Jeremy Lin remembers the packet. The Houston Rockets distributed it at season's end, an ocean of personalized data slimmed down to a few pages. It showed that the point guard was one of the NBA's best at driving and making plays at the rim, but that he also struggled shooting from the left wing and 3-pointers off the dribble.
"Things like that, I didn't know," Lin said. It helped shape his offseason training regimen.
Lin is savvy on the subject, one he has been interested in dating back to his playing days at Harvard. He said his agent even doubles as a personal analytics assistant. "I'm not going to overreact to some numbers," Lin said. "I want to know what they are, though."
The franchise catered to Lin's interests well. The Rockets are not only considered to be one of the NBA's most aggressive teams in the field, but also, perhaps, in any sport on any level.
On the other end of the spectrum, there's the team the Rockets traded Lin to last offseason, the Los Angeles Lakers.
"[Byron Scott] told us a couple stats," Lin said, "but I don't know if they're necessarily that deep into analytics. They were stats about our efficiency when we score in pick-and-rolls versus isolations and some defensive numbers. But besides that, I haven't seen that much."
He's not alone. Although teams guard the inner workings of their analytics operations as if they were protecting nuclear missile launch codes, there's almost no public and little private information about what -- if anything -- the Lakers have done or are doing on this front.
"If you ask the analytics people who work in the NBA, 'Who does work for the Lakers?' Nobody knows," said Ben Alamar, director of production analytics at ESPN and a former analytics official for multiple NBA teams, including the Oklahoma City Thunder and Cleveland Cavaliers.
"It could be that they're being really secretive and that they're really good at being secretive. But they haven't hired anybody that anybody has respect for in the analytics community."
His point was echoed by several analytics officials employed by NBA teams.
For years, the franchise has appeared afraid of change, or arrogant, or both. The Lakers didn't send a basketball operations representative for the first six years of the annual MIT Sloan Analytics Conference in Boston, the latest incarnation of which will be held Feb. 27-28. In 2013, they were the only team without a representative.
Lakers officials see it a bit differently.
"We sent people, but they weren't basketball people," Lakers general manager Mitch Kupchak contends. "Nobody has known about it. It's better to go under the radar, if you ask me. That's just the way we do it."
Lakers assistant coach Mark Madsen acts as a liaison between the coaches and analytics crew. When asked how the Lakers' analytics operation compares to others from around the league, a smile spreads across his face.
"All I can say is this operation here is second to none," Madsen said.
However, interviews with those around the league and considerable evidence suggest that the Lakers' progress with analytics matches their record this season: almost dead last.
The staff
The Lakers say they have six people working on stats, four of whom focus on data provided by SportVU, the six in-arena cameras located in the rafters of every NBA arena that use Israeli missile-tracking technology to record the movements of each player, the referees and the ball 25 times per second every game.
Two of those SportVU people are based in L.A.; two others -- including the Lakers' top SportVU consultant -- are based in the home of the Lakers' biggest rival: Boston.
"Just so you know, he's born and raised in Southern California, not Boston," Kupchak said of their top SportVU consultant, whom we'll call Jack. (The Lakers declined to reveal the identity of their analytics staffers, provide much information about them or allow them to be interviewed.)
Jack attended graduate school at UCLA with Kupchak. "He's somewhat of a basketball fan," Kupchak said. They stayed in touch over the years and talked about analytics, eventually discussing the type of person an NBA team needs to hire for such a task. The topic came up as SportVU started to become more and more available; all 30 teams had access to it beginning in the 2013-14 season. "And then he was able to say, this is the kind of person that you need," Kupchak said.
Kupchak said he hired Jack about two years ago.
Jack keeps in touch with the two SportVU staffers in L.A. through conference calls and visits. The other employee in Boston, Kupchak said, is tasked with helping break down the data into a more digestible form. The Lakers' other two analytics staffers are former Lakers coach Rudy Tomjanovich and his son, Trey, both of whom are based in L.A. Kupchak said Rudy has worked in a "statistics-based analysis role" for several years with the team, working with more conventional stats, such as information found in box scores, and also using NBA.com's StatsCube, the league's official advanced statistical analysis tool.
