Stealing From Yourself
How one analyst crossed over from baseball to basketball to drive innovation
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.
The following is the second post in a series on the Wake Forest Bridge Seminar, which was put on by Wake Forest Dec 17th and 18th 2022.
While nothing matches being there, I wanted to share what I learned for those who missed it, and to set the stage for an ever bigger turnout in 2023.
—
TJ Barra gave a different view than most at the Bridge Seminar because, well, he has a different view.
A few years ago, after 15 years in MLB front offices, Barra took on a new challenge as the Senior Director of Basketball Research and Innovation for the Milwaukee Bucks.
While he describes the recruitment pitch from Bucks’ GM Jon Horst more eloquently, he basically was tasked with “doing some of that baseball stuff in basketball”.
For TJ, this was the perfect role.
As he explains, he loves thinking outside of the box and stealing from other sports. In fact, he did this for the Mets nearly a decade ago when he saw the growth of geospatial data and how soccer clubs were using it to build passing networks.
Now was his chance to transport his baseball learnings to the NBA.
While his presentation dove deep into Bucks-specific-takeaways, I want to focus on how he set the stage: with 4 differences between the two sports.
Understanding how MLB and the NBA are different helped TJ cross the chasm and help drive decision making for the Bucks…and any sports tech company making a similar move could learn from TJ’s experience.
Difference 1: Infrastructure
Baseball R&D departments are much more established. There is more data available, more tools to work with, and larger staffs to do the work.
Barra mentions the Bucks having an 8-person R&D team that, while likely being one the NBA’s largest, is dwarfed by nearly every MLB organization.
The Sports Tech Lesson
The smaller an organization’s R&D department, the more saving time is of the essence. Vendors should maximize for ease of use- either via 1) immediate takeaways or 2) easily fitting into an organization’s work flow. Likely both.
Difference 2: Using Data for Player Development
Basketball currently lags behind baseball in terms of utilizing data for player development.
While the sport has made strides in moving beyond outcomes metrics to understand the why, the next step is breaking down How Data, specifically through skeletal data.
As of now, basketball mainly relies on shot tracking, which only shows the result of a shot rather than the mechanics behind it. This is something Barra expects to develop rapidly in the coming years.
The Sports Tech Lesson
This difference brings both excitement and caution.
Excitement is obvious. Compared to baseball, basketball is wide open. While many are using data to improve team outcomes, very few are using it to make the individual better. Doing so would give any team a major advantage…which means vendors who make that easier will create massive value.
At the same time, we need to proceed with caution.
Behavior change takes time. Even if a vendor can analyze ball flight, repeatability, balance, and more, nothing matters without organizational buy-in .
Tech companies should look for incremental wins when it comes to buy-in, and may need to start with improved tools for scouting or weight room development, which come with less of a demand for behavior change.
Difference 3: Individual vs Team Game
Next, Barra talks about how baseball is an individual game- pitcher vs hitter- while basketball is a team game with countless variables.
Here, I think there are two differences worth mentioning. First, in basketball, not everything shows up in the data. The best front offices can’t just analyze outcome data. They need to use their sport specific knowledge to make sure they’re asking the right questions.
Second, basketball is reactionary by nature. Given the countless variables, no two plays are alike.
Here, I disagree with Barra a bit. While the reactionary nature of basketball is different than baseball pitching, it is similar to what we see in hitting…
The Sports Tech Lesson
As Barra mentioned, when looking to solve a new problem, stealing from other sports is a great strategy. One of the things Reboot is looking to take with us from baseball is our use of control theory to analyze hitters.
In baseball, hitters predict where the ball will end up when it crosses the plate- and they usually start doing so before the pitcher releases it. By analyzing the bat path, we can estimate how strong those predictions were at different times in the swing.
While it isn’t 1:1, we can use similar logic to analyze how well a jump shooter is getting his body to do what he wants regardless of the situation.
Difference 4: Staffing
The final difference Bara mentioned is hands on player development.
Here, basketball has an advantage as the coach to player ratio is greater- mostly due to smaller rosters. This allows for more specialized player development, with dedicated coaches for 1) each position and 2) specific skills- shooting, rim protection, etc.
The Sports Tech Lesson
I will admit I am still unsure of my takeaway here, but I do have a hypothesis. In MLB, a lot of our work revolves around scale. We are not just dealing with 25 person rosters, but multiple minor league affiliates as well.
Coaches need to get their eyes on an entire organization, and do so in a way that does not take up their entire day.
In basketball, we may not be solving for scale- but rather for certainty. Coaches spend their entire year dedicated to working with a small number of athletes. If they can help one or two get 5% better, they massively improve the team’s outlook.
Therefore, the best sports tech may be a hypothesis testing engine, allowing coaches to double down on the changes they have the most confidence in.