LemmaEOF

Your favorite chubby cuddlebot

Hey! I'm Lemma, and I'm a chubby queer robot VTuber who both makes and plays games on stream! I also occasionally write short stories and tinker with other projects, so keep an eye out! See you around~

Chubbyposting and IRL NSFW alt: @cuddlebot

name-color: #39B366



cathoderaydude
@cathoderaydude

it's wild to me that

  • in baseball, half of the umpires job is to call strikes
  • as far as i can tell, it's agreed upon that they fuck this up all the time
  • it's incredibly hard for a human to do
  • it's one of the very few tasks that a computer is actually good at
  • there's no other sport where computer analysis could add any value. cameras can't keep track of hockey pucks or footballs, and a lot of other sports don't have the level of precision of baseball pitching
  • baseball is the only sport where a camera has an absolutely unfettered view of the ball on 100% of plays
  • they've already built computer systems to analyze pitches
  • they've been completely universal and reliable for over a decade
  • they still make it an umpire's call

it's just so bizarre. there's no value in this being a human decision, there's no "analog warmth" element to an umpire staring into the audience and daydreaming during a pitch and getting the call wrong. it's not like there's a judgment call to be made. it's a literal, scientific question of where the ball was, and nobody could come up with a good reason that answering that precisely on every single pitch with electronics would be a bad thing, that somehow makes the game less human.

i saw some shit saying that players would still argue with the "computerized ump" and i'm like, okay, right now they have an excuse, because the umps are sometimes wrong. whereas, if you want to argue with a machine that's provably 100% accurate... i mean, hell, you can't get thrown out for it, so it would actually make some games run smoother! sure, go bounce your helmet off the LCD screen. nobody cares. there's acrylic over it. i'm making the hand signal for time, get it all out

this is one of the very few cases where I think it's absurd to not use a computer. we only use them for shit they're like 15% good at, and then we pretend they're 100% accurate. this is a case where 100% accuracy is actually possible! we have sixty years of military R&D behind this, and it's just not a very hard thing to do! people write code that does this on a weekend, for fun, and it works!

apparently computerized calls might be coming in the next couple years, maybe, but there's no reason it should be this late. it doesn't even require computers, they could have solved it in the 50s with Electronic Eyes on either side of the plate and left only the altitude vs. height question up to the umpire. absolutely inexplicable

footnote: i do not actually care viscerally about any of this, i am just confused


rem
@rem

as somebody who loves baseball and has thought about this a lot, here are a few points that i think cohere into my pretty pro-umpire position.

  • prelude: if you believe a priori that there's no sense of 'analog warmth' in this specific dynamic, in baseball, or even sports in general, that's fine, but my tastes differ. sports to me are about humans doing cool stuff and the narratives that fall out of humans being placed in deliberately fanciful situation. they're about all the amazing stuff, cool stuff bad stuff, stupid stuff, human stuff that humans do, not about particular abstract conditions being fulfilled or not, in my mind.

  • there are plenty of other rules across sports that i find to be qualitatively comparable in intent to the concept of the strike zone (that being, 1) the pitcher is supposed to throw the ball to the batter in a manner where it could theoretically be hit, and 2) the hitter is supposed to try and swing at a ball that can be theoretically hit in turn. this is, philosophically, an imperfect, generalized, but common-sense metric to facilitate play in which people see good pitcher-batter matchups, the core gameplay loop and the heart of the sport.

  • there's a rule in basketball called the three-second rule: to be brief, defending players are not allowed to stay in the paint, a region around the hoop, for more than three seconds. the intent of this rule is, of course, to facilitate play in which people see good offense-defense matchups, because otherwise the defense could just sit under the hoop and there's no game to be played. this rule is not, and has literally never been to my understanding in the modern game, enforced as written, and nobody in the universe would want it to be, even though it could be tracked by a computer. borderline violations would completely disrupt the rhythm of play, and put a hard wall on what i think is the value of human officials, in that they are human. the one difference that makes the three-second rule more egregious if computer-mediated would be that basketball has a time clock, while baseball has an out clock, and fouls necessitate an interruption in normal play, while in baseball each PA ends in an advancement in play of some kind, with no time-outs to speak of.

the point being, no one cares about the three-second rule being enforced because it is successful in what incentives are created, regardless of whether it is legalistically violated as a matter of course or not. i would argue that's largely also true with the strike zone.