Another element is Lakers executive vice president Jim Buss, who in a 2012 interview with ESPN.com said he has a "defensive rating, basically, that involves the offense, defense and then the impact depending on how much they play." It's a formula that Buss said then that he developed himself. "I'm up to all hours of the night doing numbers. Trying to tweak it. Trying to get it better. Seeing if it truly [corresponds]," he said in the interview. It's unclear what involvement -- if any -- Buss has with the Lakers' day-to-day analytics operations. (The Lakers declined to make Buss available for an interview.)
The coaches
One of the tasks the analytics staff faces is gathering information requested by Scott. The first-year Lakers coach said he met with the staff before the season and told them that on a day-to-day basis he wants to know four things, though he declined to identify them. "It's really that simple for me," Scott said.
Scott, a former NBA guard, said he falls somewhere in the middle when it comes to value of analytics in the game today. He calls himself "old school." He added, "I'm not so old school that I look at that stuff as a bunch of garbage. I think it has a lot of value. But I'm not so into it that I think that it's the best thing in the world."
Scott appeared particularly old school earlier this season when he said he didn't believe that 3-pointers win championships. Indeed, the Lakers are 29th in 3-pointers made this season, attempting fewer per game than three of the past four NBA champions. This appears to be ignorant of massive changes in how coaches use the long ball. This season is on track to be the first in which, league-wide, 3s outnumber free throws. As recently as the early 1990s free throws outnumbered 3s four to one. The trend is driven in no small part by analytics showing 3s are, all in all, incredibly beneficial to offenses.
Yet Scott has cited analytics during media sessions on at least two occasions. In one instance, he referenced how the Lakers were tracking which players were running harder than others during games, a feature that SportVU provides. In another, he said advanced statistics informed them that Kobe Bryant's performances suffered when he played beyond 32 minutes. Scott said the information came from their "eye in the sky," a nod to the cameras.
Madsen said Scott is far more into statistics than outsiders realize, noting that Scott instructed him to attend and take copious notes at an analytics conference held in San Francisco last fall.
As for Madsen, the former Lakers forward said he has been interested in analytics since studying economics at Stanford, where he also acquired his MBA. "I did take statistics and then I took data indecisions in business school," Madsen said. "I have a bird's-eye view of the world of analytics. By no means am I an expert. By no means am I running specific regressions on a daily basis."
He added, "I think from a basketball coaching side of things, I don't think analytics is going to reinvent the wheel. That being said, if you can look at all the aspects of analytics, and if you can find a way offensively to score one more point and defensively if you can shave another point off what a team can do? If one point results in one win, then it's worth it."
Scott credited Madsen, calling him a valuable conduit between the basketball staff and the analytics staff. "Again, I'm old school," Scott added. "A computer like that, I'm looking at it like, 'I don't have a clue what Mark is doing right now.' I'm computer illiterate when it comes to that stuff. I think a lot of it has a ton of good use. I'm just not the one that's so sold on it because I don't do it that much, because I am so old school."
The GM
Kupchak is considered somewhat old school as well, even to his boss. In that 2012 ESPN.com interview, Jim Buss said Kupchak "relies on his intuitiveness of watching players" and "he's not a numbers guy." When asked about Buss's remark, Kupchak said he first wasn't aware of it.
"I think things have changed and if I take it with a grain of salt, the thing that you're saying about Jim, I didn't rely on traditional data analysis as much as some other people do," Kupchak said.
"But SportVU and the data that is available now has really changed [things]. In the last two or three years, because of what's available now, I guess you could say my passion for analytics has grown."
Kupchak also said that their investment in analytics came after years of tracking the progress of that technology, even as other teams were already diving in headfirst.
