Can Data Sharing Solve the Pitcher Injury Epidemic?

Bill Streicher-USA TODAY Sports

Martijn Verhoeven wears many hats. As the research lead for the Twins sports science department, he is in conversation with all sorts of people, including baseball operations staffers, coaches, hitters, pitchers, and the medical staff. Verhoeven is armed with biomechanical data from KinaTrax, and the insights from the data help all these people do their jobs.

For understandable reasons, the Twins want to keep these insights private. Baseball is a zero-sum game — only one team can win the AL Central, and so the Twins would prefer their divisional opponents not know what they’re thinking.

But there is one area where this tendency for teams to hunt competitive advantages might be working against their interests: identifying solutions to the pitcher injury crisis.

“We have this massive injury epidemic,” Verhoeven told me. “There are times where I wish [teams] could share more and collaborate more because ultimately I think everyone would benefit from just having the best players on the field longer and more often. You can tell that people who’ve worked with this data for a long time are sort of moving toward [asking], ‘What can we do from a collective point of view in terms of making some of this understanding available?’”

In some ways, it feels that the sport is moving in the opposite direction. Teams hoover up talent from academia and wield their talents for internal use to gain slight edges over their competition. Meanwhile, the biomechanical data that KinaTrax or Hawk-Eye gathers with multiple synchronized high-speed cameras remain in the hands of the clubs, cutting off public researchers from data that could advance our understanding of pitcher injuries.

Which leads to a question of what the sport owes to itself. In July, The Dallas Morning News reported that the league had interviewed 160 individuals on the subject of pitcher injuries. A league source told me that a pitcher injury task force will be appointed after the interview project concludes, though the source could not confirm a specific date when the task force would begin its work.

Let’s assume that, in addition to the planned interview project and task force, Major League Baseball could accelerate its understanding of the injury crisis by permitting the public to access biomechanical data and by facilitating greater collaboration between teams. What steps could the league take to make that happen?

MLB could release the biomechanical data gathered by Hawk-Eye, allowing for public research to be conducted. It could fund an open-source journal for articles written by team employees. Teams could share their in-house data with researchers working at universities.

Whatever form it takes, the sport ought to embrace an open-source ethos toward its data and research. Doing so could kick off a movement to unlock solutions to one of baseball’s most intractable issues.

“I think a lot of people in my position come from an academic background,” Verhoeven said. “These are people that were academically trained, have advanced degrees in biomechanics or biomedical engineering. And what I think appeals [to these people] is that idea of cross collaboration. You have findings, you share those, you see findings from others, you try to replicate them or build upon them.”

***

Tyler Zombro is a recently retired minor league pitcher and a performance specialist at Tread Athletics, a pitching lab based in Charlotte, North Carolina. Zombro works with dozens of active MLB pitchers. He also suffered significant injuries as a result of a comebacker in 2021. These two facts give him a particular perspective on the question of pitcher injuries.

Zombro, who has communicated directly with MLB about the injury epidemic, frames the conversation in terms of “controllables.” One of these controllable factors, from his vantage, is a change to the “sticky stuff” policy. The league could allow pitchers to use sticky stuff to get a better grip on the baseball and, as a result, put less stress on their arms. During the 2021 season, MLB cracked down on the widespread use of illegal grip-enhancing substances, which resulted in artificially high spin rates and contributed to a depressed offensive environment. Pitchers have argued since the explicit ban that the baseballs are too slick and that the increased grip from the sticky stuff could help prevent arm injuries.

“Every major league baseball [is] different,” Zombro told me. “You don’t know how the baseball is going to come out texture-wise. And if you don’t have a substance and you need to increase feel for the baseball, you’re going to increase tension on your flexor mass to some degree. Call it 2%, call it 10% — [some pitchers] would say it’s 50%.”

The second controllable factor, in Zombro’s view, is the pitch clock. Zombro sees the addition of the clock as a potential contributor to pitcher injuries — but also allows that more research is needed.

“If you’re a muscle-driven thrower and your muscle fatigues, you have to find a new way to transfer energy across joints, which opens up the susceptibility to injury,” Zombro said. “Looking at how the pitch clock affects those energy systems and muscular fatigue, recovery times, etc. — no idea why that’s not a thing.”

More research is necessary to conclusively determine whether there is a connection between sticky stuff, the pitch clock, and pitcher injuries, though many pitchers have made arguments along the same lines as Zombro.

The league sees the issue differently. According to a league source, a central takeaway from the interview project is that frequently throwing at maximum effort is a primary driver of pitcher injuries. And Rob Manfred rejects both of Zombro’s suggestions.

