2015 SSAC

Adam Silver says data may prove Gregg Popovich right in terms of resting multiple players


BOSTON — When David Stern held the office of NBA Commissioner, he fined the Spurs $250,000 after head coach Gregg Popovich sent four starters home the day of a nationally televised contest against LeBron James and the Miami Heat.

Adam Silver, who took over for Stern in February of 2014, isn’t likely to follow in his predecessor’s footsteps.

At the time, Stern blamed the sudden nature of the Spurs’ decision (along with the lack of any sort of advanced communication) for coming down so hard.

But Popovich has long felt that resting his players during certain stretches of the league’s long, 82-game grind would be beneficial, and Silver, speaking at the MIT Sloan Sports Analytics Conference on Friday, was willing to concede that as the league looks to make injury data available for all teams to analyze, it may ultimately prove what Popovich has known all along.

“There’s been a lot of talk recently about back-to-backs in our league, or four games out of five nights,” Silver said. “There’s a sense in terms of how tired guys are, but we should be at a point where using better analytics in terms of creating the schedule — (not only) how big a difference does it make if maybe there’s a back-to-back, but when they’re playing back-to-back and they cross two time zones, does that have a different impact, for example?”

“I think our teams, there maybe was a time where that was viewed as a competitive advantage, to the extent that ‘we have better trainers, better doctors,’ Silver said. “You’d like to believe that as a league, that ultimately, that’s not how we want to compete, in terms of the health and welfare of our players — that we all have an interest in ensuring the best players are on the floor. And so there are additional areas where I’m going to be having discussions with our general managers and our owners, and where everybody’s going to be willing to take a long-term view and say it’s in the collective interest of the league to focus on the health and well-being of the players, and taking the best wisdom from all the teams and distributing that — which is exactly what we do now on the business side.”

The moderator of the panel then quipped that once all teams start embracing that sharing of data and begin to analyze it, Popovich won’t be able to sit four starters for a particular game anymore, because the information won’t support the decision.

“Or,” Silver said, “Maybe he’s right.”

MORE: Shane Battier: ‘A 60-game season would be perfect’

Shane Battier on Carmelo Anthony: ‘Last two years I guarded him, he did exactly what the scouting report said he would’


BOSTON — Shane Battier is the poster-boy where advanced analytics are concerned, and had a reputation during his playing days of being one of the game’s more knowledgable defenders, taking in as much statistical information as possible in order to help him gain a competitive advantage.

While he admitted having difficulty guarding the game’s elite scorers regardless, there was one player that was so predictable in the way he played that Battier felt good in his ability to consistently shut him down: Carmelo Anthony.

“I had some success against him,” Battier said. “The numbers really play out with him, and as he’s gotten older, he has not been about reinvention. The last two years I played him, he did exactly what the scouting report said he was going to do, every single time.

“A lot of players will deviate — Kobe was so tough in his prime, or Durant or LeBron. But Carmelo, I knew what he was doing. He was on the left block, he’s going to dip his right shoulder and go to his left hand every single time — he travels every time, by the way; he travels every single time. But if I made him use his right hand and go over his left shoulder, he didn’t want to do that. And as a result, I was able to drain his efficiency.”

Battier may have had success most of the time against Anthony, at least in terms of forcing him into more difficult attempts that resulted in a lower shooting percentage. But given that Anthony is one of the league’s best scorers, all of the information in the world won’t be enough for someone with Battier’s skill set to stop him every time out — like the time Anthony dropped 50 on Battier’s Heat back in 2013.

Even after that one, however, Battier was unwavering in his praise of analytics, and felt good about his defensive performance in spite of the end result.

“The only time he got me was, he had the most unbelievable game,” Battier said. “This was like, the outlier of all outliers in terms of the numbers and analytics. He scored 50 points, and didn’t have one paint point. So, yeah. I was really feeling good about ‘any long non-paint two is a good shot’ after that game, right? He scored 50 points on me, ZERO paint points.”

“As far as the numbers were concerned, I always enjoyed Carmelo, because he knew what he was,” Battier said. “And he wasn’t apologizing for it; he’s really good at it. It was always fun to intellectually pit the numbers against his game.”

Daryl Morey believes two bad shots are better than one good shot (always), but data wasn’t significant enough for Jeff Van Gundy to change


BOSTON — There was a time when Rockets general manager Daryl Morey believed that getting two bad shots was better than one good shot — always. And maybe, he still does.

But while the statistical data may show that, it’s not always easy to get a head coach to buy in, and make that strategy a part of his team’s routine — especially when the amount it may change a club’s fortunes in terms of wins and losses is proven to be statistically minimal, at best.

Speaking at the MIT Sloan Sports Analytics Conference on Friday, Morey recounted a story about trying to change then-head coach Jeff Van Gundy’s mind where two-for-one opportunities are concerned. These situations can come up at the end of each quarter, when a team has the chance to shoot the ball with maybe 30 or more seconds remaining in the period, in order to ensure they get it back in time (thanks to the other team needing to shoot it before the 24-second clock expires) for one more potentially-critical possession.

“Jeff had never been in favor of two-for-one,” Morey said. “It made no sense (to him) whatsoever. And basically I was like, why are we not doing this? There’s 100,000 trials, and it doesn’t matter who’s on the floor, it doesn’t matter the context — two bad shots are better than one good shot. Always. There’s like, no exception.