"We just felt we were in the loop and on top of what's out there and what's available," Kupchak said. "We knew SportVU was not complete. It would've been a waste of resources. Now, it's not, but it would have been. And there's a right time and a wrong time to jump in."
He stressed that, as of the moment, he believes the human element is still the biggest factor. "That's what we think -- the emotions, the character," Kupchak said.
Could analytics alone one day lead to a championship?
"Well, I'm not going to close the door to that," Kupchak said. "Let's see what the next year or two or three bring. Right now, it's a factor that we use to make decisions. But the pie chart [of how much we use it to make decisions] is getting larger."
The big question
If the Lakers have indeed truly embraced analytics, those around the league still have one big question: Where's the evidence?
"It's hard for me to believe that they're even close to fixing things right now, not given the results but given the way they're making decisions," said Jeff Ma, a predictive analytics expert for ESPN who has worked as an analytics consultant for several NBA teams.
In fact, just this month, ESPN The Magazine and ESPN.com have put the Lakers in the bottom 10 across professional sports in terms of their commitment to analytics. In the NBA, only the Knicks and Nets have shown more hostility to incorporating advanced stats into their operation.
"There is no evidence that I'm aware of that suggests progressive thought, finding innovative ways to improve or progress the team -- either in team construction/composition, or on-court play," said one NBA analytics official. "If anything, their on-court play is actively counter to more widely accepted analytic insights: in the absence of specific personnel pointing you in another direction, increase pace, attack the paint in transition, shoot more 3s, protect the paint and deny the corners on defense."
The Lakers rank third in percentage of shots taken from midrange, widely regarded as the least efficient area on the court, and in the bottom five in corner 3-pointers, among the most efficient. The Lakers have long had an offense -- especially in crunch time -- dominated by Bryant taking contested shots, even as analytics show open shooters to be the more efficient option, regardless of who is taking them. Similarly, in his heyday, Andrew Bynum was among the most efficient scorers in the league, but often lagged teammates in number of attempts.
A hallmark of outmoded hoops assessment is to overvalue players with big points per game totals. Chief among those who fell in value in the era of efficiency, where misses and lost opportunities are factored in, are big scorers. Allen Iverson's reputation has taken a beating from quants who worry over his many misses. The same goes for Carmelo Anthony -- scouts love his footwork; quants note he is a so-so defender who's addicted to covered midrange shots, and skilled as he is, those twin realities mar his overall contributions.
Similarly, Bryant is among those titans of points per game who comes to earth through the lens of efficiency. In the highlights, he's hitting incredibly hard shots. On SportVu, he's making the Lakers easy to defend by predictably ignoring open teammates while taking tough, covered shots that actually don't go in at a great rate.
In other words, as analytics have dawned around the NBA, a player whose game has been an advertisement for a pre-analytic approach to the game has starred for the Lakers.
The response
Despite Madsen's enthusiasm for the numbers, one analytics official familiar with the Lakers' operation said that Madsen's work is "not well-received" by Scott and the other coaches, and that any advanced stats the Lakers are using are "really, really basic."
For instance, although Scott cited how analytics revealed that Bryant's performance suffered after reaching a certain minute mark, he noted it only after Bryant was so fatigued that he had started to miss games. It wasn't long until Bryant was lost for the season to a shoulder injury.
Meanwhile, the reigning champion San Antonio Spurs, who rank seventh overall in ESPN's analytics rankings, have used rest to great effect, even going so far as to sit forward Tim Duncan for games dating back to 1999, Spurs head coach Gregg Popovich recently said.
"Don't get me wrong -- I don't know that the Lakers need to be at the forefront of analytics usage," one analytics official said. "They have some underlying competitive advantages in terms of finances and ability to attract free agents -- in theory, though not in recent practice. They've demonstrated in the past dozen years that they can eschew analytics and be very successful."
A multibillion-dollar television deal, the L.A. market and their 16-title legacy certainly help. This is the same team that turned Bynum and some spare change into Dwight Howard less than three years ago.