“Getting out there and saying, for example, that the pitch clock is causing pitcher injuries, when the only study that’s been done suggests that’s not true, isn’t helpful,” Manfred told The Dallas Morning News in July. “The same thing with sticky substances. The idea that sticky substances, or the regulation of sticky substances, is causing pitching injuries [isn’t accurate]. The fact of the matter is pitching injuries have actually gone down since we started enforcing the sticky substance rule. We need to talk about what the facts are, not a narrative driven by other motivations.”

In any case, investigating these relationships is challenging. Part of this difficulty is the lack of public data on, say, energy systems and muscle decay among the major league pitcher population, and part of it is attributable to the culture of secrecy in major league front offices.

In reporting this story, I reached out to 10 front office members for major league teams. Some chose to speak to me on the record; some spoke on the condition of anonymity because they weren’t authorized to speak to the media. Others expressed an interest in speaking but declined due to the proprietary nature of data. Teams, generally, don’t like to let their front office employees speak to reporters.

Verhoeven, the Twins sports science research lead, made a point to speak on the record. He wanted me to know that the KinaTrax data contain the potential for furthering the sport’s understanding of how and why pitchers get injured.

“There are things where you start seeing, okay, these are just tendencies associated with guys who fatigue a lot during a game,” Verhoeven said.

One team employee, who requested anonymity to speak candidly for this story, expanded on this process.

“Even within a game, you’re able to see changes that are potentially signaling fatigue,” this team employee said. “We know if you’re fatiguing, that means your muscles aren’t working as hard, which means — in theory — that you’re putting more stress on the joints. With in-game data, you’re able to much more accurately quantify how that’s changing within a game and game to game. And so from that perspective, we’re definitely a lot more equipped to pick up on those things, identify those things early, advocate for more rest or workload restrictions on guys.”

Per Verhoeven, biomechanical data could reveal useful information about what exactly happened at the specific moment when an injury occurs.

“​​If something happens during a game, you now have data on that pitch where it happened,” Verhoeven said. “In a theoretical case, where something happens in-game, and the pitcher signals to the dugout to be taken out, you will have a record of what happened in that game.”

Verhoeven also pointed to the potential for these data to illuminate population-level trends, such as the biomechanical qualities of pitchers who get Tommy John surgery versus those who don’t.

“At a more aggregate level, you can ask the question, ‘Are there commonalities within those groups that you don’t see outside those groups?’” Verhoeven said. “How can they inform you about what it takes to be a pitcher with longevity?”

One ambitious goal for biomechanists is attempting to predict future injuries. A recent paper authored by researchers at the Mayo Clinic and three employees of the Twins (including Verhoeven) hinted at how that could shake out.

Using ball flight data from Statcast — velocity, pitch break, among other metrics — these researchers were able to predict whether a pitcher would get injured the next season to a startlingly accurate degree. (They also found — unsurprisingly — that velocity on non-fastball pitches was the strongest indicator of future injury.)

Another team employee, who was also granted anonymity to discuss the issue candidly, suggested that the strong relationship between the ball flight metrics and future shoulder/elbow injuries might actually be a proxy for the deeper causal relationship between these injuries and specific mechanical factors.

“It’s likely there’s an influence,” the team employee said. “The way that you’re manipulating the ball is going to change muscle activation at joints that stabilize to protect from the forces. So if you are taxing those muscles more, you’re probably going to fatigue faster and then expose yourself to higher injury risk.”

Even as clubs get better at determining when injuries might occur, one team employee speculates that injury predictions will still be framed in terms of rough estimates, and that ultimately team decisions will remain, as they are now, a matter of balancing risk with reward.

“Even if we did get really, really accurate with our ability to predict injury, at the very best we’ll be able to give some sort of likelihood to injury risk for a player,” one team employee told me. “The conversation would become something like, ‘Okay Player X, if you pitch today your likelihood of injury is 50%, and if you wait two days it goes down to 30%.’”

How to perfectly balance this tradeoff between injury risk and player performance remains an open question. And the people best equipped to answer that question, increasingly, are absorbed by teams. Moira Pryhoda, for example, was a postdoc in the Department of Mechanical and Materials Engineering at the University at Denver. While there, she wrote a paper titled “Lower body energy generation, absorption, and transfer in youth baseball pitchers.” A year later, the Guardians hired her as a biomechanical analyst.

If Pryhoda is working behind closed doors, she likely isn’t writing papers on pitching that I can easily find on Google Scholar. Instead, she is keeping those insights in-house, so the Guardians accrue the benefits.

This isn’t to single out Pryhoda and the Guardians. Pull up the staff directory for any given front office and you’ll likely find someone with a biomedical engineering PhD. But the cumulative effect of information secrecy and talent siloing is working against the game’s interests.

