Technology, data and innovative training all have a role to play in ensuring its safe implementation.
Before all that, the Rhode Island native was a lifelong lover of the sport. From that vantage point – baseball fan who helps protect pitchers’ health as a profession – Buffi has a unique perspective of the pitch clocks that are now universal in the minor leagues and seem to be imminently headed to the majors, likely as soon as the 2023 season.
“Selfishly, and personally, I kind of like it. I do feel like the time between pitches has gotten a little crazy, just as a fan of the game,” Buffi says, adding this important proviso:
“But we’ve got to make sure we are smart about implementing it.”
After experimenting with a tightly enforced pitch clock in the Low-A West League (just re-named the California League) last year, Major League Baseball mandated its use across all four minor league levels in 2022. After a pitcher receives the baseball, he has 14 seconds to deliver a pitch when the bases are empty. That limit is extended to 18 seconds with runners on, except at Triple A where they have 19 seconds. Hitters must be ready to bat by the time there are nine seconds left. Infractions are penalized with an automatic ball (for pitchers) or strike (for hitters).
The resulting changes to time of game has been nothing short of dramatic. According to data provided by MLB, the average length of a minor league game through May 11 was 2 hours, 35 minutes—that’s a full 28 minutes faster than the 3-hour, 3-minute average of minor league games without the pitch clock in 2021. (The clock went into effect on 15 April; games played in the 10 days prior averaged 2 hours, 59 minutes.) Violations of the clock steadily declined during April and May to less than one per game.
The potential catch to what is very welcome news for a sport cognizant of hastening its pace and duration of games is that speeding up pitchers could induce higher levels of fatigue, according to preliminary research on the topic. Fatigue is the No. 1 predictor of pitcher injury.
Mike Sonne wrote his Ph.D. thesis on muscle fatigue prediction and has studied workplace ergonomics extensively. Among his other projects was consulting on the design of assembly lines at Ford Motor Company and devising the appropriate number of rest breaks to prevent injuries.
When interest in a baseball pitch clock heightened in 2015, Sonne and co-author Peter Keir researched its effect through a series of computer simulations. They concluded, “This study has shown the implementation of pitch clocks, or enforcement of existing pace of play rules, will increase the fatigue accumulated in the forearm and elbow musculature and could jeopardize joint stability.”
That finding was published in the Journal of Sports Sciences in 2016, but Sonne says now that the premise is rather intuitive, too.
“It doesn’t require any machine learning or anything like that,” says Sonne, now the Chief Scientific Officer of 3MotionAI, maker of ProPlayAI. “You just think about, if you go to the gym and you do 10 reps and you shorten the time between the 10 reps, it’s a lot more tiring by the end of it.”
Later that year, a research group in Taiwan conducted a similar study but experimented with college pitchers instead of computer simulations. The study was admittedly small – only seven pitchers completed all three phases – but gave empirical, physiological evidence to the effects of pitch clocks. Each pitcher was evaluated for pitching performance (velocity and location) and muscle inflammation and damage (via blood biomarkers) when throwing pitches every eight, 12 and 20 seconds for seven innings.
Their results, which were published in the December 2016 issue of Medicine & Science in Sports & Exercise, indicated that an eight-second pace was untenable, and there was also notable muscle inflammation lingering after the 12-second paced outing.
“Our data suggests that 20 is much better than 12,” says Donovan Fogt, one of the study’s co-authors and formerly a kinesiology professor before working as a scientist reviewer, vetting military research at Brooke Army Medical Center. He advises using the two completed papers as a guide to drive the study design of larger-scale research.
For years, MLB has had Rule 8.04 on the books, which mandates that a pitcher throw to the plate within 12 seconds of receiving the ball, although that, quite clearly, has not been enforced. That duration informed the design of the Taiwanese study, although the data is not apples to apples. In the experiment, a pitcher literally threw the ball every 12 seconds whereas the minor league pitch clock only starts after the pitcher receives the ball – either a return throw from the catcher or from a fielder if the ball had been hit into play.