"On the other hand," continued the official, "I'm not sure it's reasonable to expect that to continue indefinitely as other teams find and exploit new competitive edges. In short, we know that many teams are getting smarter and pushing the envelope. And none of the available information suggests that the Lakers are in that group, or even making up any ground on the leaders."
Howard, for instance, left the Lakers and their richer contract offer to sign with the Rockets the year after arriving in L.A.
"If you have the resources and you have all the opportunities to get a much better return on investment by doing things correctly, why wouldn't you?" asked Ma. "This is not like a religious [question] -- 'Am I going to be analytically-driven or not?' This is, 'Am I going to try to make good decisions or am I not?'"
One analytics official was highly critical of the Lakers' recent spate of transactions, chief among them the two-year, $48.5 million extension handed to Bryant in late 2013.
"Extending Kobe was an unmitigated disaster," the official said, "at least where winning basketball games is a concern."
Said Ma: "The whole 'Moneyball' concept was just because the A's were cap-strained and they had to come up with better ways to do things. But if you look at a team like the Red Sox that is not cash-constrained, they've had a lot of success incorporating a lot of analytical framework into their decision-making.
"If you think about the responsibility that you have to your fans or the responsibility that you have to your organization, you should always be trying to make the best decisions possible."
The Lakers didn't appear to have clear answers when asked if they could point to specific ways in which analytics have affected team decision-making, eventually offering the following Madsen quote but declining to comment further:
"Analytics has to be simple. If Byron tells the team that another team's center has attempted a single shot outside the paint in two years, this helps us plan strategy. If we notice that an opposing wing scored 150 points on catch and shoot and 25 points on his dribble pull-up, then that can be brought to Byron to consider. Or if a post guy has only scored six points going right shoulder in two years on terrible percentage and 40 going the other way, which way do you want to force him? Some people say analytics is garbage. I say that's wrong. Even if you can prevent two points in a game and you win, and that game gets you into the eighth spot, then it's worth it. Players say, 'We all know each other's tendencies.' But do we really? Do you know the rookies' tendencies perfectly? Do you know the Eastern Conference players? Do you know the Euro guys who just came over? And the preparation part is one branch of the whole analytics movement. So is there any value? I think there is."
The road ahead
Several analytics officials -- both current and former -- pointed out that analytics is too often considered to be a tool that might, in the end, help a team win an extra game or two. In truth, it's much more than that.
"Using it to identify player value better and player development to make sure our guys are actually progressing -- to make sure our core players going forward are actually getting better -- that's a huge valuable thing that analytics can do," Alamar said.
"I don't know if the Lakers are trying to do that or what they're trying to do to understand the value of draft picks, cap space and tying that all together. Those are all things analytics does that has nothing to do with getting an extra win this season."
Kupchak said he was open to change, calling analytics a "whole new world."
"You know, whoever hired analytics people for the first time, they started something and they do look at the world differently, and maybe they always did, or maybe because of all the information, they do," he said.
"But they do look at things differently. I go back and sit with our guys. We'll start talking and they'll ask me questions and I'll say, 'Man, I never thought of that.' So that is refreshing, and whoever hired the first analytics person 10 years ago or whatever it is, I've got that person to be thankful for."
Yet analytics officials and others around the league say they have serious doubts that the Lakers are believers in any sense.
ESPN's rankings characterized the Lakers as "nonbelievers." "I wouldn't deem it true," Scott told reporters Wednesday night in Salt Lake City before the Lakers faced the Jazz. "I think we've got a few guys who truly believe in it -- I'm not one of them, but I listen to it and all that stuff."
Scott said he receives analytics reports on a weekly basis.
"It sits on my desk and I look at it for a little while," he said. "Then I take it to Mark Madsen and we'll talk about it for a minute, and then I say 'OK,' and I take it from there."
Scott was then asked if those meetings have resulted in any major changes this season.
"No."
