If you’re a basketball fan, you know your team’s
big stars, head coach, maybe even the rookie bench warmers. But you probably know very little about the Ph.D. mathematicians
like this woman. Her job is to crunch
the numbers of the game. We’re all in sort of an arm’s
race to acquire information. And she’s turning basketball into a hard, cold science. My name is Ivana Seric
and I’m a data scientist. For many years, Philly was a tough place to be a basketball fan. But having reached the top eight in the playoffs for a
second season in a row, the 76ers right now are in a renaissance. Embiid fakes, Embiid down the lane. Just across the river is where the team’s been hard at work and it’s where I went to
go meet Ivana in April. This 125,000 square foot facility comes with all the features you’d expect from an NBA training complex. It also features the
office of people like Ivana who support the team behind the scenes. My job involves analyzing the data that we have from NBA games. My focus is mostly on
the coaching strategy. Coaches have a lot of intuition. They know the game, they know the players, but we try to complete
their picture of the players with data, which is hopefully unbiased. Ivana is part of the Sixers 10-person analytics team. The work of these statisticians
has already forced a profound change in
the game of basketball. Teams are taking more three point shots. Embiid for three. Even though we make
them at lower percentage than the shots closer to the basket, but they’re worth more points and the trade off actually pays off. Take a look at this clip from the ’80’s. See how all the players are
bunched up right by the hoop? Compare that with today. More and more of the action is taking place at the three point line. Of course, it’s not just basketball that’s been reshaped by analytics. Businesses of all kinds are
tracking a ton of information they never tracked before
thanks to new sensors, improvements in computing power, implementing cost and data storage. Data scientists are the people making sense of that information with a combination of statistics
and computer programming. The profession’s expanded
rapidly in recent years and comes with an average
salary of $130,000 a year. Much of the data Ivana
works with comes from technology the NBA adopted in 2013. Every NBA arena has cameras
that record the games and then from those cameras, they can extract player locations on a court. These cameras record 25 frames per second, so for each basketball game, there’s a million of throes of data. And that allows Ivana to analyze plays that were previously difficult to track. For example, take the pick and roll. It’s one of the most
important plays in basketball. It involves one player
setting up a human shield to help a teammate shake off an opponent. So we can look at each player, how much they run pick and rolls and how good they are at that and then we can select a player and see how often he’s going to pass
out of the pick and roll, how often he’s going to shoot. And how would that help you develop a strategy for the coaches? So we could, for example this player, he’s going to pass pretty frequently. We can say this to the coaches and they’re going to decide on the defensive strategy for the player. But this makes Ivana’s job look simpler than it actually is because she spends a lot of her time coding to extract the
information she needs. And I tried really hard
to get her to talk about what exactly she looks for. And there’s some stuff that you do that’s beyond just these interfaces that you showed me today, right? Yep. Can you tell me more about that? Um, not really. So that’s like a trade secret? Yes. NBA’s a very competitive league, so whatever can give us that
advantage, we try to keep it. When advising the coaches
with her analysis, Ivana has a decided advantage
over many of her peers. I started playing when
I was seven years old and I just loved it from the first day. I think ’cause the game is so dynamic and there’s so many different skills. It’s what I wanted to do since I was seven years old, really. I wanted to be a professional
basketball player. Yeah. At 19, Ivana moved from
her home in Croatia to New Jersey to go to college
on a basketball scholarship. And while she was a star
player on her Division I team, she was also an exceptional math student. Much of her life has been like that, balancing her love of basketball
with her love of math. I always thought I’m going to have to choose between the two. I went to graduate school and that’s when I really thought that,
okay, I’m really choosing one or the other this time. So I thought I really chose just math. The plan was to become a
researcher or a professor. Then, three years into
pursuing her Ph.D. in math, she heard that NBA teams
were starting to hire data scientists, but only 26%
of data scientists are women and Ivana didn’t like her odds. When I saw the job posting, I didn’t think I would
actually get the job. I thought because being a woman in such a male dominated field, they would never really consider
me or give me a fair shot. But this prediction
turned out to be wrong. She’s got both the technical
and basketball backgrounds, which is sort of the ideal mix. Her ability to capture
these complex insights and then share them
with players or coaches, with executives in a way
that makes sense to us is super valuable and
frankly, not that common. The next evening, I attended my very first NBA game. It was the last game of the regular season before the playoffs and
my chance to catch up with the ultimate
beneficiary of Ivana’s work, the Sixers head coach, Brett Brown. The team has had a great season so far. What role do you think analytics
has played in that success? I think it’s played a significant role in our success and many, many others. The NBA playoffs are going to
start in three or four days and immediately, we’ll get
an analytical assessment on the strengths or
weaknesses of an opponent. You can assess play
calls, good or bad ones, ones you should do more,
ones you should avoid. Really, I think it’s
going to continue to grow and play a significant role
in the design of organizations and coaching staff’s beliefs. That night, even though
none of the starting players played in the game, the Sixers
ended up crushing the Bulls. A month later, the
Sixers ended up advancing to the conference
semi-finals and suffered a tough defeat at the hands
of the Toronto Raptors. Is this the tie breaker? Ivana and her colleagues will keep looking for
ways to help their team do better in the seasons ahead. But the Sixers victories on the court won’t be the only way she’s
measuring her success. It’s really exciting to be able to show young girls that they
can actually have careers, but it also feels like
a big responsibility because if I don’t do well, it’s going to seem like a woman cannot do this job, because there are not so many of us. It’s a big part of what
motivates me every day to be able to show young girls that they can succeed in
STEM fields and in sports.