Thus, in the minors, the average interval between pitches is inherently a bit longer. This is borne out in the MLB-supplied data that says the actual average time between pitches has been 18.8 seconds. The number of runs, hits and pitches per game have all been stable this year compared to last year – and, most importantly, so too have injuries.
“We monitor injury rates very closely at the minor league level, and we have an excellent injury tracking system that allows us to get some really useful information about the injury impacts of the different rules that we experiment with,” says MLB Executive Vice President of Baseball Operations Morgan Sword. “So far, we are one season and one month in, there’s no evidence to suggest that the timer is having a negative impact on pitching health.”
Interestingly, the Low-A West League – the only one with the strict pitch clock in 2021 – had “the lowest rate of pitching injuries of any of our minor leagues,” Sword says. “We’ll see what ends up being supported by the data, but there’s a theory that the quicker pace that’s forced by the timer prevents the max-effort pitching style, or at least mitigates the max-effort pitching style that is more common in today’s game and maybe contributing the more injury. So forcing pitchers to manage their effort level a little bit may actually have a positive effect on arm health.”
Pitcher velocities in the minors have been “pretty flat,” Sword adds, “and I guess that’s probably worth noting because they’ve been going up every year. We don’t yet have a ton of data to suggest a big change one way or the other.”
Under the terms of the new collective bargaining agreement, a Competition Committee consisting of six MLB-appointed members, four active players and one umpire will decide on rule changes. They can be implemented with as little notice as 45 days, but Sword pledges that the league is “cognizant that more notice is better than less notice on rule changes.”
The exact time intervals for the likely MLB pitch clock are not set in stone, either, with some calling for a relaxation of the limit.
“We’ve cut so much time from the minor league game that some games are being done in less than two hours now. And that’s good, I guess, but I would hope that maybe we can give the pitchers back a couple seconds because maybe we were too aggressive with the pitch clock, you know?” says Driveline Baseball founder Kyle Boddy, who spent the past two years overseeing minor league pitchers in the Cincinnati Reds’ organization. He adds, “At the end of the day, though, we do need to increase the pace between pitches, and I think the pitch clock is definitely a part of that weapon.”
Here’s how technology, data and innovative training can play a role in making sure it is implemented safely:
New training programs
“The body,” as Buffi recently wrote, “is great at adapting to what we train it to do.” Updated training will be the most important pillar of any new protocol.
“There’s a physiological limit where you can’t just do things rapid-fire over and over and over, but I do think that we could find a pace where, as long as the pitchers are training for that pace, they should be able to sustain that pace,” Buffi says.
The experts consulted for this story say pitchers should train with a clock for bullpens, simulated games and side sessions. Pitching and strength & conditioning coaches will have to adapt their throwing and training programs.
“You can plan for building athletes to face that requirement,” says Kinetic Pro Performance Founder Casey Mulholland. “What you’re building up for – and this is general workload principles – is you’re trying to get your body ready to handle the stimulus you’re going to face.”
When the minor leagues began implementing new pickoff rules in the minors last year, Boddy says the Reds coaches used their iPhone stopwatches in the bullpens to track pitcher delivery times. That data was then posted to keep every pitcher accountable in adhering to new standards. He believes proper game planning – such as presetting pitch calling sequences – can help, too. The introduction of the PitchCom system for sign delivery has been well received by some pitchers and may be adopted by more going forward.
Pitch tracking
Every minor league ballpark is outfitted with a pitch tracking device, whether it’s a TrackMan radar or Hawk-Eye camera system. These technologies provide a wealth of information. On the most basic level, they provide time-stamped pitch release times for monitoring pace. They also record pitch velocities, locations, movements and spin rates for tracking deviations in performance or signs of fatigue. Researchers can use these large datasets to track trends that may arise in either individual pitchers or the entire minor league system, both in the short term and longitudinally over multiple seasons.
