fbpx

Fatigue Units: A Conversation with BDG Research Director Dr. Mike Sonne

by Jan 24, 2019

 

Last week, Baseball Development Group had it’s preliminary 2019 ‘Research Meeting’ to discuss our upcoming plans, projects, and talk shop about what we want to accomplish in the next twelve months (there are some cool things that we can get into that later on in this blog). 

At the end of the meeting, we decided that it would probably be a good idea to have a conversation with our Research Director, Dr. Mike Sonne (@drmikesonne), to go over his background, what he’s looking into, and our future endeavours at BDG. Here’s a copy of our conversation – buffed up afterwards with some editing. 

 Dr. Mike Sonne has a PhD in biomechanics from McMaster University in Hamilton, where he studied muscle fatigue and how it influences performance. In the last few years he’s gotten very involved in baseball; having been a recurring guest to Pitch Talks and the Fan 590 Sportsnet, writing for The Athletic and his blog, and now working as the Director of Research with us!

[Steve]  Alright Mike, time to get down to business and show off that brain of yours. Can you maybe elaborate a little bit on your PhD thesis, your interest in baseball and why you publicly boo the PA announcement of pitch clocks at baseball games? (he actually did that with me)

[Mike] Sure! I started my degree in Human Kinetics, thinking I’d full well get into the field of Athletic Therapy. I had been involved in AT as a student trainer, all throughout high school. By the end of my undergrad, I took a bit of a different turn, and got into the world of ergonomics – and primarily, coming up with new ways to evaluate the risks in how people work.

In my PhD work, I studied muscle fatigue. (You can find Mike’s research here) The goal of this work, was to come up with a model that could be used to evaluate jobs, and determine which ones would cause people to fatigue more. So, I did a bunch of studies using different techniques to evaluate fatigue. We ended up modifying an existing model, and integrating the physiological principles of motor units, to help better predict fatigue. The final product was a model that required a series of muscle demands, and specified rest breaks, and we could get an idea of how much fatigue a muscle had incurred.

 I’ve always been a big baseball fan. I pitched a bit in the earlier part of high school, but my mid 50’s cheese didn’t really hack it. My family have been massive Blue Jays fans for my whole life. As I got further into the fatigue research, I found some interesting articles that evaluated the muscle demands of pitching, and felt I could use those to assess fatigue – using that same fatigue model – in pitchers.

 

A common issue that we have in ergonomics, is when engineers add jobs to the assembly line, thus decreasing the amount of rest workers have while doing their job. The demands themselves, are acceptable still – they don’t exceed the strength demands of the workers, but now they don’t have enough time to recover. Over the course of an 8 hour day, the fatigue levels increase, and the risk of injury increases.

When I heard that baseball wanted to implement a pitch clock, I decided to use the very same fatigue model to see what impact that would have on current MLB pitchers. It turns out, we could expect to see a significant increase in fatigue if the clocks were implemented. In the era of player safety (eliminating home plate collisions, take out slides at second), and a massive increase in elbow injuries – it made absolutely no sense to me that MLB was thinking of implementing this rule.

 

[Steve]Logically this makes a lot of sense. You’re asking pitchers to complete the same amount of work in a smaller time period. But then again, this line of thought can escape baseball  in even the most ritualistic tasks. How many coaches use a stopwatch to ensure appropriate rest intervals in between pitches in a bullpen? Or progressively tweak their work:rest ratios throughout the off-season? What about taking all those hacks in a short amount of time during batting practice? These are things that we think about all the time.

Your investigation of fatigue modelling is an interesting avenue, especially what you’ve done in creating an algorithm for Fatigue Units (FU). Can you explain this model a little bit for us?
 

[Mike] Well, in the first example – how many times are coaches getting on their pitchers to tell them – ok, now you need to throw again… RIGHT NOW!!! In ergonomics, we commonly rely on the field of psychophysics – which is, if we train someone well enough in a task, we can rely on their inner physiological, psychological, and biomechanical feedback to say “yes, I have recovered enough to perform another effort”, or “this is the maximum amount of force I could perform in a given amount of time”. The same thing goes for pitchers – how many pitches have you thrown in your life? I think it’s safe to say, by the time someone makes it to the MLB, their body is pretty in tune with what it can, or can’t do, and to force them out of their natural pace that they have self selected – that’s counter productive to player health and well-being. 

 
The inspiration for the Fatigue Units model was the disconnect between traditional workload metrics, and UCL injury trends in Major League Baseball. Selfishly, it was my academic backlash against the Blue Jays threatening to send Aaron Sanchez to the bullpen in 2016! If you look at the injury rates of relievers vs. starters, the relievers have a rate of UCL injury more than double than that of the starters. Aren’t limiting innings, and lowering pitch counts supposed to protect the pitchers? I tried to incorporate a few items from a paper by Whiteside and colleagues (Predictors of UCL reconstruction – you know it – yes we do, we wrote a Research Review on it) into a metric that built on my pitch clock fatigue model, and integrated research on UCL injuries.
 