“So I was like, why are we not doing this? And Jeff, of course, is very smart. So he said, ‘OK. If I do that two-for-one thing, every time instead of not, how much more are we going to win?’ I was like, oh. That’s a good question. So I went back. And I was like, we’ll win … one more game every two years.

“And he was like, ‘I’m not doing that!'”

Extra possessions are to be valued in basketball, certainly. But it’s long been believed that two-for-one shots indeed end up in two poor attempts — the first one is rushed due to time constraints, and the second one usually comes in isolation, forced up over one or more defenders just before the clock expires.

But while the percentages of success are extremely small (i.e., making both of these unnecessarily difficult, somewhat forced shots), the numbers do show an advantage to adopting this strategy, however small it may be. Still, this is the challenge for those at the forefront of using the data to make sound decisions — convincing those on the court calling the shots that what the numbers say to do is indeed the right choice.

“I think this is where analytics gets a little confused,” Van Gundy said. “It’s not what you know, it’s what you can impart on people like coaches that don’t believe in it, so that they will then get it to their team.

“So I think what’s been undersold in all this is not this term (analytics), it’s trying to analyze and make the best decision for the team. And what he did was, he made me as a coach — if I disagreed with him for something like this two-for-one — you have to sit back, and if the numbers say I should do it, it makes you think. If the numbers are saying it and I’m not doing it, why am I not doing it? And then try to come to some reasonable explanation.

“And often times, and even on that one, if I’m wrong, I should change.”

Shane Battier: ‘I think a 60-game season would be perfect’


BOSTON — During All-Star weekend, one of the things Adam Silver said he wanted to focus on in his second year as commissioner was reducing the wear-and-tear on the players by reducing the number of back-to-backs on the schedule, and attempting to eliminate stretches of four games in five nights almost entirely.

Beginning the season earlier and/or extending it later into June would allow more time to accomplish that, and those options remain on the table. But what hasn’t seriously been discussed is reducing the number of games in place on the regular season schedule.

“To me, 82 is here because somebody is making a lot of money,” Mike D’Antoni said Friday, as part of a panel discussion at the MIT Sloan Sports Analytics Conference.  “Usually that’s the bottom line. They’re making money, it hasn’t been a disaster, and it’s a little more like a marathon, and that’s just the rules. 82 isn’t going anywhere.”

As D’Antoni summed up succinctly, without a large amount of data available to essentially prove that an 82-game schedule significantly puts the league’s players at risk, the financial incentive not to touch that magic number of 82 will remain too strong. And Celtics assistant GM Mike Zarren echoed those remarks.

“It’s not just the number of games, it’s in what time frame,” Zarren said. “So there may be some tweaks that happen soon in the NBA to that. It’s a much more realistic thing than cutting games, because it’s in everyone’s interest to grow the pie, and cutting the number of games cuts ticket sales, which shrinks the pie.”

Those are realistic perspectives, but they’re ones that come from a coach and a member of the front office.

On the player side, Shane Battier came up with a number of games that he believes would be ideal — not only to protect the athletes, but also to make the games that are played much more compelling.

“Personally, I think a 60-game season would be perfect,” Battier said. “Every game matters more. You can’t sleepwalk through a few weeks of the season — it does happen — and then all of a sudden wake up near the All-Star break and turn it on. Fans just want to see the best basketball players in the world at their highest level going head-to-head.

“Every team has a certain number of throwaway games. You just know. You just know you’re not winning tonight. You don’t have it. And then after the game, coach knows it, everybody knows it, coach comes in, says ‘Alright, bring it in guys. We’ll get ’em tomorrow. 1-2-3 team!'”

Money, however, will remain the primary reason this idea faces such great and continual opposition. And while Battier firmly believes less would be more in terms of a reduction of the schedule, he certainly realizes it.

“Obviously, the economics, that’s not my forte so I can’t figure that out,” he said. “But from a competitive standpoint, from a product standpoint, the game would be immensely better with 22 less games.”

Jeff Van Gundy to Daryl Morey when he flinched at using analytics to make a decision: ‘Either you believe in this sh-t or you don’t’


BOSTON — Jeff Van Gundy was the head coach of the Houston Rockets from 2003-07, and along with Daryl Morey as GM and Shane Battier on the roster, those teams have been credited with being the first to use advanced analytics to make decisions more than any other, and pushing the statistical movement into the spotlight.

But despite what the data showed, not all of those decisions were easy.

Speaking at the MIT Sloan Sports Analytics Conference on Friday, Morey recounted a tale of when the numbers showed that Dikembe Mutombo was a better on-court option than Yao Ming was. But when he wasn’t ready to use the data to pull the trigger on making the decision, Van Gundy called him out for questioning what the numbers showed.

“One of the things we showed, and I don’t know if you remember this Jeff, but we basically showed that Dikembe Mutombo was better than Yao Ming,” Morey said. “And Jeff correctly called me out, because I was like, well, it shows he’s better, but, nah, we’re going to play Yao. And he was like, ‘Well you either believe this sh-t or you don’t.'”

Morey still wasn’t ready to play Mutombo over Yao. But when at one point injuries forced his hand, it turned out that the numbers happened to be right.

“What was interesting was, when we had our winning streak — this was after Jeff left — about 12 games in, Yao goes down, and Dikembe comes in,” Morey said. “And we’re better! We were better.”

“I’m not saying [Mutombo] is better,” Morey said. “But it was interesting.”