Player monitoring
Several on-body and camera-based technologies can provide insights into how individual pitchers are faring, both in training and in games. “You just need to manage it, you need to measure it,” Boddy says.
Driveline Pulse (née Motus) is an elbow-worn compression sleeve with a sensor that can track a variety of metrics related to pitching volume and intensity. The KP Sleeve, made by KineticPro Performance and Nextiles, is a newer option.
Many MLB clubs have multi-camera motion capture systems installed in their ballparks, such as KinaTrax or Simi Motion. These solutions provide their own biomechanics analysis dashboards and can be augmented with additional tools such as Reboot Motion. Outside of MLB venues, smartphone options such as ProPlayAI, Mustard and Uplift Capture are alternatives.
Deviations in mechanics, as measured by individual baselines, can indicate signs of fatigue.
“The way we do it at Reboot is for a selection of the metrics, we try to calculate a normal range as a function of pitch count,” Buffi says. “So then we try to see, as the game goes on, at any point is there a time when that person fell outside of their normal range?”
Fatigue Units
About a year after his research paper on pitch clocks published, Sonne set out to develop a metric to monitor the accumulated strain on pitchers. This new stat, called Fatigue Units, takes into account the number of pitches thrown in an inning, the type of pitches thrown, the velocity of the pitches and the interval of time between them. That is then aggregated across games and seasons, accounting for days of rest. Relievers in particular populated the Fatigue Units leaderboard.
“Some of these big relievers, I hope they can stay healthy, but they’re going to have to really monitor their workloads, and maybe it’s time to let off the gas a little bit,” Sonne says. “You can’t throw as hard as you were before.”
Mulholland has assessed pitchers’ warmup routines and discovered that too many guys go right from long toss to the bullpen to the first inning, causing the largest fatigue spike in the first inning. That’s an avoidable one. The other pratfall is the long inning in which a pitcher needs 30 or more pitches to escape a jam.
“You just want to be as efficient as possible as a pitcher, which might change the way some guys approach going into the game and pitching,” Mulholland says. “But you just don’t want to amass a lot of pitches in a short amount of time.”
Majors vs Minors
“It’s possible that the rules in the minor leagues and major league shouldn’t be the same,” Boddy says. The objectives of the leagues are not completely aligned. Organizations want their players to develop winning habits and competitive instincts in the minors, but winning a championship is not of such primal importance as it is in the majors.
“At the end of the day, the goal is not to prevent all injury,” Boddy says of the minors. “The goal is to develop big league pitchers. And so how do you balance those things that are constantly at odds – giving them enough workload and enough stimulus to make them better and give them the opportunity to improve with not doing too much.”
Changes implemented in the minors typically take three to five years before they manifest in the big leagues, he adds. That lag is because pitchers build up stamina and durability over time. Boddy believes limiting the number of pitchers on a roster will help provide the incentive for pitchers to develop those physical tools and hopefully help quicken the pace of games by limiting pitching changes.
While minor league managers might be more inclined to let a pitcher work out of trouble for the sake of his development, Sonne doesn’t think a big league club would offer the same leash to its pitchers.
“The major league game is so different that, if the command goes, if the velocity drops, you may have more base runners, you may have more walks, you may have more pitching changes, and does it actually help?” Sonne says. “Are you going to get 20 minutes off of a major league game? I don’t know if I’m convinced of that yet.”
More research
“Generally speaking, it’s a reasonable idea, but there just hasn’t been enough research into it,” Boddy says, noting that there are several motion capture laboratories who credibly can conduct the appropriate studies on the topic and calling for a clear timeline of implementation.
“How do you individualize this?” Mulholland says. “I just don’t think we have enough data on it. And I don’t think it’s been to the point yet where players have really overly thought about it.”
For now, MLB is opting to study game data – with the giant sample size of 120 minor teams – through its injury surveillance database and the various pitch tracking systems rather than commissioning lab research, which has its own limitations when pitchers aren’t throwing with full game adrenaline.