1. The first step was to look at what the predicted fatigue would be for a pitcher in a given inning, given their number of pitches, and the amount of time they take between pitches.
2. I put a weighting on the amount of fatigue incurred in a given outing based on how many days of rest the pitchers had before that outing. For 5+ days, the weighting was 1, for 2-4 days, the weighting was 3, and for 1 day or less (there were a couple double header appearances!) the weighting was 5.
3. I had used some data that was made publicly available online from Driveline baseball to examine what happened to UCL stress when pitchers threw harder. It turns out, there is more stress on the UCL the harder you throw – so, i weighted the velocity with respect to league average, and multiplied each pitcher’s velocity by the sum of their fatigue units. The result was the Fatigue Unit Metric!
 
This is still in its infancy, and extensive testing is still required. The biggest challenge right now is accurately reflecting demands in young pitchers. Once I find a way to accurately reflect minor league pitching demands, then I think we can look at fatigue units as a predictor of overload injury.
 

[Steve] So FU’s take into consideration the number of pitches per inning, the pace at which they’re thrown, the number of days rest in between outings, and the velocity of the pitcher. I remember when you and I first talked about this, years ago now, I brought up the challenge of trying to customize this to a player’s previous throwing history. 

 
Is it possible for a player to become adapted to their FU, even if it’s high? Is there a specific range or threshold in which we can say that pitchers are significantly more likely to sustain an injury? We know that avoiding acute workload spikes is all the rage right now, has your model lended any support for that idea?  
 
[Mike] I think one of the misunderstandings about Fatigue Units, is that a high fatigue unit number is directly correlated with a risk of injury. Of course, when we’re looking at MLB pitchers, those who have ended up at the highest level of fatigue units in a season all seem to have had injuries – and pretty severe ones. Those who were in the 90th% percentile of workload using FUs had a 300% higher chance of having Tommy John Surgery in subsequent seasons. 
 
However, in reality, you can’t make a major impact on a baseball team if you don’t have a relatively high workload. If you look at things like the more traditional workload metrics of innings pitched, or pitches thrown, Fatigue Units can give you insight into two players who have had similarly high traditional workloads, but which one had their workload managed better. For example, Josh Hader appeared in 54 games in the 2018 season, amassing 21.42 Fatigue Units. In 2015, Travis Wood also appeared in 54 games, but with 29.43 workloads. That’s 37% higher of a workload than what Hader had seen from a Fatigue Units perspective.
A player can most definitely become accustomed to their workload – which is what you’re saying with the acute workload spikes. Combining fatigue units with other more advanced metrics like acute to chronic workload ratio can give you insight into how a workload is managed. Back to the Hader example, he only appeared in 5 back to back games all season – which was essential to managing his workload, and preventing those large spikes.  
[Steve] Speaking of pairing FU’s with acute:chronic workload, I know that you’ve outlined how coaches and players can establish their own FU’s (check this document) but how can someone integrate them into their own workload monitoring? Or is that a project that you’re actively working on and it’s still a secrets  

[Mike] Using FUs alongside your traditional metrics, is a great way to compare workloads between your players. For example, the Josh Hader example. Let’s say you have two pitchers going into your team’s Tournament, and both have thrown the same number of innings – though one has been appearing on back to back days more frequently. Despite similarities in workload from an innings and pitches perspective, you can see which pitcher has had a higher workload. The same works for checking to see how efficient a pitcher is. Perhaps two pitchers have thrown 30 innings, but one has a 15% higher Fatigue Unit workload than the other – this pitcher needs to be come more effective in the number of pitches they throw per inning to reduce their own workload. 

While the excel sheet works great for now, we’re always looking at new ways to make it safer for pitchers, and easier for their coaches and parents to monitor their workload. We’ll keep working on new tools through the BDG, and everyone will know when they’re coming out! 
[Steve] To me, utilizing this sort of model in conjunction with appropriate pitch guidelines could really help the youth pitching problem. Hopefully this is something that can be worked into the new Baseball Ontario pitch count guidelines that you’re on the advisory board for. Otherwise, we’ll have to consider cranking out an adjustment to the workload document we currently have running.  

The Baseball Ontario Pitch Guidelines are going to change in 2020 to be much better aligned with contemporary concepts regarding workload management. Dr. Mike Sonne is on the advisory board for the committee addressing these changes.  