Adam Bloebaum, a reliever for the Nationals Single-A affiliate and a computer vision intern at Driveline, summarized the fundamental tension between pursuing competitive advantages and prioritizing pitcher health.

“As a player and even as a fan, you don’t like seeing the amount of injuries,” Bloebaum told me. “It’s obviously not good for the game. But I also understand the competitive advantage side, that if one team is able to keep their players healthy at a greater clip than the next, that’s the same as being able to take a hitter and give them 100 points of wOBA or whatever it is. I think we may struggle to get a lot of publicly available data until the competitive advantage is fairly minimal.”

An anonymous team employee concurred, saying: “We’re not as close as we think [to understanding why pitchers get hurt]. And the more people you can get working on it collaboratively, the closer you’re going to get. But yeah, there’s also the team aspect of it, the competitive advantage. And predicting injury better than another team is beneficial.”

***

With some of the best researchers and thinkers working behind the scenes and the data reserved for teams, the process of solving the injury problem will necessarily advance at a slower pace. Facing this collective action problem, the commissioner’s office could make a series of choices. For one, the league could publicly release the biomechanical data that it has already.

MLB, through Hawk-Eye Pro in partnership with Reboot Motion, provides biomechanical data to the 30 clubs on an opt-in basis. Imagine what public researchers could do with access to those data. As Stephen Sutton-Brown wrote on Twitter earlier this month, we are in a “golden age of open source baseball code.” Intrepid public researchers with day jobs are doing all sorts of fascinating stuff with data, but they’re using mostly the same data they’ve had available to them for years. Drop the biomechanical data, and who knows what the hundreds or thousands of people out there on the baseball internet might unearth.

“I think the more smart people you can tack onto an issue and the more transparency you have, the odds of you solving some large scale problem is better,” Bloebaum, the reliever, told me.

But there are also serious concerns with a full public release of this information. Biomechanical data, after all, are on some level private information about an individual’s body. Publicizing that information may lead to unintended consequences if the data are not properly contextualized; for example, teams could use that information to make roster decisions or to argue in arbitration that a player deserves less money.

“There’s a lot of excitement about [biomechanical data], but there also is a little bit of risk in terms of, ‘What if we get too focused on the wrong thing?’” Verhoeven said. “And I’m not saying that I know what the right thing is, but there is a risk focusing on the wrong thing.”

Verhoeven warned against the potential for these data to be used in ways that are less than judicious.

“My broader word of warning — mostly to myself, but to everyone else that’s interested in this is that you’re talking about human bodies, and there is just an inherent risk to that,” Verhoeven said. “With some of this stuff, I think there is a risk that people could get hurt.”

In the absence of a full release of the Hawk-Eye biomechanical data, the league could expand its Research Committee activities. The Research Committee, an internal MLB operation, reviews research proposals and facilitates publications in peer-reviewed journals. Historically, this league source said, the committee has been composed of team doctors and trainers, with less of an emphasis on biomechanics and the new generation of data.

Along these lines, teams could even fund a quasi-academic journal on pitcher injuries. The clubs could each contribute, say, the equivalent of a 15th round pick’s signing bonus; with this money, two or three editors could commission research from analysts around the league. It could be something like a trade journal, essentially, where biomechanists collaborate in the open to advance the question of how to stop pitcher injuries. All pieces in the journal would remain open source, providing a forum for all of the intelligence wrapped up in clubs to instead be interacting openly. The specific data shared in published articles would be aggregated or anonymized, and the writing would be accessible to the public.

“Is there a way where MLB can facilitate some of that while maintaining competitive advantages, but allowing for teams to collaborate or contribute in a way that can move the game forward?” Verhoeven asked.

Ultimately, the information, or at least an approximation of it, might end up coming out anyhow. Driveline provides some motion capture data as part of its OpenBiomechanics Project; it’s possible the organization releases even more. Dylan Drummey, who in partnership with Carlos Marcano recently unveiled the BaseballCV Github Repo, is at work on a project that appears to resemble KinaTrax or Hawk-Eye’s markerless motion capture technology, presenting the possibility of a public alternative.

Several people interviewed for this story argued that releasing more data is in the best interests of the sport, even if it requires sacrificing the competitive edge of some teams. As Tom Tango told Dave Cameron in 2016, “There’s no doubt that having 10,000 people looking at the data is better than 10 or 100. Open Source is the best thing about the internet… We’ve probably learned 1% of what we need to learn. And in 10 years, we’ll be at 10%. And that’s if everything is public and everyone works together.”

As I wrote in May, in response to release angles derived from Statcast, the public research community is “at both the literal and metaphorical threshold of the shifting movement from process-related statistics, freely provided in a Statcast .csv, toward biomechanical variables, largely the purview of the teams.” This is true not only from a performance standpoint but from an injury-understanding perspective as well: With these data, questions around arm health can be answered with greater clarity and confidence.