  • that argument, combined with the fact that umpire accuracy and consistency is actually pretty high compared to what a lot of people think (and absolutely compared to what it used to be) is why the human strike zone works, but i even think it's good on top of that. because it adds depth to the game. the batters and pitchers learning the particular nuance of the "enforced strike zone" that night is a genuine skill that is generally interesting to observe, but that pales in comparison to its specific distillate in catcher framing. yes, framing is lying to umpires, but lying is fun and cool, and the contrasts in this skill which is somewhat unrelated to the other skills make for a good catcher (though not totally, deception is useful for, say, put-out attempts mid-PA, though with the pitch clock that's going to go down) make for deeply fascinating strategic considerations when balancing your catching tandem and considering individual matchups. a huge dimension of the catching profession that has built up over time would largely disappear, and i think that's a bummer. again, i really can't make an argument if you think it sucks, but that's true with anything about sports.

also, running out of steam here but this is arguably the most important, but simplest point: i'm pro labor and i like to see people make money, and almost every anti-umpire argument online is mum at best to the question as to how exactly the back of their union should be broken, or why it's inherently okay to remove these jobs from existence unilaterally.


bruno
@bruno

This makes me think of another phenomenon in basketball: flopping. I am resolutely pro-flopping. I think flopping is funny and cool. I think Marcus Smart and Kyle Lowry are both entertaining and great basketball players.

Specifically, I think that the charge rule adds a lot of dimension and texture to basketball that wouldn't be there otherwise, and applying the charge rule in ways that favor habitual floppers like Kyle Lowry (Seen here trying to get a charge called on LeBron James in the all-star game) is cool and fun. So much of modern basketball is about matchup hunting; the charge rule means that bigs who are switched onto a guard can't just automatically back them up in the post for a layup, and that transition attempts can be meaningfully defended even when there's a big numbers disadvantage.

The flipside of the charge is that it also incentivizes defenders to get in the way of driving dunkers, which leads to more posters. Everyone goes home happy!

Flopping on offense leads to more bad shot attempts, which makes the game funnier and more chaotic, which are good things. Chris Paul having the option to try and extract two free throws from some mark by doing the rip thru leads to less possessions where the ball is just moving around the perimeter and the offense is just probing fruitlessly then settling for a mediocre shot.

Ultimately, pro sports need these unfair and random wrinkles to them because otherwise fandom would be very difficult. Imagine if instead of just tweeting "FUCK scott foster" you had to take a long hard look at the problems with your team's roster and coaching? God, perish the thought!


garudina
@garudina

BAD RULES, RULES THAT BREAK DOWN UNDER SCRUTINY, REFEREES!


cathoderaydude
@cathoderaydude
Play ▶
FUCK YOU ORIOLES!

Tracking
|

You must log in to comment.

in reply to @cathoderaydude's post:

My understanding, only barely following baseball, is at least the on screen graphics are not as obviously correct as they lead on. They're probably still more accurate than the computer umpire, but not always.

I want to say the technical exact edge of the box depends on the size (+ stance?????) of the batter as well. The on screen graphics are always just "put the box here relative to the surroundings and it always stays the same.

A computer COULD do this even if the box depends on how the batter is squatting or how tall they are, my understanding (not a baseball expert, citation needed, if this is something I'm wrong about please let me know because I would be interested in reading about it and learning more) is that it's not a 100% solved problem as things stand right now.

Also having a single human being make calls in sports without a computer to verify sucks ass. There's enough cameras and math to prove any call right or wrong depending on how much time and effort you want to put in, and they're probably gonna show that to the audience so why try and pull rank?

idk, the parameters are so simple. even if the tv graphics aren't 100% accurate - and honestly they look pretty damn accurate, i think jomboy does a good job of proving that - it wouldn't be hard to trim up the remaining slop. the camera is a known distance from the plate, known focal length lens, known size ball, known everything. this is literally something we could do with mechanical computers in the 40s. i think they just don't want to solve it because the MLB is like that.

on that point, and re: the last paragraph, MLB hates replays. they have a thousand cameras watching the game, and they hate it. they don't want anyone to see the replay footage during the game, they have incredibly strict rules about who can see it and when and who's allowed to make decisions based on it and it's just very clearly meant to keep things chaotic. the only possible explanation is that they like the fact that wrong calls happen and they want them to continue happening.

