Measure What Matters
Early thoughts on what we're learning about the relationship between momentum, alignment, and fastball velocity
Every other week I write an email discussing what I learn launching and growing Reboot Motion. If you would like to receive it directly in your inbox, subscribe below.
Whenever we talk with a potential partner, the first question we always get is some version of “What makes Reboot Motion different?”
While the answer is a few different things, one of the key focuses is our momentum based approach, explained by Jimmy Buffi here:
Using momentum over velocity gives better insight into an athlete’s movement, which allows Reboot to do two things:
Produce a comprehensive report, looking at an athlete’s sequence, the alignment of their rotation planes, the magnitude of their momentum peaks, their overall balance, and more.
Deliver a unique and valuable dataset for pro organizations to build on top of with their own proprietary insights.
While we can talk about the advantages of a momentum based approach until we are blue in the face (and we will), we know we need to do more to prove momentum should be the gold standard for baseball biomechanics…and eventually sports biomechanics.
Proving Momentum Based Biomechanics
To show momentum based biomechanics is superior to velocity, we are in the process of creating a “What Matters for Movement” White Paper.
Our long term vision is to:
Determine the key variables that drive performance generally
Embed these takeaways into our reports so coaches know what aspects of an athlete’s movement can be modified to yield the highest results.
For our MLB teams, use this as the basis to make individual models for each athlete.
We expect this work to take a bit of time and, in true Reboot Motion fashion, wanted to get our thoughts out to the universe as fast as possible. We expect to constantly tweak our models, update our conclusions, and look for more and more ways our partner teams can push the envelope on player development.
This post will focus on pitching, and we will follow up with additional posts as we dive deeper.
Early Pitching Takeaways
After a bit of cleaning, we took a first pass combing through pitching metrics to learn which ones correlate most to velocity.
The first thing we notice is metrics labeled “proj_max” are at the top of the list. These metrics are the peak momentum of that body part in the direction of the most important body part/object (i.e. the pitching hand or the bat).
As an example, we notice the thing most correlated to velocity is the peak momentum of the pitching hand itself. In other words, the faster the hand is moving, the faster the ball will move.
This is both 1) obvious and 2) unhelpful, as telling an athlete to move their hand faster will likely result in a frustrated pitcher.
However, seeing this as step one is helpful in a couple ways. One, it is a point in favor of the model doing what it is supposed to. And two, it allows us to remove variables that, while correlated with the true independent variables we are after, the athlete may have less control over.
Early Returns
After removing these variables, we were able to spot major areas that correlate significantly with improved performance.
Before diving into them, it is important to note a couple things:
The data shows correlation, not causation. However, our understanding of how the body works gives us confidence in a causal relationship.
Data is just information. We do not believe these metrics should be maximized without context. We trust in great coaches to know when tweaks should be made and how best to make them.
With that said, here are some of our early takeaways:
The Value of the Torso
Surprise surprise, the body is connected…and the center of it it is pretty important.
We noted that if we want to get the ball moving fast, we need the hand, and the throwing arm, moving fast.
This is obvious, so the better question is “how do we make that happen?”.
Since the ideal sequence for transferring momentum from the ground through the body to the ball goes rear leg → lead leg → lead arm → torso → throwing arm, it should not come as a surprise that 1) maximizing torso momentum and 2) aligning the torso with the throwing arm, leads to increased velocity.
Our early model suggests that both 1) a 12.5 meters per second gain in torso momentum and 2) a 12.5% gain in torso alignment, each correlate to a 1 mph increase in fastball velocity.
One Step Back in the Sequence
When possible, we like to move as far back in the sequence as possible. We believe the further back we go, the more the athlete can control and the more we understand how things happen.
And that is the case here. Yes, increasing peak torso momentum will help pitchers throw harder. But looking a step earlier, we see the same thing for lead arm momentum.
Specifically, we notice a similar 12.5 meters per second increase in lead arm momentum leading to a 1 mph gain in velocity.
Induced Vertical Break
The correlation between velocity and performance…and therefore the correlation between velocity and dollars is gospel in MLB.
We include this slide in many of our presentations, but assume front offices regularly rolls their eyes. “Yes, yes, we know!”
While it is not talked about as much, pitching coaches, pitchers, and front offices generally agree that, when it comes to induced vertical break- just like velocity- more is better.
IVB is the difference between where the pitch actually crosses the front of home plate height-wise, and where it would have crossed had it traveled in a perfectly straight line from release, but affected by gravity.
In other words, IVB is simply a measure of how much the ball’s spin caused it to rise and counteract gravity, and is important to pitchers due to the role it plays in “deception”- or how different the result of a pitch is from what the batter expects.
Rotation Planes
When looking at the data, we again see what we would expect from our understanding of physics: steeper rotation planes lead to greater induced vertical break.
Taking a step back, we measure each body part’s rotation plane as the tilt of the plane in which the body part is rotating and therefore generating momentum.
A completely vertical plane would be 90 degrees, while a sidearm pitcher could see their throwing arm below 0 degrees.
Below, we see a pitcher whose pitching hand (the thing most closely related to the thing we care about- the ball) is extremely vertical, at 83 degrees.
The data shows that the more vertical a pitcher moves, the more induced vertical break they will create. Also, unsurprisingly, the effect is greatest in the pitching hand, where a 6.25 degree increase will result in an additional inch of induced vertical break.
(It is no surprise that the plane of the torso, and then the lead arm, are the next most impactful as that is the reverse order of the desired pitching sequence.)
The final thing to note here is our confidence that this relationship is a causal one, despite the data only proving correlation.
The more vertical a pitcher’s hand, the more the ball will spin in that direction. And that spin is what causes the ball to lift, counteract gravity, and appear to the batter as if it is rising.
It’s just physics!
Next Steps
We are just scratching the surface on better understanding “How Data” and its impact on performance.
While the questions we can ask are limitless, this initial work brought up two key areas to look into:
What would introducing more variance do?
Currently, we are looking at pro pitchers who, more or less, do most things right. There may be more low hanging fruit as we enter the college and high school ranks.
Can we measure the tradeoff between velocity and induced vertical break?
All the work above was done “holding all else equal”. We know that is not how the body works.
Today, we can show that increased alignment leads to higher velocity, which, in a vacuum, is good. We can also show that the more vertically a pitcher rotates, the more their fastball will rise, which, again, is good.
However, we notice that alignment generally goes down as the pitching hand rotation plane increases, which reduces velocity. This is mostly due to it being a less natural movement, based on how one would have to tilt their spine.
The pitcher below (the same one posted above) shows what we are talking about. He does a great job, but it is easy to see 1) how much his spine is tilted, 2) how we would have to exaggerate that even more if he wanted to increase his ball hand rotation plane, and 3) how difficult replicating that motion may be for most pitchers.
Today’s Conclusion
Early returns tell us:
Moving your lead arm and torso faster will increase velocity.
Aligning your torso with your throwing hand will increase velocity.
Moving your pitching hand, torso, and lead arm in a more vertical plane will increase induced vertical break.
None of these are revolutionary. In fact, they are anything but revolutionary- they are incredibly intuitive. But that is what is so exciting.
Lead arm and torso speed obviously matter.
Alignment throughout the rotation obviously maters.
Verticality obviously matters.
How the body moves obviously matters.
And if you are a coach, you need to measure what matters.