Privacy & Security

Analytics: 4 Lessons Schools Can Learn From the NBA (and Vice Versa)

By Benjamin Herold — January 13, 2016 8 min read
  • Save to favorites
  • Print

Our overlords at Google and other companies tracking users’ web behavior undoubtedly already know that I spend a lot of time reading online about two topics: education and pro basketball.

Typically, those interests are pretty divergent. But for Education Week’s new special report on fresh directions in personalized learning, I had the chance to talk extensively with Benjamin Alamar, a former National Basketball Association executive who is now the director of sports analytics for ESPN.

In nearly every sector of society, there’s been a push to collect ever-more data, then analyze that information in ever-more complex ways in search of insights that can help guide (hopefully) smarter decisions. The analytics movement has already transformed the NBA. An analogous revolution in K-12 schools is progressing in fits and starts.

While every industry is of course different, there are parallels across fields. Alamar and I talked about lessons that schools might learn from the NBA—and vice versa.

Lesson 1: In a world of limitless data, it’s crucial to determine what metrics really matter.

In pro basketball, there’s been a tremendous shift towards valuing efficiency over raw production: A player who scores 15 points on 10 shots is likely to now be seen as a more valuable contributor than a chucker who scores 20 points on 25 shots.

NBA teams are also investing heavily in evaluating players based on their progress honing specific skills and handling specific situations. And there’s a clear bottom line—wins and losses—against which everything else can be measured.

The K-12 education system, however, is still struggling to operationalize success. There’s growing dissatisfaction with standardized test scores as the be-all-end-all outcome measure. But while there’s plenty of research to support newer (and older) notions about what constitutes student learning and engagement, schools still have a long way to go to identify clear, feasible metrics for determining when systems, schools, teachers and students are doing a good job.

Lesson 2: Being smart isn’t the same as being effective.

Even the best analytics-driven insights can end up collecting dust on a shelf if key decision makers don’t see their value. It’s a lesson that Alamar said he had to learn on the fly when he made the transition from academia to working for the NBA’s Seattle Supersonics (now Oklahoma City Thunder.)

One example: Alamar at one point was testing ideas to measure prospects’ “basketball IQ” and ability to anticipate plays before they happened. Rather than presenting a theoretical model or requesting the budget for a comprehensive study, he developed a quick-and-dirty video-based game, which he then gave to team executives to play during their down time. They loved it—and that turned into support for his efforts to build and use a more complete version.

It’s a lesson that many ed-tech companies would be smart to heed. Just because analytics make sense to you doesn’t mean that a busy, cash-strapped superintendent or principal is going to find a reason to give a new tool or strategy a try.

Lesson 3: Data are most valuable when they can be looked at together.

For years, Alamar said, even the most forward-thinking NBA teams housed critical data—such as on-court performance results, scouting reports, salary information, practice reports, and player health records—in separate silos. But over the past several years, there’s been a huge push to bring that information together in one place, so it can be easily accessed by key decision-makers and new insights can be gleaned by looking at information side-by-side.

That’s still an uphill battle for most school systems, particularly when it comes to merging the learning data generated by the dozens of software tools, games, digital content, and apps with which students often interact. Systems such as the Houston Independent School District are pushing for new solutions to this problem, and third-party efforts by groups such as Knewton and the now-defunct inBloom have sought to shake things up by creating central warehouses that would do some of this work for schools. But there’s still a long way to go to overcome the technical, privacy, and bureaucratic hurdles that have stymied such efforts to date.

Lesson 4: There’s always more information to be found.

NBA arenas now have sophisticated camera systems to track player movement during games. The inspiration? Israeli missile-defense technology. Teams have also invested heavily in understanding the recovery process from various injuries. Biometrics are also huge: Players on many teams are now outfitted with wearable devices that track their exertion, nutrition, rest, and more. The Milwaukee Bucks even retained an expert in understanding facial expressions to help evaluate the psychological makeup of draft prospects.

In the education sector, meanwhile, groups such as AltSchool (which I featured in our recent special report) are seeking to bring such techniques into the classroom. But the discussion about whether that is appropriate, valuable, and even legal is only just beginning.

Lesson 5: Have real conversations about privacy.

This is one where education is light years ahead of the NBA. The involvement of children and the existence of laws such as FERPA play a major role, and parent advocates have fought back hard against some efforts they believe go too far. Lawmakers are listening: For the past two years, Congress and state legislatures have been wrestling with the challenge of how to protect students’ sensitive information without unduly limiting the potential for innovation.

To date, though, privacy has been mostly a non-issue in professional basketball (and other sports.) But there’s a good chance we hear a lot more in the coming years. The rules about what data teams can collect on players, and what they can do with that information, are governed by the league’s collective bargaining agreement with the players’ union. Given the huge growth in the collection of biometric and other data in recent years, Alamar said, expect those rules to be a hot topic when the CBA is renegotiated, possibly in 2017.

An even more sensitive subject, Alamar said, has to do with what information teams can collect from colleges and universities about the players they are considering drafting. The techniques used in the NBA are filtering down to the amateur ranks, but laws such as FERPA and HIPAA make it a very touchy question about whether and how the resulting data can be shared with potential employers.

Lesson 6: Giving learners access to their own information is powerful.

With the rise of the personalized learning movement in K-12, it is now the norm for digital curricula, software, games and content to provide students with ways to see their own progress. The general theory is that having access to such information will help students better understand not only what they’ve learned, but how they learn best. In turn, proponents hope that will increasing students’ motivation and sense of ownership over their own learning process, accelerating positive results in the process.

In the NBA, however, teams’ analytics staffs still tend to focus on providing information to intermediaries, rather than directly to players.

“There’s a belief that coaches are the best people to deal with players,” Alamar said. “When an analytics group comes up with a piece of information, it’s competing with so many other things. [There’s a belief that] the coach has to act as a filter for what can be put in front of players.”

Lesson 7: How data are presented is critical.

Perhaps because there is such an emphasis in the K-12 sector on presenting information to students and teachers, vendors have paid a lot of attention to creating well-designed, user-friendly analytics dashboards. The result has been an influx of new tools that synthesize a wide mix of information in smart, intuitive ways, all with the goal of helping make it easier for users to make decisions such as how to group students for a particular lesson. My colleague Malia Herman did a great job covering this trend in Education Week’s new special report.

Alamar said such work is only now taking off in the NBA.

Teams “are definitely concerned with how people are interacting with information day-to-day,” he said. “Anything that lowers the barrier to getting people involved with data is a good thing.”

Lesson 8: Open can be powerful.

This is one where neither sector has shown much progress.

The education sector has seen huge growth in “open” educational resources that are licensed to be free to use, share, and revise. The U.S. Department of Education is among the groups pushing that growth, touting OER as a valuable tool that can help level the playing field between districts.

But there hasn’t been a similar push in K-12 and the ed-tech sector to also make “open” what is happening under the hood. In 2015, for example, tech company Knewton announced that it would “power” thousands of OER items with its analytics engine, but the news was greeted cautiously in the open community, largely because Knewton would not commit to making open its algorithms.

And the NBA?

“There’s more and more information in the public domain all the time, and there’s a lot of exciting stuff that I’d like to share with fans here in my role at ESPN,” Alamar said.

“But teams keep their most advanced [and proprietary] stuff behind a wall. They’re looking for a competitive advantage.”

Photo: Houston Rockets guard James Harden warms up before an NBA basketball game on Jan. 12, in Memphis.--Brandon Dill/AP

See also:

A version of this news article first appeared in the Digital Education blog.