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.
R̶e̶b̶o̶o̶t̶ ̶M̶o̶t̶i̶o̶n̶ ̶i̶s̶ ̶a̶ ̶b̶a̶s̶e̶b̶a̶l̶l̶ ̶c̶o̶m̶p̶a̶n̶y̶…̶s̶t̶a̶r̶t̶i̶n̶g̶ ̶w̶i̶t̶h̶ ̶m̶o̶v̶e̶m̶e̶n̶t̶ ̶a̶n̶a̶l̶y̶s̶i̶s̶.̶
Reboot Motion is a movement analysis company…starting in baseball.
I’ve talked before about why we started in baseball. It was where we had the best chance of immediate success- due to 1) our background in the sport and 2) teams having the ancillary services to fill in the gaps in our first offering.
While MLB was and is a great first market, we knew it had to be just that…a first market. Thirty customers can only get a company so far.
So, the question was “where do we go from here?”
To keep it simple, we had two options:
Expand vertically- bring our work to college baseball, academies, and beyond
Expand horizontally- bring our elite biomechanics focus to other sports
This decision was a tough one. Moving vertically is intriguing. Our brand is strong, we know what maximizes performance, and we understand what coaches are looking for.
This is not true elsewhere.
We do not know what specific movements correlate to performance. Nor do we know how open front offices, coaches, and athletes will be to change.
We may have intuition. But we do not know.
With all these unknowns, why would we move to a new sport? Why not double down on baseball?
Because that is not who we are.
We are not a baseball company.
We are movement company.
Exploring Basketball
Being true to our DNA as a movement company allows us to compete where we are strongest- physics and biomechanics.
We want to help coaches help athletes perform better…and we want to do so throughout sports.
—
Recently, we’ve been exploring an expansion to basketball- which we specifically like for how different the problem is.
In baseball, the most important movements are max effort movements. How hard can pitchers pitch? How hard can hitters hit? Basketball- specifically shooting- is different. It is about control.
Additionally, basketball is a green space. As far as we know, no one is looking at biomechanics. Very few are even analyzing ball flight (or at least tackling the problem the way we would).
This means opportunity.
In baseball, we barely even look at ball flight- not because we can’t, but because people are already doing it…and doing it well.
In other words, it is not a customer pain point. But in basketball, it is.
Yet, there is a downside to green spaces- we have a lot to do before we get started:
Finding data: teams are not capturing data the way they are in baseball. While that will make selling harder when we get there, it more importantly makes testing and research more difficult today.
Teaching: coaches aren’t conversational in biomechanics. In baseball, especially pitching, the industry has accepted biomechanics as a player development tool.
We are fluent in biomechanics- but we cannot overstate how helpful it is having coaches who are conversational.
Learning: Our background is baseball. Not only do coaches understand us, we understand coaches. We have spent time learning what they look at, how they talk to players, and more. We are not fluent in basketball and need to be.
A Proof of Concept
With that said, we needed to get started. And the obvious place to start is shooting.
In the NBA, shooting is expensive. If a team can teach it- or identify it early- they can gain a massive advantage.
Question # 1 What makes great shooters great?
While correlation is not causation, it is a good place to start.
We set out looking for teams that could provide body data and ball data. We wanted to analyze how each player’s body moved, how the ball moved, and sort by the two categories teams care about: makes and misses.
We quickly failed.
Few teams have reliable biomechanics data, especially not the in-game data we really want. Even fewer know who we are.
The lack of data, combined with our NBA brand recognition being nowhere near where it is in baseball, meant we had to look elsewhere.
Luckily, we found some help with Inpredictable- a site that describes itself as a “general playground for sports analytics work”.
We discovered shot trajectory data we could use to get started…and did that just that.
We analyzed the vertical velocity of the ball from when the shot starts to when the ball leaves the players hand. We then divided groups into the top and bottom 40% of shooters (all data is from 2016).
The first thing we notice is a clear difference between the better shooters- shown in green- and the worse ones- in red.
The “good shooters” have both a lower peak velocity and a less intense dip afterwards.
Intuitively, this makes sense. We believe this gives good shooters a higher margin for error as they spend more time near their desired velocity window.
But what about the best shooter?
Steph Curry is an outlier. The most obvious difference is his quick release, which every casual fan already knows.
What is more interesting is- despite the quick release- he lines up his shot and gets it to his desired velocity (normalized to 0 here) better than the rest of the NBA.
This is far from statistical proof, but it is another nudge towards a lower variance of velocity being an advantage.
What This Means
Right now, all we can say is we think we are on to something.
