The data National Basketball Association teams use to analyze potential trades never will replace instinct and assessment of players by scouts and coaches, Dallas Mavericks owner Mark Cuban and other panelists at a sports conference said.
Jeremy Lin was never mentioned.
Team strategy, salary cap and roster space are as important in deciding whether to trade a player as data, Cuban, owner of the NBA-champion Mavericks; Mike Zarren, in charge of analytics as assistant general manager of the Boston Celtics; and Kevin Pritchard, director of player personnel for the Indiana Pacers, told the Massachusetts Institute of Technology Sloan Sports Analytics Conference in Boston on March 2-3.
“You need people to watch basketball players for a long time and know what an 18-year-old kid is going to be able to do when he’s 25, because the numerical data is really noisy and the visual data isn’t,” Zarren said. “You still need all the other stuff.”
The inaugural one-day meeting on MIT’s campus in 2007 had 175 attendees. This year the conference had 2,200 visitors, with representatives from 27 NBA franchises among the 73 professional teams present at the Hynes Convention Center. The growth of the event, co-founded by Sloan School of Management graduate and Houston Rockets General Manager Daryl Morey, may indicate the increasing influence of data in sports.
Attendees this year included Scott O’Neil, president of Madison Square Garden Sports; Los Angeles Lakers executive Jeanie Buss; and Denver Nuggets General Manager Masai Ujiri. Other panelists included NBA Deputy Commissioner Adam Silver, National Hockey League Commissioner Gary Bettman, Major League Baseball Executive Vice President Rob Manfred and Brian Rolapp, chief operating officer of National Football League Media.
Cuban, Zarren and Pritchard all referred to “Moneyball,” the book by Michael Lewis about the use of data analytics at baseball’s Oakland Athletics. Data analysis is commonplace at MLB clubs and became a topic of popular conversation after Lewis’s book was made into an Oscar-nominated movie starring Brad Pitt.
Analytics five years ago accounted for perhaps 5 percent of his trade decisions, Pritchard said. While the influence of data will rise over the next decade, it never will dominate, he said.
“I don’t think it ever takes over the eyes, I don’t think it ever takes over the gut,” Pritchard, 44, said.
Since no one has managed to devise an analytical tool that predicts how a player will perform once he moves to a new team, the use of data is limited to historical appraisal, Zarren said.
“You may do a lot of analysis on how a player affected Team X,” he said. “That player may come to your team and act totally differently because he’s put in a different position. That’s the coaching piece and that’s a very hard piece to analyze.”
As a result, the coach’s plan for the team and any potential recruit is of paramount importance in trade decisions, Zarren said.
“You just can’t do an important trade without the understanding of what your coach will do after you do the trade,” Zarren said. “There’s too much invested in these guys not only financially but roster-spot wise.”
No one on the panel mentioned Lin, the 23-year-old Harvard University graduate who has led the New York Knicks back into the playoff picture since becoming their starting point guard on Feb. 6. Draft bloggers Arturo Galletti and Ed Weiland said their analysis showed Lin was among the best prospects.
When a player scores in a game can be more beneficial in decision-making than looking at the total points he scores, Cuban and Zarren said.
If a player scores 20 points in the first quarter and two points the rest of the game and the team loses, his importance is less than if he doesn’t score through the first three quarters before contributing 22 points in a two-point win, they said.
“If you’re lighting them up when it’s 0-0 or you’re up by 30 points, on an averages basis you still look like a stud,” said Cuban, who wore a light blue T-shirt with “Talk nerdy to me” printed on the front. “A lot of people overweight that.”
Some stats are just worthless, Zarren said.
“It’s amazing the number of times you get an e-mail from someone with another player-ranking system,” the Harvard Law graduate said. “I never sit around thinking, ‘Who’s better, Kevin Garnett or Tim Duncan?’ It’s not a conversation we ever have in our front office. It’s all about what will affect our team.”
It’s up to general managers to educate owners and coaches about the usefulness of data and encourage them to drop their skepticism, Pritchard said. Ten of 30 NBA teams have no one working on analytics, he added.
The teams that do invest in it have no intention of divulging what they know, the panelists said.
“There are certain things we won’t talk about but for the most part, data is data,” Cuban, 53, said. “It’s how you use the data and are you in a position to take advantage of the data.”
To contact the reporter on this story: Dex McLuskey in Dallas at firstname.lastname@example.org; Scott Soshnick in Boston email@example.com
To contact the editor responsible for this story: Michael Sillup at firstname.lastname@example.org.