Analytics, Film Study, and the Human Element: Core Pillars of Basketball Analysis

Our own Nathan Grubel shares his personal philosophy around the elements of basketball analysis and why each pillar isn't more important than the next.

There’s nothing like current events to stir up a piece that I wasn’t expecting to publish for No Ceilings NBA.

Jaylen Brown…what a DIVISIVE conversation that’s taken place over recent days based on a quote from former NBA executive and current ESPN analyst Bobby Marks.

Marks, in an interview on SiriusXM, had this to say about Brown in terms of what he heard regarding Brown’s potential trade value from an anonymous NBA front office member:

“I had an analytics guy tell me, ‘We view [Jaylen Brown] as the seventh-best player on a TEAM.”

Well… that certainly stirred up the pot with many around basketball, including former players, analytics scouts, and media, weighing in on the topic with subjective opinions regarding Brown and how any analysis could come to such a conclusion.

Which is why I felt it was necessary to comment on what I believe to be my core philosophy as a basketball scout and analyst. If you’re looking for further discussion around Brown’s value, there are plenty of other pieces handling that specific topic.

My goal is to outline where I stand on the core pillars of basketball analysis: analytics, film study, and the human elements of intel and intangibles. The deeper I’ve dug into basketball, the more I’ve come to believe that all three of these elements MUST work hand in hand, and one isn’t more important than the other.

In terms of how I’ve used these different components to analyze the game in the past, present, and future, here’s the line I’ve walked when using and combining these different pillars: analytics and data help ASK the questions relevant to a player or team, the film helps me ANSWER those questions by providing the necessary on-court context that I can derive solutions from, not just confirm the problems present. And finally, the intel portion of analysis, i.e., gathering information about a player’s specific situation, role, and any factors off the court, works hand-in-hand with the other two elements to provide additional context that could be beneficial in actually solving the problems at hand OR suggesting a potential solution.

This is where basketball analysis gets a little complicated nowadays. There’s this immense focus on debating which player is better than the other, or who is actually good enough to stick around the league—or better yet, the one I’m most guilty of: which player should be ranked higher on a big board, or is deserving of this financial contract?

These are all fine conversations and studies to engage in, as it’s a way for fans to interact with the game and other fans. Obviously, these are real-life problems teams and executives are trying to solve that affect the competitive livelihoods of themselves and everyone around them within the organization.

What can get lost in these debates or studies is actually FINDING a solution to the problem at hand, rather than just pointing out what’s wrong with a player.

If an NBA player is a bad shooter, how can we identify that while also offering a potential solution to the problem?

Using this as a baseline example, each of the three core pillars I outlined above can help solve this problem, or at least present solutions to said problem.

Analytics are an incredible set of tools to utilize in basketball analysis because they’re able to quantify basketball at a large sample size that’s far quicker to complete the analysis of than purely trying to watch as many basketball games as it may take to solve a short-term problem or complete any historical analysis. There is no human being who can watch the amount of basketball from start to finish that it takes to compare shooting metrics across the entire league within a season, or multiple seasons, to identify patterns and come to reasonable conclusions. That’s just a fact. And it’s a great point that data scientists have repeatedly articulated.

So, having data, even just for one player that’s able to capture a large sample size of shooting metrics across various plays and shot types, is critical to solving the issues at hand because said data gives you an ACTUAL starting point for further analysis. Where does the player struggle shooting the most? Are there touch indicators anywhere else that suggest a foundation of potential shooting success?

Those numbers can help you ask the right questions and point you in a clear direction for further film analysis.

When watching back the tape of specific situations that have been identified for said player in our example, one can start coming to conclusions around the context of WHY a player has shot poorly. Is there a mechanical issue with the player’s jump shot? Is the player not taking high-quality shots, and WHY is the player continuing to find themselves in those situations if that’s the case? These are things that analytics models can help you identify, but they can’t provide the same context as actually watching back the tape to figure out how to solve the problem.