And, of course, pitchers may not be the primary culprit to slow pace of games. Fogt wonders why the onus on speeding the game is placed on the pitcher. “When you think about all the screwing around that the batter can do – get into the box, get out of the box, get into the box, get out of the box – how can you blame the pitcher for that kind of shenanigans?” he says.
This article was brought to you by SportTechie, a Leaders Group company. As a Leaders Performance Institute member, you are able to enjoy exclusive access to SportTechie content in the field of athletic performance.
A data & innovation article brought to you by our Partners
“I experienced it first-hand as a practitioner,” Rich Buchanan tells the Leaders Performance Institute, having previously worked behind the scenes at Swansea City FC, the Wales FA, and other organizations. “The people tasked with making sense out of assorted data are under immense pressure to interpret information every single day for every single player throughout an entire season. The weight of delivering ROI lies in their hands.”
Buchanan, who has also worked for organizations in the US and continental Europe, is now the Performance Director at Zone7. The company’s artificial intelligence [AI] system is being used by a growing number of top-tier sports outfits as they look to convert data collection into actionable insights for higher levels of athlete performance and availability.
“If you have one person undertaking manual analysis and interpretation, it is difficult, if not impossible, to do that consistently well over the course of a campaign,” he continues. “Just as we’ve seen in other industries, I believe AI will play an increasingly important role in helping professionals identify complex risk patterns from an athlete’s data that would otherwise be invisible to the human eye.”
How does data fatigue manifest? “Typically, sports scientists or fitness coaches are the ones tasked with monitoring players using data-generating tools, like GPS or biometric wearables,” says Buchanan. “Many of these professionals, however, are not data specialists by trade. They are generally sports practitioners whose expertise is better served in athlete-facing environments – on a pitch, in a gym, etc. – where they work directly with athletes in a practical manner.”
“Right now, the conventional norm for making sense of athlete performance data in sport requires significant time commitment to examine large datasets. It is a laborious manual analysis and interpretation process, one which often runs counter to why these professionals embarked on a career in sport in the first place. As a result, disenchantment becomes quickly apparent and you see signs of data fatigue setting in as they’re pressured to deliver meaningful insights that can then be practically applied.
“Even now, as many organizations employ data science personnel, it’s not humanly possible to do this kind of manual analysis and interpretation effectively and consistently for multiple athletes over the course of a season. Combined with the frequency and complexity of new datasets becoming available, the manual approach often results in flawed insights that reduce the value that organizations receive from investments in data collection tools.”
Buchanan works closely with Tal Brown, CEO and Co-Founder of Zone7, whose extensive background in creating and deploying AI technology with the likes of Salesforce and Oracle has allowed him to witness data fatigue first-hand.
“The need to find accurate solutions for harmonizing, analyzing, and interpreting such large volumes of data has never been clearer,” says Brown. “I’ve spent much of my career creating intelligent tools that minimize data fatigue while creating more efficient data analysis processes. This is now becoming a significant challenge for decision makers in sport, especially as they’re tasked with validating and correctly interpreting data from a growing array of different sources. It’s not just game and competition data, we’re also seeing increased amounts of medical, strength & conditioning, sleep and general wellbeing data generated by wearable technologies.”
The scenario of data fatigue described by Buchanan is not uncommon and Zone7 is aware of the vital need to adopt a ‘practitioner’s lens,’ continues Brown. “Tech innovation and evolution is driven by the need to answer harder questions in more efficient and reliable ways. You need to collaborate with practitioners and ultimately provide the insights that add value in their specific environments.”
One such practitioner is Javier Vidal, a Performance Coach with Spanish La Liga club Valencia CF. Vidal has used Zone7 in a number of different team environments. “Zone7 is a tool I’ve used for several years,” said Vidal. “Its AI has allowed me to adapt my day-to-day routine and get more value of out new data generating technologies that are arising all the time.