And the league is showing that it is comfortable with publicly sharing some amount of biomechanical data. On Thursday, Tango and MLB Advanced Media released aggregated arm angle data, which look at the placement of the pitcher’s shoulder, the pitcher’s center of mass, and the position of the pitcher’s hand at release. How far will the league ultimately go? Could releasing even more data to the public accelerate our understanding of pitcher injuries? Which data should be protected, and which should be shared?

These are difficult questions to answer, but they ought to be approached with an affirmative spirit toward the growth and preservation of the game.





Michael Rosen is a transportation researcher and the author of pitchplots.substack.com. He can be found on Twitter at @bymichaelrosen.

27 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Buhners Rocket Armmember since 2017
11 days ago

I’m inclined to believe it’s the pitch clock. Take a look at all the pitchers who had the longest pace in 2022, and see how they have done in 2023 and 2024.

Ohtani – TJ
Manoah – constantly hurt
Luis Garcia – TJ
Burnes – consecutive years of lost effectiveness
Nola – same
Patrick Sandoval – TJ
Gausman – great in 2023, down in 2024 and suffered from shoulder fatigue in spring
Darvish – many injuries
Blackburn – many injuries
Montas – immediate labrum surgery
Josiah Gray – TJ after stalling as a prospect
Clevinger – TJ
Montgomery – pitching himself out of the league in 2024
Springs – hurt both years after breakout in 2022
Giolito – ineffective, then TJ
Woodruff – TJ

I do not think the pitch clock is actually an issue. It will weed out pitchers who aren’t human rain delays at a younger age. If a guy can’t work quickly he will end up in the bullpen and I am fine with that.

Last edited 11 days ago by Buhners Rocket Arm
Kylemember since 2024
11 days ago

What’s the list of the quickest pace pitchers? How does that look now? Are they all healthy? How did you get this list?

Last edited 11 days ago by Kyle
Kylemember since 2024
11 days ago
Reply to  Kyle

Found it.

Shane Bieber – TJCole Irvin – hurt, bad, or just declining?German Marquez – TJMiles Mikolas – lots of inningsShane McClanahan – TJ
I found your idea interesting but I’m not convinced.

source: https://www.fangraphs.com/leaders/major-league?pos=all&stats=pit&lg=all&qual=y&type=15&season=2022&month=0&season1=2022&ind=0

Last edited 11 days ago by Kyle
Buhners Rocket Armmember since 2017
11 days ago
Reply to  Kyle

The list of quickest pitchers had mixed results because many of the guys who worked quickly were not as good generally as the ones who worked slowly. It seemed pretty well understood that taking additional time between pitches was advantageous for pitchers.

Kylemember since 2024
11 days ago

I’m not sure what your thesis is. It sounded like you are saying that the pitch clock (a faster pace) leads to injury, especially for players who took their time previously. But three of the 5 fastest pitchers from 2022 have had TJ since the pitch clock started. So it’s not exclusive to pitchers who had to adjust.

You still might be right, I just can’t grab the proof yet.

Aside: My favorite pitcher ever is Mark Buehrle. He pitched super fast and almost never got hurt. Probably an outlier though. I want data!

MikeSmember since 2020
11 days ago

Thee NCAA has had a pitch clock since 2011. There should be enough data there to show an effect if there is one. You could argue that it is different since MLB is a higher level of competition, but I think then you are just arguing that max effort all the time is a bigger factor than the pitch clock.

Albymember since 2024
11 days ago
Reply to  MikeS

College guys usually pitch once a week, don’t they? So it’s more than just the competition level that’s different.

NATS Fanmember since 2018
11 days ago
Reply to  MikeS

Perhaps it’s the combination. Pitching all out could hurt you faster with less time between pitches.

Albymember since 2024
11 days ago

I’d prefer giving them their four extra seconds, because I’d rather see good performance than a fast game.

Smiling Politelymember since 2018
11 days ago

I think the pitch clock is another reason why starters don’t go deep anymore; it leads to more fatigue and since injuries occur (and performance declines) when one is more fatigued, lots of clubs pull their guys earlier, so it’s noisy and messy.

This also ignores the massive elephant in the room, which is what happens to them before they hit the pros; what about the little league, high school, and college games where they throw a ton of pitches? where their mechanics and recovery might not be great? the throwing/stretching/conditioning in between? By the time a prospect even gets to the minors, *so* much has happened (and we haven’t even talked genetics, yet). I just don’t know how you measure and track the variables.

bernardgilkeyhasapossemember since 2019
11 days ago

I think the issue is that MLB will categorically refuse to look at pitch clock impacts, because it’s Manfred’s hobby horse.