Strike zone data and suggested changes to how we do things now

Regardless of which time period you examine in baseball history, there are always big issues that surround how the game is played. Right now the most talked about issue are the abundance of home runs being hit, why that is, and if the game should take action to restore some balance. But perhaps an even more important issue is one that’s most fundamental to the game – the calling of balls and strikes.

It has to be said that calling balls and strikes is probably an impossible task for human beings and that umpires do a better job today than what they did 20 years ago, when each umpire seemingly had a different strike zone. While recognizing the improvements that have been made, we still get daily reminders of calls that were simply wrong. So, what – if anything – should be done about that?

Before we get to that, let’s try to inject some context into things. Of all places, BloombergBusinessWeek keeps track of umpire calls. Unfortunately, they don’t provide a lot of information on how these rankings are compiled. It appears that all calls, not just ball-strike calls, are included in their lists. According to Bloomberg, all umpires get between 85 and 90% of calls correct. This seems pretty good until you consider the number of calls that are pretty cut and dried, like the pitch that bounces at the plate or the call at first base where the runner is out by 3 steps.

While the Bloomberg lists may not be exactly what we’re looking for, they do provide a section called “Worst Calls,” which lists the worst calls of the day and how much they missed by. On the Dodgers-White Sox game of August 16, Bloomberg notes: “With a 0-0 count in the bottom of the 6th inning, umpire Mike Muchlinski called a strike on a pitch that missed the strike zone by 7.62 inches.”

The latest call involving the Mets on this “Worst Call” list happened on July 20 in the game against the Cardinals. Bloomberg reports, “With a 2-0 count in the bottom of the 2nd inning, umpire Joe West called a strike on a pitch that missed the strike zone by 7.72 inches.”

There are always going to be calls that could go either way, ones right on the edge of being inside or outside the strike zone. But when pitches that are seven inches away from the zone are being called wrong, that’s an issue. It seems we can all agree that these types of calls just shouldn’t happen.

Back in 2016, HBO’s Real Sports ran an episode entitled, Man vs. Machine, in which it made the case for using technology to call the strike zone. Star-Telegram reporter Gil Lebreton wrote up a piece of the episode and he said the following in his piece:

They asked Yale professor Dr. Toby Moskowitz to study the Pitch f/x data from MLB games in recent seasons. Moskowitz analyzed every pitch called by major league umpires – nearly a million in all – over the last 3 ½ years.

MLB claims its umpires call 97 percent of balls and strikes correctly. But according to Moscowitz (sic) and HBO, the study showed that only about 88 percent of the calls were accurate.

Roughly one of every eight pitches, in other words, were called incorrectly. As Frankel pointed out, that adds up to more than 30,000 bogus balls or strikes each season.

That figure includes the obvious calls where the pitches are right down the middle or way outside. When Moscowitz (sic) narrowed his analysis to pitches that were within two inches, either way, of the corners of the plate, the umpires got the call wrong 31.7 percent of the time – nearly one of every three pitches!

That last figure is the most damning. On close pitches, the umpires are calling the pitch incorrectly nearly 1/3 of the time. Again, calling these pitches is an incredibly hard thing to do. We’re asking humans to do a job that they’re not ideally suited to do. This is not a problem of either effort or willingness. No one should look at the umpires as lazy or not caring. Just like we don’t expect a batter to get a hit every time up, we can’t expect an umpire to get every close call right. It’s just not realistic.

The ideal solution would be for an electronic system to call the balls and strikes and instantly send a signal to the umpire on the field, so that he could officially make the call. It seems unlikely that we have that technology for that right now. It seems even more unlikely that the umpires’ union would allow such a system to be used without a fight.

For non ball/strike calls, we have a replay system in place right now which almost no one likes. It allows teams the chance to use technology to decide if a call should be challenged and then the replay itself takes way too long. Instead of being used on obvious blown calls, it’s being used to question bang-bang plays.

Perhaps we can kill two birds with one stone and fix the current system while also mixing in the ability to challenge ball/strike calls.

Taking a page from the NFL, all scoring plays should be reviewed without a challenge. Close play at the plate? Instead of taking 30 seconds to decide whether or not to challenge, this decision is removed from play, with the replay being automatic.