Zone7 has been deployed and operated real time by Vidal at Valencia CF since the start of the 21/22 season and the number of confirmed injuries has dropped by 52% compared to the previous season. This closely resembles results during his tenure in Getafe another La Liga team, where Vidal saw a drop of 70%, with a 65% reduction in days that first-team players were lost due to injury. “It would be the work of many people analyzing data all day to gain such useful information, but with Zone7 I get accurate, usable information within minutes that I can immediately put into practice.”
Buchanan adds that, in the case of forecasting injury risk, AI can also take a complex, multifaceted problem and present it in an easy-to-understand way. “At the top line, Zone7 presents ‘athletes at high, medium, or low risk. On top of that, it presents potential actionable solutions, such as, ‘do more or less in this specific area than you originally intended.”
Multifaceted problems also create greater risk of blind spots emerging in athlete monitoring. In this case, the relationship between classic sports performance data streams such as external workload, strength and internal workload, and ‘next gen’ of data points such as heartrate variability [HRV] and biomarkers monitoring stress, sleep or diet presents a margin for error. “We can create visibility into how those nodes interact in the day-to-day environment,” says Brown. “More data allows us to calibrate tools that can understand those relationships. Traditional spreadsheet-like tools just don’t offer that functionality.”
One of the missions driving Zone7 is to give practitioners a sense of security at moments when they are under significant pressure to deliver. “One of the hardest parts of a coach, analyst, or performance specialist’s job is giving definitive advice on decisions made around athletes,” says Buchanan. “The number of times these professionals are tasked with doing this under pressure from all directions, with only their own subjective opinion to draw upon, is concerning. Equipping medical, fitness or coaching professionals with the objectivity that AI provides, driven by complex computations, creates more certainty and a stronger case for the advice they provide in that scenario.”
“Humans, by nature, already have in-built biases,” Buchanan continues. “We already have opinions about certain athletes; who’s robust and who’s not robust, who’s likely to report muscle tightness, etc. Now, if Zone7 corroborates those opinions, it assures the user. If what Zone7 suggests turns out to be true, then the trust builds. When we suggest something that’s counterintuitive, people may find that uncomfortable because it’s challenging preconceived beliefs. But I would say, in a way, that’s the true value-add that Zone7 brings, highlighting those blind spots. We’re there to make sure those players, and the opportunities to pre-emptively intervene, don’t slip through the net.”
Brown and Buchanan are excited about the opportunities that lie ahead for data science in professional sports, whether that be 360-degree support for any interaction a player has related to performance, the possibility for longitudinal injury risk profiling, or the ability to support practitioner across a variety of different sports, each with their own cadences. The latter is already underway, with Zone7 being actively applied across American football, basketball, baseball, and rugby, in addition to extensive work in soccer.
Zone7 will also continue to challenge accepted wisdoms. “Regardless of the value individual practitioners and sports organizations place on data to manage athletes/players, the industry has been heavily reliant on simplistic data analysis and interpretation for a number of years and, quite simply, there’s now a more refined way of doing it,” says Buchanan. “Using AI to simulate different training load scenarios with the aim of physically peaking on certain days on the training cycle is just around the corner. “This is where people can get very precise on how they prepare their players or athletes for future events, rather than best guess periodization models.”
“Ultimately, future sporting success belongs to those who leverage their data in the most efficient, effective and accurate manner. We’re here, primed and ready to help practitioners in sport do just that.”
To connect with Zone7 directly, please email [email protected]
“There could be something in a group chat that was buried under half a dozen different birthday messages,” he tells the Leaders Performance Institute.
That missing information could be the details of a pitch the team is training on, the details of a medical appointment or a sponsor engagement.
Trimble, a former rugby player with 229 caps for Ulster and 70 caps for Ireland, explains that teams across all sports struggle with communication channels. “There are teams with 10-15 WhatsApp groups or – just as bad – they have one group for every single topic and there are numerous threads being discussed; and no one can understand everything,” he continues.