The final component is the trickiest for those not working directly in basketball to actually apply to their analysis, but the longer I’ve been around the game, the more I’ve understood how critical it is to actually understand what TYPE of solution to propose in order to solve the problem.

Intel and context around a player’s situation, be it his teammates, coaches, staff, and people around him, past, present, and future, can all help to contextualize what needs to happen next in solving a problem. If a player has struggled to convert a high percentage of catch-and-shoot shots, has that player been put in that type of role before? Has he been taught how to operate and work off screens, relocate, and play in that context? Or has said player always grown up with the ball in their hands and has only really been an off-the-dribble scorer? What type of role does the staff expect him to play? What is THEIR developmental plan for said player to improve in this regard? And what is the player’s support system and work ethic like to actually put in the time to fix the issue?

I can present a potential solution to helping a player improve in this regard, after having properly identified the areas of shortcoming using the data and watching the film to understand the on-court context. But if said issue isn’t a priority to the player’s current on-court role, OR those around him have other plans and priorities for his development, then will the issue itself even be actioned with a proposed solution?

Not only that, but actually fixing and improving elements of a player’s shot is generally not a 1-2-3 process and requires several layers of work that focus on balance, mechanics, and mental conditioning in terms of confidence and composure, actually letting those shots fly in opportune situations.

By now, you’re probably thinking I went down a convoluted rabbit hole in trying to present my personal philosophy around where all of these elements coincide with one another. But if you’re left asking more questions about just one example above, rather than feeling like I’ve answered something for you, then you’re exactly where I want you to be. Basketball is simple in nature, but complex and layered when actually breaking down what all goes into succeeding at a high level.

And NO success comes in the league without utilizing all three pillars to identify and solve basketball problems. If you’re a player who is focused on the on-court work, I’d venture to say you have coaches or developmental staff who are analyzing the data AND film to outline a plan and help you execute it to better serve your team and win more games. Coaches in the film room are being provided with data to better map and route where they need to focus next when presenting game tape to players. Data scientists are most certainly receiving feedback from coaches and players on what they feel are improvement areas, and their ideas around where to focus next, and data scientists are taking those suggestions and using the data to find those areas of opportunity, which get the feedback loop going again because THAT is where the questions come from!

All three of these pillars are working in unison to create a feedback loop. Basketball is played by humans, so the intel and organizational context are as critical as what the film says and what the data shows. But without the data, how would we be able to perform macro and micro analysis on areas of improvement in the time people actually have to prepare and train? Without the game tape, how would we have any context behind what the numbers say, and how to properly execute a game plan put in place by a coaching staff?

I say all of this to come back to one key point: NONE OF THESE ELEMENTS IS BETTER THAN THE OTHER! Use of one by itself doesn’t leave anyone with the type of analysis needed to actually SOLVE basketball problems at hand, which is what we all should strive to focus a bit more on than just high-level conclusions, myself included.

Will this piece solve the great debate between “Spreadsheets” and “Eye Test”? Absolutely not. Hopefully, I’ve at least sparked some ideas in anyone reading this as to how they can grow and evolve their process to get closer to solving basketball problems than just saying “Player X is a bad shooter” and stopping there. I’ve never been one to try to swing a room on an argument in 30 seconds, because if I’m able to accomplish that, said room probably didn’t have much conviction behind what was bandied about in the first place. What I always hope to do is inspire thought behind a subject and provide a new way to think about a topic or issue at hand that leads to a self-found answer and deepens one’s personal philosophy. And I know I’m not alone on that front, as all of us at No Ceilings NBA strive to do that very same thing.

So to bring us full circle, I ask you, the reader, this question: if you have a strong conviction about a player, what is the WHY behind the conviction, and HOW can you help solve the problem? If you were in the room with said player, or a member of his staff, what would be a potential solution to the problem beyond just pointing out the problem?