October 12, 2017 Will Carroll I spent most of a morning digging through data, looking at both the kind of data we can get with the motusTHROW and velocity logs from both this season and some notes I took from a research project I did back at Baseball Prospectus in 2004. While this is hardly definitive, I believe that it’s clear that fatigue is both steady and predictable when it comes to pitchers. This is not groundbreaking, but it is important. What does “steady and predictable” mean? At it’s most basic, it appears that pitching fatigue, as measured by velocity, arm speed, or arm stress is almost entirely based on muscular fatigue. This is not to imply that pitching mechanics are unimportant, but that they are individualized to such an extent that measuring them is a variable that can be relatively ignored when measuring fatigue. It’s clearly important for stress and for long term viability, but my assumption is that at upper levels, mechanics are an established and largely repeated pattern, excluding it largely from this kind of calculation. Let me reiterate this for clarity: pitching mechanics are very important up to the point a pitcher comes to the mound. At that point, they must be considered fixed for that period. Work on your mechanics and efficiency in practice, but once you get to the mound, a pitcher should never think about them and largely, neither should the coach, absent a major breakdown. Research on other muscular fatigue, whether its from running or simple weight, follows this pattern. A muscle can give a certain amount of effort, then deteriorates or if pushed too far, it breaks or causes a breakdown. Pitching doesn’t focus the stress on any one muscle or muscle group. Fatigue may come in a number of ways, but as with any muscular endurance function, it is influenced largely by training and recovery. Focusing on shoulder strength or fatigue is likely not a winning strategy. In pitching, the legs, torso, and arm all have to work in sync as well as mechanically, with the individualization making it difficult to generally focus on any one thing. This is where a good coach comes in, being able to use collected data to inform him about an athlete that he understands and can communicate with across training and game conditions. All that said, there does need to be a focus. If the pronator is the key to preventing UCL damage as it appears to be through both experience and research, it is both the strength of the muscle and the endurance of that muscle, as well as an element of resilience, allowing the muscle to handle the load necessary. Further research is needed to determine if the pronator’s muscular strength and endurance can be adjusted to further protect the ligament, or if the failure of the muscle is predictable or detectable, in order to know when to cease an activity. There is no battery meter on an arm, the way a smartphone has. There is no way to know currently if a pitcher is truly one hundred percent or if he’s down at 80. Studies have shown that velocity is a good proxy, but not an ideal one. Moreover, the perception of velocity – watching a pitch rather than relying on a radar gun reading – has shown time and again to be as accurate in terms of fatigue measurement. Research from Dr. Glenn Fleisig of ASMI, presented pre-publication at the 2017 SaberSeminar in Boston, indicated that “pitch cost” – the difference in stress between various pitches – is consistent, but individualized. Most pitchers have a continuum of pitch costs for fastball, breaking ball, and change, but that while the differences tend to be predictable within a range, the starting values and the intervals are very individualized. Early indications are that pitching fatigue is similarly individual, though the unknown and very confounding variable of training will clearly have an impact. Though we do not know, from the outside, what the specific training for a pitcher is, it is knowable, especially through the use of a device like the motusTHROW. Preliminary indications in that data set indicate that this sort of individualization is also present, but there are some other confounding variables that make this a more difficult problem. That problem is recovery. In general, it is a complete unknown for athletes, even at elite levels. Again, we lack that theoretical battery meter on any athlete. Coaches and associated medical personnel simply can’t say with any certainty and certainly not with any strong objective measure where a player lies on their readiness continuum, but for a moment, let’s pretend that we can. Imagine a pitcher comes into a start at his peak readiness. His battery is at 100 and he works to about 50. Assuming a consistent level of muscular fatigue across the system, the key shifts from ability to recovery. In whatever allotted period of time, the athlete is going to need to “recharge” back to as close to 100 as possible. Not taking on a full charge might leave a pitcher without his best fastball or might show more in the endurance side. An extended period of failed recovery could cause further problems. However, elite level pitching not only cannot calculate a true readiness measure, but fails to account for recovery in any usable manner. Pitchers are asked to recover based on a schedule, rather than on a most-effective basis. Starters are asked to go every five days, while relievers are thought to need less endurance and therefore less recovery, though the extra workload and stress may make that assumption very incorrect. This suggests that what high level baseball has is a resource management issue. Just like an airline needs planes fueled up and maintained, in the right place and with a flight crew that can handle it, a baseball team needs to have a collection of pitchers who are ready to go on any given day in the season, as well as having the mix of skills that allow a manager to exploit advantages. The proper management of resources would not focus on a schedule, but on maintaining the proper mix of pitchers to be available at the maximum opportunities, while never falling below a certain threshold. This resource management issue is solvable, but requires a measured, objective approach. Managing workload is the first key, while optimizing recovery is the second. Objectively and accurately assessing readiness allows a team or manager to then model out the expectation to avoid a resource shortfall while maximizing the available resources. How this is done is almost infinitely variable, but once again, it is individualized. All of the key measurement points and problems come down to measuring and then optimizing. Currently, no one is doing this because our ability to accurate measure these sorts of things have not existed. This is no longer the case. The resource management issue is also very complicated and individualized at two levels (player and team), but advances in both computational speed, computational availability and machine learning are making this less of an issue every day. Once we have these measurements, we can start working on individualized responses. Low stress pitchers might have advantages in terms of pitch counts and recovery. Players that recovery quickly or use technologies like MarcPro might have advantages in terms of usage and recovery time. Managing workload so that pitchers aren’t taxed beyond their ability to recover will become a key skill for pitching coaches and medical staffs. To once again use the battery metaphor, the pitcher of the future will need to not drain their battery completely and be able to quickly charge the battery up without significant loss of function. In this, the problem is not much different than what Apple or Samsung face in creating a consumer phone. Pitching coaches might need to learn as much from Tim Cook as they do Clayton Kershaw. Each of these pillars – measurement, optimization, and management – bear more discussion in the near future. I believe that it is clear that there are some strategies that pitchers and coaches can focus on in the near term. Focusing on readiness, optimizing and balancing workload and recovery, and gaining a resource advantage through smarter scheduling is ripe for both research and application. Have ideas? Want to discuss? Motus is interested in helping with research and training. Contact us online at @injuryexpert, @motusglobal, or firstname.lastname@example.org.