It sounds confusing. “Then there’s numerous screen grabs of whiteboards, which are obviously non-live. In some instances, an athlete has to walk into the medical room, pick up a marker and book an appointment that way, which is very strange in this day and age.”
Trimble, who retired from playing in 2018, heads up the Belfast-based Kairos, who have created a unified digital planning platform that enables better communication within sports organisations, from operations and management to coaching and performance. The platform was designed to be not only sport agnostic, but has tools to support every level from first team down through academies.
Kairos – a Greek term meaning ‘the decisive moment’ – enables teams, through their app and desktop-based platform, to solve problems by eradicating the distractions caused by multi-channel approaches. Trimble and Kairos Chief Operating Officer Gareth Quinn, who developed the first iteration of their platform in the mid-2010s, soon realised there was no suitable tool on the market. “We received strong validation that this is a problem that’s really worth solving,” says Trimble, who explains that Kairos is compatible with third party calendars such as Outlook and Google. “With our platform, it’s all very clear, there’s no clutter, all the athletes know where they need to be and there’s greater levels of accountability.”
Their platform is currently used by teams in the English Premier League, United Rugby Championship, Premiership Rugby and a series of teams across North America and South East Asia. Trimble points out that these teams may not just be suffering from a problem of unclear communication. “From a staff member’s perspective, it could be getting assurance that if you send something important to an athlete that it’s going to be delivered, received and understood and engaged with correctly,” he says.
“If it’s one place, then it doesn’t take up any cognitive load for the athlete. They can spend 100 per cent of their time thinking about performance. If that’s compromised in any way, if they have to scroll through their screen grabs or pictures of a PDF, and they have to scroll through their email for something else and look through their WhatsApp group to find a thread, then all of that is a distraction and all of that impacts on performance. Equally, if multiple departments are speaking different languages then you’re asking an athlete to be a goalkeeper with ten different goals to defend.
Trimble delves further into the issues that can exist within a single team. “Even within one department you can have three or four different behaviours,” he says. “Take a medical department. One medic may create a block of availability that allows athletes to book appointments, another medic may book that same slot of availability but then allocate slots to athletes, there may then be another medic who bypasses all availability and pushes bespoke events or appointments to players.
Kairos helps to solve such problems. “Athletes and staff get the assurance that everything is on the platform and they can see it. It’s all live. Any department that wants to communicate with the athlete will use the one platform. Ultimately, they’ve got one goal to defend and, if anything changes, they can see notifications, reminders or updates on their notification channel; it’s very clear and very easy for them to know where they have to be, what they have to do, what the requirements are, and then how to get the best out of themselves.”
The platform can also be adapted to the prevailing culture at a team. “There’s ownership on one end and management on the other, and every team lies somewhere on that spectrum, but it’s important that we can support everybody, whatever that team culture is or environment or what the expectation of the players is; either to tell them where to be or what to do, or to allow them to manage all that themselves. We’ve got tools to capture both behaviours.
“There’s a number of different ways that you can use our software and it’s important that we can work with a team and find a way that works best for them and gets them the best results and, ultimately, gets their athletes performing the best.”
On that note, Trimble says that Kairos is continually reiterating its platform. “Every conversation for us is about discovery, finding where the club is at, and deciding what their unique issues are and, nine times out of ten, we will have encountered something similar before and there will be a mechanism in place to be able to provide a solution in the software,” he says. “We can take them through that, but often there can be something unique and there’s learnings in that for us too. That could even just be a coach with a new way of thinking about the game, a new way of communicating, or a new operational procedure. We have to capture that development.”
This attitude points the way forward for the next 12 months. “The next phase we’re going to be working on is the integration with third parties, be that GPS providers, sleep data, or nutrition and diet.”
The feedback from Kairos’ ever-growing client base has been positive too. “They say it is unthinkable that they’d be able to go back to the way it was before,” says Trimble. “Professional athletes require a platform that treats them like they’re a valued professional and communicate in a way that gives them back time and takes away distractions.”
