By the numbers: SVG believes in stats and looks to beef up Pistons analytics staff
Rocky Widner (NBAE/Getty)
ESPN ranks the Pistons among the NBA's more evolved teams when it comes to embracing the value of analytics in roster building and implementation. They were in a group that included Oklahoma City, Golden State, Portland and Cleveland labeled "believers" when the media giant last week ranked professional sports franchises across all four major sports for their utilization of statistical analysis.
They might as well have been ranked with a bullet, because Stan Van Gundy's goal is to beef up the analytics department even more.
The Pistons were represented by four front-office members at the weekend Sloan Sports Analytics Conference held annually at MIT: assistant general manager Ken Catanella, who heads up the Pistons analytics department; Pat Garrity, director of strategic planning; Andrew Loomis, executive director of basketball operations; and Jorge Costa, software engineer who works hand in hand with Catanella.
While Van Gundy was out of coaching, he was a featured guest at the conference and came away with an appreciation for the value of attending and, especially, for the bright minds who populate it.
"You make some really good contacts there," Van Gundy said. "They will run into some very, very bright people for sure – and some people that we may want to either use their businesses or bring on board in the future. You get some great ideas. They always are presenting new research, which not only is the research itself valuable, but it gives you ideas on other things you might want to do. I found it an eye-opening thing of how much can be done and how many smart people are out there.
"That's sort of a must for people to be there, for us to be there, because we're trying to build that area up even more and those guys will come up with good ideas. That's why we had four of them out there."
A few Van Gundy comments from his Sloan appearance were taken out of context and interpreted to mean he wasn't the biggest fan of analytics. What he really said was that the value of the numbers is only as good as the data being inputted and the way it's interpreted. Unless everybody can agree how to define an isolation play or a side pick and roll, then taking seriously numbers that purport to speak to the efficiency of such plays can lead to poor decision making.
Figuring out what numbers are meaningful and developing the most efficient means of arriving at those numbers is Van Gundy's aim for the future with the Pistons' use of analytics. It's one of the primary items on the plate of Pat Garrity, who played for Van Gundy at the end of his 10-year NBA career and reached out to him when Tom Gores hired Van Gundy last May. Garrity, like Catanella, earned his master's degree in business at Duke and worked at a major Connecticut hedge fund before deciding he wanted to pursue a career in basketball – starting at Van Gundy's side in large measure due to Garrity's respect for his intelligence and straightforward approach.
Van Gundy has always had an appreciation for statistics. Long before "analytics" became a component of professional sports management, Van Gundy intuitively understood there was value in the story told by numbers. He doesn't put blind faith in analytics, but neither does he think the eye test and hunches trump anything the statistics might argue against.
"I look at the numbers every day. Some of them are numbers we've looked at for 50 years, when my dad was a high school coach. Some of those numbers are still the same," he said. "It's hard to say I looked at that number and it made me do this because it's all part of a bigger picture, but it is something that I use.
"There are a lot of things that can't be quantified with numbers and you have to know the player you're getting involved with. You have to know the fit with your team. Things like that you can't quantify, but that doesn't mean that the numbers aren't useful. They're part of the entire picture. If you're relying only on numbers, you're making a mistake. And if you're not using them, I think you're making a mistake. You've got to put everything together. It's not an exact science, but the analytics make it a little bit easier and a little bit better."