Additionally, on all challenges, the manager has to decide within five seconds whether he wants to challenge the call. This will go a long way towards eliminating the reviews on bang-bang plays and get back to the original purpose of fixing the truly blown call. Sure, managers will continue to challenge bang-bang plays but hopefully these will be limited to ones at extremely important times, rather than at any time at all during the game like we see now.

How to handle replay on ball/strike calls is not so easy to handle. What’s a bang-bang play at first base is not the same as what’s a borderline pitch. Some hitters, like Joey Votto, know their strike zone amazingly well. What might be a borderline call to me and you might seem like a clear ball to Votto. Picking numbers out of the air, if we view a pitch three inches out of the zone as a blown call but Votto views a pitch one inch out of the zone as a blown call – why should he be penalized?

Perhaps there could be a separate challenge category for ball/strike reviews, allowing each team a set number, say three, per game.

It’s important to recognize that we’re trying to implement a system that’s better than what we have now, not one that’s foolproof. If umpires are missing 31% of close pitches and we can introduce a replay system based on efficiency that reduces that to 25% – that’s a good thing.

Of course, we’ve seen how a system is designed and how it’s instituted are two different things.

Personally, my goal is to see the calls be made right as often as we can, with the understanding there will have to be tradeoffs. We’ll never be able to have all calls both right and quick. And even with an electronic strike zone, it’s unlikely that accuracy would reach 100%. My preference would be not to have a gimmicky challenge system. But that may be the most efficient way to handle things, at least here in the short term.

The goal is to always move towards a system that’s better than what we had previously. We can debate what makes something “better.” But hopefully we can all agree that what we have now can use improvement.

10 comments for “Strike zone data and suggested changes to how we do things now

  1. Jimmy P
    August 18, 2017 at 1:05 pm

    A good, well-considered piece about technology that one day might exist. I wonder what pitch it would hurt most. Curveballs, I’d guess.

    But I find myself not really caring. And I hate the idea of more challenges.

    Baseball has a huge problem right now. The length and pace of game is terrible, and still moving in the wrong direction. Up 5 minutes this season. Combined with all the non-events per pitch, we are looking at a product that is not as exciting, interesting, or fast-paced (relatively) as it used to be. The bullpen usage is a big part of the problem.

    It’s becoming a sport for old white guys and mathematicians. Reversing that trend has to become the #1 priority. The kids aren’t watching.

  2. Pete In Iowa
    August 18, 2017 at 1:34 pm

    Nice piece Brian.
    The “five second rule,” as I outlined a few weeks back, is most definitely the way to go. Why should teams be allowed to have someone who is not even on the field look over a replay (and from several angles to boot) prior to a challenge being made? Under a five second scenario, it will be up to eyeballs on the field (just like the umpires, BTW) to call for a challenge of a call which, presumably, would be so bad as to not need a time-wasting replay study by some staffer off the field to make the call. In addition, managers should be allowed 3 losing challenges per game. And, the “umpire” challenge should be eliminated entirely.
    As far as balls and strikes are concerned, I certainly agree that something must be done. I just don’t think a challenge of these calls is the way to go. For example, what evidence would there be to justify the call? The ridiculous “K-Zone” box which is used on telecasts? Or, even worse, a camera which doesn’t have the correct angle to definitively make such a call to begin with? I think an electronic technology must be perfected and instituted, however long that may take.
    I am by no means knocking umpires. They have a difficult job. As far as balls and strikes go, I disagree with your premise that calling them “is an incredibly hard thing to do.” I don’t see it as any more difficult as a close play on the bases. Could possibly be easier as there is only one object in motion to be judged, unlike close plays at bases when both the ball and a runner’s foot must be judged to make a call.

    • August 18, 2017 at 8:31 pm

      Thanks for the kind words!

      It may be no more difficult to call a strike on a pitch 1/2 inch either way from the edge of the strike zone than a bang-bang play at first base. But you’re going to have 30 of those calls every night as a home plate ump and maybe two at first base. Plus you have to make calls on 270 other pitches. Plus any plays at the plate. Plus any balls hit before the bag.

