if everything were always perfect.

[S5Dewey]

As I've said many times, I am okay with a 14% error rate if that's the best that humans can do. I think they can improve upon that holding umpires to a higher standard, but even if they can't, I'm good with it.

 

IMHO if you tell someone that a 14% error rate is acceptable they will meet that 14%. Of course that 14% is the average of all umpires which means some are worse. We just don't know who, or how much worse. But if 14% is the average then some are worse, and if you don't like 14% you're probably going to like 16%, 18% or 20% even less. (i'm hoping that even if you do like 14% you'd still like 16%, 18%, and 20% less).

 

I'm not going to go back and repost my posting back on the early pages of this thread, but I'm still in favor of using computerized strike zone identification to grade umpires with the worst of them being retrained and eventually removed from umpiring if they can't meet a minimum standard. Without some kind of "incentive" to improve they simply go on doing as they've been doing with impunity.

[User Name]

They haven't done it because apparently it has become accepted. I'm not saying that making umpires stick to the true zone is going to improve borderline calls. But it will make the borderline calls surround the actual zone rather than their redefined zone, and umpires can be trained to know what the actual zone is.

 

From this blog:

 

But for our purposes the point is the obvious rounding of the zone at the four corners. That, as you can plainly see, is where a great many (probably most) of the under-recognition errors occur.

 

 

That statement says to me that they are counted as errors.

 

We'd have to refer to the actual studies, since not only is there more than one study, and the "soft zone" conclusion is the blog writer's not either study. Then again, there are studies that use "true zone" instead of the one defined by the rulebook, and the error rate difference is not statistically significant. (It's there though)

[User Name]

A couple of thoughts that are related:

 

1. Should fans not be allowed to sit so close to the HR lines where fan interference could affect the game? A fan last night may have affected the outcome of the game. Isn't that similar to the umpire affecting the game with an 'unfair' call?

 

Also, should we make all stadiums be dome stadiums so that, for instance, an outfielder does not 'unfairly' lose a ball in the sun? Is it fair that Wright's one bad outing this year was due to rain?

 

2. As far as the frustration with umpires goes, I think it would be extremely boring knowing that a ball/strike call is always going to be right. Part of the experience of being a fan is suffering the frustration of bad calls (I know there's plenty of frustration outside of umpire calls). The fan experience would become rather mundane if everything were always perfect.

 

None of those are directly referenced in the rule book in a way that's meant to be specifically enforced. It's an apples to oranges comparison.

[Bellhorn04]

As far as the frustration with umpires goes, I think it would be extremely boring knowing that a ball/strike call is always going to be right. Part of the experience of being a fan is suffering the frustration of bad calls (I know there's plenty of frustration outside of umpire calls). The fan experience would become rather mundane if everything were always perfect.

 

Don't worry about that Kimmi, the players will never stop entertaining us with boneheaded mistakes like Leon getting picked off third last night.

[Kimmi]

IMHO if you tell someone that a 14% error rate is acceptable they will meet that 14%. Of course that 14% is the average of all umpires which means some are worse. We just don't know who, or how much worse. But if 14% is the average then some are worse, and if you don't like 14% you're probably going to like 16%, 18% or 20% even less. (i'm hoping that even if you do like 14% you'd still like 16%, 18%, and 20% less).

 

I'm not going to go back and repost my posting back on the early pages of this thread, but I'm still in favor of using computerized strike zone identification to grade umpires with the worst of them being retrained and eventually removed from umpiring if they can't meet a minimum standard. Without some kind of "incentive" to improve they simply go on doing as they've been doing with impunity.

 

I remember your post from the early pages, and I agree with your idea. There are some umpires who are worse than 14%, which means there are some who are better. The ones that are not meeting a certain standard should be replaced.

[Kimmi]

We'd have to refer to the actual studies, since not only is there more than one study, and the "soft zone" conclusion is the blog writer's not either study. Then again, there are studies that use "true zone" instead of the one defined by the rulebook, and the error rate difference is not statistically significant. (It's there though)

 

But the umpires are not technically making errors on pitches in the true zone. That is what the accepted zone has become. That's why the error rate doesn't change.

[SoxnCycles]

I remember your post from the early pages, and I agree with your idea. There are some umpires who are worse than 14%, which means there are some who are better. The ones that are not meeting a certain standard should be replaced.

 

Yup, using Pitch\Fx to "encourage" the umps to do their jobs better makes more sense than infrared googles or equipment loaded up with sensors or other wing-nut ideas.

[Maxbialystock]

IMHO if you tell someone that a 14% error rate is acceptable they will meet that 14%. Of course that 14% is the average of all umpires which means some are worse. We just don't know who, or how much worse. But if 14% is the average then some are worse, and if you don't like 14% you're probably going to like 16%, 18% or 20% even less. (i'm hoping that even if you do like 14% you'd still like 16%, 18%, and 20% less).

 

I'm not going to go back and repost my posting back on the early pages of this thread, but I'm still in favor of using computerized strike zone identification to grade umpires with the worst of them being retrained and eventually removed from umpiring if they can't meet a minimum standard. Without some kind of "incentive" to improve they simply go on doing as they've been doing with impunity.

 

That's actually a good idea. The problem with removing umpires who don't meet the standard is that I think they are unionized. But I agree the tools are there now to encourage/make umpire to improve on balls and strikes.

[mvp 78]

That's actually a good idea. The problem with removing umpires who don't meet the standard is that I think they are unionized. But I agree the tools are there now to encourage/make umpire to improve on balls and strikes.

 

And the next union contract isn't until 2020. We're stuck with the same system until then at least.

[Bellhorn04]

That's actually a good idea. The problem with removing umpires who don't meet the standard is that I think they are unionized. But I agree the tools are there now to encourage/make umpire to improve on balls and strikes.

 

And they'll inevitably reach a point where they simply can't get better and they're probably pretty close to it already.

 

With an automated zone none of those umps who are a little weak on balls/strikes would lose their jobs.

[User Name]

Yup, using Pitch\Fx to "encourage" the umps to do their jobs better makes more sense than infrared googles or equipment loaded up with sensors or other wing-nut ideas.

 

The playing field is already loaded up with cameras. It's just using existing technology. How in the actual f*** is that "wing-nut"?

[SoxnCycles]

The playing field is already loaded up with cameras. It's just using existing technology. How in the actual f*** is that "wing-nut"?

 

Yeah, that camera pointed along the left field wall did a bang up job of capturing the Bogaerts hit.

[User Name]

Yeah, that camera pointed along the left field wall did a bang up job of capturing the Bogaerts hit.

 

I'm talking about the pitch f/x technology. How do you think they actually capture the data? Voodoo magic?

[SoxnCycles]

I'm talking about the pitch f/x technology. How do you think they actually capture the data? Voodoo magic?

 

So, your plan is to transmit pitch\Fx data to an umpire and do it in under 1.5 -2 sec?

 

You think games are too long now? Wait untill that plan is implemented.

 

GameDay feeds aren't much slower than realtime games because MLB thinks it's cool to delay them.

It takes time to process the raw data into a graphic of the balls path.

 

You've been watching too many Star Trek mvies.

 

Again, MLB can't be bothered to track foul ball over a pole or a HR over a wall.

They're not going to bother with any expensive data feeds to umpires.

[mvp 78]

So, your plan is to transmit pitch\Fx data to an umpire and do it in under 1.5 -2 sec?

 

You think games are too long now? Wait untill that plan is implemented.

 

This is why they just need to let the robots take over the balls and strikes duties. Just take it completely out of the hands of the ump.