100 thoughts on “The NBA Data Scientist

  1. As the Philadelphia 76ers were gearing up for the NBA playoffs last month, I spent a few days with one of their data scientists who's playing a behind-the-scenes part in their recent comeback. Do you think that analytics is changing the way your favorite pro sport is played?

  2. Bloomberg is always coming up with interesting insightful stuff. Unlike others in Mainstream media outlets.

  3. Very interesting what the data scientist is doing. If only you wouldn´t give this gender victimhood to the whole reportage, it would be a great story.

  4. Hopefully I can get some kind of data analyst job before i graduate =( this looks real cool, but sadly I think ill need a masters.

  5. All that logistics stuff has won….0 rings. Both Houston and Philly are at home salty af. 🤷‍♂️

  6. Damn some scientists sound like robots. Probably doesn't help that she's from another country tho

  7. Solve the ben simmons 0/nth, 3 pt field goal. Only Ben wasnt able to adapt with players concentrating on the 3 pt field goal

  8. What is the probability that Kwahi would have made the Game 7 buzzer beater after ball bounces four times on the rim?

  9. You tried to imply that the three-point revolution came about because of analytics. If anyone has any doubts, it wasn't analytics that did it, it didn't even start with the Splash Brothers.

  10. Can you become this with just a Bachelor's in Engineering and a Masters in Data Analytics? Like not have to get a PhD? Because being a data scientist for the NBA sounds hella cool ngl

  11. Daryl Morey and the analytics folks have been doing this for the last 20 damn years. Sabermetrics, shot value, efficiency … and it hasn't won them a single damn Championship. Numbers only take you so far: in the end, it's all down to talent, grit and effort when everyone's arms are too damn tired to shoot the same 45 percent arc they do in the regular season. Not to say that it isn't useful, of course, but the team with the better chemistry and the better coaching adjustments beats a bunch of numbers adjudging tendency and conversion rates any day, every day.

  12. this data works better for MLB or NFL cause in the NHL/NBA the coaches usually go with gut feeling and a name like Reddick or Leonard. Though I will say that thanks to this data now every team has deadly 3 point average compared to 20 yrs ago where you had 1 or 2 players who could shoot from behind the 3 pt line.