MLB specifically does feel a bit, stuck in the mud like that. Unfortunately. Glad I only care about it when the Mariners are fighting for the playoffs :)

quietly moves a couple of eggs into the NBA... erm, basket

And it for sure, 100%, would not be difficult at all to automate the calls with a computer and a couple of cameras. It's not hard, there doesn't need to be a noisy guy behind the plate to do it. Big agree. With my pedantic hat on, I'd say bad calls mean engagement? That'd be pretty lame.

i like to think of baseball as a game between three teams. the umps can never "win", but they sure can play spoiler. this would unfairly nerf the umps mlb pls fix

more seriously: the rules around the strike zone are completely bonkers. the ball needs to pass through a pentagonal prism extending above the shape of home plate, so it's not as simple of a question of "well did the ball go left or right of the plate". the ability to electronically adjudicate the other dimension, between "the midpoint between the top of the batter's shoulders and the top of the uniform pants" and "the hollow beneath the kneecap", is fairly recent.

i think we're only at this level of "this could simply be automated" because of statcast, and in the statcast era it seems like (most) umpires have taken the home plate position more seriously. hell last season was a home plate ump's first "perfect game". baseball is an incredibly slow sport to change, and given we've only played 8 seasons with statcast, i'm not surprised we're still here.

maybe we'll have automation in a few seasons, with the ability for umps to reverse a call on the spot if they feel like it. or maybe umps will continue to get better? you still have to have a guy stand back there to watch for other violations, like balks, which everybody definitely completely understands

computerized calling balls/strikes wouldn't lose anyone their jobs, you still need a guy behind home plate to watch for specific things outside of the pitch. if anything it'd reduce the amount of abuse they get from players and fans about balls/strikes

yeah but they don't want to do that, they enjoy calling balls and strikes. also, if they have less talent expected of them, there is less leverage that they have with their employer.

One thing that comes up when the robo-ump is brought up is the concept of "framing", where in catchers uses essential slight of hand to make balls look like strikes to the umpire. It's one of the ways that a catcher can add value to a team. I don't think it's enough of a reason to not use a robo-ump, but it would adversely affect the careers of some of the catchers who are really good at it.

They've had ball-tracking for LBW decisions in cricket for years now. It is only for (player requested) reviews though.

It's a similar problem to balls vs strike and it's been pretty accurate, though not foolproof. One thing it isn't though, is quick. There would have to be significant advances for the MLB's system not to disrupt the flow of the game, particularly with their new time limits.

Another interesting side-effect is that human umpires have improved in a lot of edgecases as a result of being corrected by machines repeatedly.

if they have a computer call the shots, baseball is no longer a game you and your friends can play at the park

to me, this has always seemed like the no-brainer reason to keep it umpire-based, and i just canonised it as the obviously correct one, based on no evidence whatsoever

came here to post this. the trickle-down incentives on aspirational players "needing" to get in front of a "real" digital umpire as soon as possible are back breaking to the sport's ability to remain affordable for casual play. I'm fine with managers having some amount of check on the system but human referees are integral to sports being approachable at all levels of play.

coming from a different cultural tradition, i don't really know baseball, but "there's no other sport where computer analysis could add any value" doesn't seem right? as someone else says, cricket has cameras for LBW calls, football (soccer) has VAR, tennis uses camera to call outs. or take fencing - admittedly that's not necessarily a computer, but it's still completely dependent on electronic scoring, even at an amateur level. none of this removes the need for referees, of course - plenty of rules that aren't clear cut, and in the case of VAR it's just a different referee making a decision based on the ability to rewind the footage.

so anyway: yeah, sounds like they should! plenty of other sports do & they still get to have all the fun of people making bad calls that everyone gets outraged about!

in reply to @bruno's post:

in reply to @cathoderaydude's post:

People think they'd be happier with clear cut, objective, automated calls. But they get just as mad at baseball video games when the computer calls a strike on the edge of the zone. I don't think automated umps would fix as many 'problems' as it's assumed they would, just because they could.

The Umpire tradition is not valuable simply because it's traditional, but it does add value to the game. The existence of an automated zone checking system wouldn't take all of that away, but it would take a big part of it. The question to ask is, would the automated system replace that value with something of equal worth?

There are options for RF-assisted puck tracking in hockey that are used in training, but I think the situation in hockey highlights the problem of refereeing based on intent. The icing rule, for example, in the NHL manual does have about twenty different bullet points enumerating when it is and is not icing. But it's sort of accepted that ultimately it just comes down to the opinion of the ref or lineman as to whether the play was both reasonable and reasonably likely to succeed. If so, it wasn't icing. If it seems to have been intended to delay play or was just a real dumb pass that was never going to work out, it probably is icing.