I want to side step a little here and get into what I think to be the best article (and idea) that you’ve shared with me – that throwing weighted baseballs may reduce joint rotational stiffness. (The post I’m referring to is called “A Theoretical Concern for Weighted Ball Training and Workload Reductions“. The idea being that by implementing heavier balls into a regular throwing routine, we may first increase co-contraction around the UCL before adapting to the stimulus and then reducing co-contraction / stiffness and leaving the UCL vulnerable.  I’ll let you get into it. 

[Mike] When it comes to the arm care guidelines with Baseball Ontario, we’re going to use all of the available scientific literature to propose some new ideas, but then rely on the expertise of coaches and trainers to help us achieve successful and reasonable implementation. These things only work if everyone is on board.

So, joint rotational stiffness is one of the ways that our body protects itself from joint injury. If you want to read more about it, look up some of the great literature from Joshua Cashaback, Michael Holmes, or Steve Brown. Essentially, it’s the ability for a group of muscles to contract surrounding a joint to protect that joint from injury – things like ligament damage or capsule injuries. As humans, when we move in novel ways, there’s typically a greater amount of that joint rotational stiffness – take a look at your friends who haven’t skated in 20 years when they first get out on the ice. People don’t move very smoothly, and they tense up. This makes it next to impossible to perform an action expertly. Experts on the other hand move very smoothly, with only the bare amount of co-contraction surrounding the joint required to protect it.

Co-contraction provides stability at a joint, but with too much of it comes rigidity in movement. 

The theory that I have, is with pitching exercises like weighted baseballs, and throwing with maximum intent. To achieve absolute maximum velocity and movement speed, you’re going to have to turn off some of those safeguards – joint rotational stiffness has to decrease. While we’ve seen some studies citing that weighted baseballs don’t increase the stress on the arm, the amount of co-contraction really isn’t taken into consideration. Dr. James Buffi – subject of Jeff Passan’s book “The Arm”, and current Dodgers employee, created a model that would look at the relative contributions of muscle activity to the stress on the UCL. What his model would show in this phenomena, is that overall stress on the elbow isn’t changed when throwing a weighted baseball, but the amount of force transmitted directly to the UCL would be increased. 

[Steve] Yes and that’s a key point for the general coach to keep in mind – that although we can use tools like a Motus sleeve to get a proxy for torque on the elbow, it does not tell us how much of that force is being directly ‘felt’ by the UCL. Surrounding muscles and soft tissues force absorption is something that is interesting…but not easy to calculate.  Speaking of Dr. Holmes, how is that Masters thesis coming along with Richard Birfer? Is he done yet?

[Mike] Richard just finished off his proposal for his thesis committee, has collected most of his kinematic and kinetic data, and is now refining the mechanical assessment took he’ll be testing with coaches. All the while, he’s running a baseball scouting company, and putting together a podcast! He’s a busy guy, and I’m sure teams are drooling over the idea of hiring him. He’s still gotta get his velocity higher than mine though.

[Steve] He’s a hard working guy with a bright future for sure. Looking forward to seeing the final touches on his study. 

Speaking of studies – why don’t you give a little run down of what we’re cooking up this year from a research perspective at BDG. I’m pretty excited to finally get the ball rolling. 

[Mike] BDG has always focused on making sure the human body is capable of performing the tasks we ask of it, and then finding the best ways to recover from those types of activities. Our first official research studies are going to be focused on this exact topics. 

We’ll be working alongside Dr. Mike Holmes and graduate student Ryan Bench at Brock University to examine the effect of different recovery protocols on subjective feelings of fatigue and exertion, as well as performance in bullpen sessions. We’re going to be looking at some new tools, including dual vector bands, as methods of recovering from resistance training and pitching.

We’ll also be working with Dr. Nick La Delfa and Dr. Adam Murphy at the University of Ontario Institute of Technology to look at the basic physiological and biomechanical factors surrounding pitching performance. Lots of videos get posted on social media talking about whether or not a pitcher should lift, or how much they should lift. Should they be focused on flexibility or strength? We’ll be capturing those factors, as well as other factors like co-contraction, on performance metrics captured by the Rapsodo system. It’s been a long time coming, but it’s time you got to meet Reviewer #2, Steve!

[Steve] We’ve talked about this for years so I’m pumped to start investigating the recovery protocols — as we both know I’m a bit skeptical of some of the common practices in the baseball world and look forward to seeing how things go!
Being able to research baseball questions in an applied setting is important to not only yourself and mine as individuals but BDG as an entity. We’re trying to solve baseball questions and having the ability to bring in experienced and esteemed researchers like Dr. Holmes, Dr. La Delfa and yourself to help us is a dream come true.
As for Reviewer #2, may have to wait for another time!
Thanks, Doc!