  13. Sounds like overengineered bullcrap, unless you believe the past is always an indication of the future which is not how the world tends to work.

  14. All this data stuff is bullshit. Keep relying on data, and enjoy losing before the conference finals

  15. A made up job. Their are super fans who are more knowledgeable and can predict who has a better jump, a better pick and roll and etc.

  16. Having a PhD in math… 26% of Data Scientist in NBA are women… just wondering the amount of women having a PhD in maths…

  17. I would love a job like this. I'm working on my Master's in Business Analytics. Hopefully, I can land a job in something similar

  18. I think this video is sort of pushing gender inequality. Which is all in a persons head. Look at this woman. She thought she couldn't get the job. But she got it. I don't see any reason why a woman could not do this job or any less capable than a man. I'm a male, diesel mechanic. I wish I was a scientist, if I wanted to, I could be one. I'm too lazy to go back to college and study for it I guess. I also suck at math. But if anyone wants to do something bad enough and the possibilites are reasonable, then it's just a matter of time. The fact that she thought she couldn't get the job was all inside of her head. Also you have to look at percentage of applicants for this job. The video stated only 26% data scientist were female, but you need to look at how many applicants were female vs male. I don't know for sure, but I'm guessing that the majority of applicants were male, therefore more males hold that position because there were simply more males to have applied than females.

  19. I think this video is sort of pushing gender inequality. Which is all in a persons head. Look at this woman. She thought she couldn't get the job. But she got it. I don't see any reason why a woman could not do this job or any less capable than a man. I'm a male, diesel mechanic. I wish I was a scientist, if I wanted to, I could be one. I'm too lazy to go back to college and study for it I guess. I also suck at math. But if anyone wants to do something bad enough and the possibilites are reasonable, then it's just a matter of time. The fact that she thought she couldn't get the job was all inside of her head. Also you have to look at percentage of applicants for this job. The video stated only 26% data scientist were female, but you need to look at how many applicants were female vs male. I don't know for sure, but I'm guessing that the majority of applicants were male, therefore more males hold that position because there were simply more males to have applied than females.

  20. I think being a woman or female is already a disadvantage for them to compete to us men or males. Whether it is in sports or academics or even in tech and engineering. One big reason is that they are the one responsible for child bearing and mostly child rearing. I'm not saying it is not doable but it is an unfair disadvantage if they're competing with men.

  21. This is the type of stat person a team needs. She is someone who is a numbers genius but knows enough about basketball to understand that the numbers change in certain situations. For example 3 pointers are better then 2 pointers but its easier for defenders to guard the 3 point line then it is the guard mid range shooters. The thing is since she is a basketball player she understands this while others will say, "all you have to do is shoot 3s and lay ups"

  22. Those who can play coach; those who can't coach go in to journalism; those who are too pretentious for that get into analytics. Thus spoke Zarathustra…

  23. Why does she needs a PHD to do this kind of work? One does not need a PHD to do that. Most of the work is already done for her by the software. I do the same type of work and I do not have a PHD, overrated.

  24. USA will lost many of bright talents if they keep on limiting FOREIGNERS entering their country… She is a rare combo indeed

  25. I can fully understand but it is really amazing how data science have enlarged the importance of such small areas which were neglected in the past.

  26. how can I get a career like this? I specfically want to work in sports and performance. will a masters degree in computer science/ data science help?

  27. Since I'm also from Croatia, currently a Computer science student and started playing basketball since I was 7, this inspired me immensely. I even wrote Ivana an email to which she unfortunately didn't respond, so I was thinking maybe you guys have some tips and guidance how to specialize in this concrete field (which literature, courses, basketball-related tips, anything).
    Thanks in advance!

  28. She is a data scientist, and she can't tell that the reason there are less women hired as data scientists, is because less women pursue that carreer relative to men, and not because of a bias in the hiring process. She had equal chances when she sent her postulation than the men ceteris paribus.

Leave a Reply

Your email address will not be published. Required fields are marked *