Generally, the goal of any sports movement is to generate whatever velocity you need in a desired direction, while minimizing unnecessary velocity in other directions.
In baseball, you have the explicit goal of maximizing velocity in the desired direction (and may care less about minimizing elsewhere).
But in shooting, you don’t care about maximizing velocity. Instead, what you really want is to figure out the exact amount of velocity you need and only generate that. You want consistency. And you want the ball to hit the rim softly to have a better chance of going in.
The Next Step- Personalized Ball Data
We love that the early data backs up our intuition.
At the same time, this is just step 1 of many. There are so many things we do not know.
First, the main takeaways from this analysis are about deviations from a “desired velocity”. We assume that desired velocity is correct, which it may not be.
The first question we would want to know is: What is a player’s desired release angle and desired velocity?
For context, a shooter can modulate two things: release angle and release velocity.
A higher release angle gives a higher arc, which gives more basket surface area to aim at- but also leads to a higher approach velocity which could cause a bad bounce.
A lower angle means less approach velocity and less chance of a bad bounce- but less rim surface area to aim at.
The biggest piece of data we would need to get is a players’ make/miss pattern, which most teams have.
If a player is consistently missing slightly front and back but it’s evenly distributed, a higher release angle may give them more rim area to aim at. If they always miss short, the angle may be fine, but we may need to increase velocity (in the right direction).
While we are not sure what the answer is, we do expect it to be highly personal.
The Next Next Step- Body Data
Ball data is great at telling us why we missed. We can measure where a shot lands, its entry angle, its approach velocity and more.
But we should go deeper than what happens at the rim.
On ball data alone, the entry angle is nowhere near as important as the release angle…because release angle is what the player controls.
For example, 45 degrees may be a great entry angle... but how does the player release the ball to create that? It is going to depend on how far the shooter is from the basket, their release velocity, their height, and their shooting motion.
Focusing on entry angle is like telling a pitcher the key is having your fastball enter the strike zone at 95 mph with 20 inches of vertical break.
But how?
What is the body doing?
Taking this a step further, deeper insights would be generated by how someone’s body is moving.
If we see a player is constantly missing short, this may be due to a lack of momentum transfer from the legs.
If the player’s release velocity has too much variability, they may be off balance.
The more we move backwards in time, the more we uncover the true answers to improve performance.
Future Questions
Correct mechanics vs constant mechanics?
The analysis above is general. It looks at what “good” shooters do on average, “bad” shooters do on average, and what Steph Curry does on average.
For every athlete, there will be variability- especially when there are defenders running at you.
We need to explore not only what a specific shooter’s ideal mechanics are (for him or her), but also how much variance exists from shot to shot and game to game.
What’s the deal with backspin?
Due to friction, backspin pushes air particles trailing the ball downward, creating an upward force on the ball (tip of the cap to Newton’s third law).
In baseball, this is huge.
A fastball has a lot of backspin, and top pitchers can cause the ball to end up more than 20 inches higher at the plate than it would have if only gravity were acting on it.
While the ball still falls relative to its apex, it appears to the batter to be rising…hence the term “rising fastball”.
In basketball, we would expect a much smaller effect.
However, shooters do not miss by much. If someone is missing ever so slightly short, a half an inch of extra lift could make all the difference.
Additionally, backspin creates friction with the rim that stops the ball's horizontal velocity, making it bounce more straight up and down after contact with the rim.
From a physics perspective there is a lot of reason to think backspin = good…but we need to do a lot more digging.
Can we use this thought process elsewhere?
One thing that is great about analyzing a completely new motion with entirely different goals, is the takeaways transcend the sport itself.
While pitching is about maximizing velocity in the desired direction, shooting is a more nuanced, controlled motion.
But guess what? Pitchers care about control too.
In fact, we get asked all the time about helping pitchers improve accuracy.
In baseball, this is really hard to study because the catcher changes the target on every pitch, and we have little visibility into where the pitcher is aiming.
In basketball, we are safe to assume shooters intend to make the shots they take.
Therefore, we can use basketball as a testing ground to understand the underlying biomechanics that shape control, and bring these ideas back to baseball.
Just The Beginning
Entering basketball analysis is a big step. It is also the start of a playbook- as every new sport will bring its own unique challenges.
We need to enter every sport knowing 1) we do not have the answers but 2) we do have the ability to find them. We will always start by asking:
What correlates with success?
What is actionable to take onto the court (or field or ice or court or course)?
How best should the information be presented to the front office?
How about the coaches?
These are daunting questions- but also exciting ones- as we hope to help basketball take a monumental leap forward in player development.