  3. Name
    August 18, 2017 at 1:52 pm

    I think camera angles are quite unreliable.

    What i want to see in future is sensors on everything – the ball, gloves, plates. That’ll allow an accurate and quick output on bang bang plays.

    As for strikes and balls at the plate, a partial solution right now would be sensors underneath home plate (maybe along the edges?) that “looks up” to see where exactly the ball crosses the plate. We should easily have tech that can do this but it’s only partial because it would only be able to determine inside/outside calls.

  4. Eraff
    August 18, 2017 at 2:17 pm

    I’m at least “halfway there” on Automated Ball/Strike….I’m even more interested in an expanded strike zone…swing the bats more….put the ball in play.

    The recent launch pad results are largely based on the use of tech and math—launch angles, swing paths…… I’d like a more athletic game with the ball in play. A larger zone may provide a counter intuitive result of more contact…maybe a bit less success, especially in regards to HR’s. I do believe it will take away the “My Pitch..My Swing” approach. It will probably lessen the proclivity of hitting into shifts.

  5. Chris F
    August 18, 2017 at 2:55 pm

    Fascinating connection with a discussion Steve Phillips and Jim Bowden had on MLB radio this morning.

    So far I haven’t seen anything that makes me think that a digital strike calling machine is on the near horizon.

    All the of the K zone boxes Ive seen so far are pure junk, and Ill take an up any day of the week. Ive never seen the size of the box change. It pretty much looks like boxes are the wodth of the plate and set at the knees and shoulders of the catcher. Batter size seems to play little or no importance in what we see in the boxes. Really every single batter needs a strike zone (see the diagram at the top of the article. Who exactly will be drawing that, or will every batter be measured digitally before the beginning of the season so that their own K zone immediately comes up? I can see how that will go.

    More importantly, the image, and what we see on TV is a 2D representation of a 3D world. Everything is flattened as if there is no value to the in/out aspect of the screen, which we all know is BS. Home plate is a 5 sided polygon which will require some serious laser arrangement to actually find the exact dimensions, especially the back part of the plate. I dont see anything in a 2d rectangle that matches that. Of course we have the vertical dimension too so in fact the strike zone is a volume, and I encourage everyone to read this interesting article (note the spaces inserted in the link):

    http: //www. hardballtimes.com/analyzing-the-strike-zone-as-a-three-dimensional-volume/

    I am curious just how a laser set up will be set up and adjusted for every single player, each of whom has their own K zone. In the article above, look at the top illustration of a back door breaking ball. That strike touches the volume of the K zone at a single point, while the flight of the ball is out of the zone at every other. instance. When we see a flattened rectangular strike zone that is an area, the possibility of making many mistakes is real. I have yet to see any digital arrangement discussed to actually make a perfect volume strike zone. Until that can be done, I’ll go with the umps, yes even Joe West and Angel Hernandez, ahead of the bots.

    • August 18, 2017 at 8:43 pm

      I just want to point out that ESPN used to have an attempt at a 3-D strike zone on its broadcasts and it would rotate the angle so you could see if it caught any part of the polygon beside the front edge. And I was the only one in the chatters who liked it.

      Anyone not connected with the technology compaines has no idea what technology is or is not available today. Likewise, no one not in that small circle has any idea how long it would take to implement such technology. We can offer opinions – as I did when I wrote that it seems unlikely – but that’s all they are, opinions.

      The issue is not to institute an electronic strike zone today. The issue is to make it a priority to happen ASAP. And that can be at the start of next season or five years from now or whenever.

  6. Eraff
    August 18, 2017 at 4:03 pm

    Wow…apparently Chris is Smarter than he’s been letting on! 😉

  7. Metsense
    August 18, 2017 at 8:00 pm

    When the technology can support calling balls and strikes then I will be all for it. It would be a more consistent strike zone.
    Managers should have to request a challenge with in five seconds without the aid of a replay and based on what they saw with their own eyes, Challenges are ruining the flow and spontaneity of the game.

Leave a Reply

Your email address will not be published. Required fields are marked *

%d bloggers like this: