SI’s Compendium of Sloan Sports Analytics Conference Quotes With Commentary


In case you don’t feel like spending hours on ESPN’s TrueHoop blog, in which new hoops ideas are analyzed and discussed in a kind of NBA speed-dating scenario which mirrors the organized chaos of the conference itself, here is the opening to the 50 notes and quotes Sports Illustrated‘s Matt Dollinger has gathered for SI’s Point Forward blog, with a few comments of my own to the first 8.

BOSTON — If you’re a mere simpleton like me, the MIT Sloan Sports Analytics Conference can be a bit of an overwhelming knowledge dump. It’s like CrossFit for your brain. A two-day binge of information.

It’s also an incredibly enjoyable experience and one I’m fortunate enough to have experienced for the second consecutive year. Where else can you shift seamlessly from the latest NBA analytics to old war stories from Phil Jackson to an ex-GM admitting to tanking to Adam Silver being interrogated by Malcolm Gladwell?

Since the geniuses at Sloan have yet to introduce any analytics on how to write the perfect blog post (I’m looking at you, 2015!), here are 50 basketball-related highlights from the 2014 conference, comprised of notes, quotes and anecdotes gleaned from the entertaining — and educational — weekend.

1. The most newsworthy revelation to come out of the 2014 Sloan conference had nothing to do with analytics.Instead, it was a bit of candor. Say what you want about former Raptors general manager Bryan Colangelo (yes, he traded for Rudy Gay and signed Landry Fields. OK, better yet, don’t say what you want), but the two-time NBA Executive of the Year was willing to go on the record and admit something that no other modern GM in any sport has yet to confess.

“I tried to tank a couple of years ago,” Colangelo said during Friday’s basketball analytics panel. “And I didn’t come out and say, ‘Coach you have to lose games.’ I want him to establish a winning tradition and culture, but I wanted him to do it in the framework of playing and developing young players.”

Colangelo said the team’s win in the regular-season finale cost them “a coin-flip for Damian Lillard,” which could have altered Toronto’s history (and Colangelo’s, who was fired in 2013). Instead, they were out-tanked that night by the Nets, who scored a sheepish 67 points.

And to Colangelo’s credit, he realized that tanking in the NBA has to go.

“There really is some ugly basketball being played,” he said.

[Note the speaker here, Colangelo, is out of the NBA right now. Regarding the tanking debate, Adam Silver is in a tough position. One provision that might be used in the future, regarding GM’s who wish to gut their roster in order to pile up losses: create a minimum number of years experience that an NBA active roster must meet at all times. Philadelphia currently has Thaddeus Young (6 years), Eric Maynor and Byron Mullens (4 years), James Anderson and Elliot Williams (3 years), and the rest are rookies and second-year players. Jason Richardson, who has yet to play a game this season due to injury, has played 12 years). By forcing a team to keep at least a few active veterans, the Sixers would not have been able to gut their roster so completely.] 

2. Feeling liberated by Colangelo’s candor, Stan Van Gundy went for the kill against the 76ers: “If you’re putting that roster on the floor, you’re doing everything you can to lose. It’s embarrassing.” Also in attendance at Sloan, 76ers general manager Sam Hinkie!

More from SVG later.

3. One of the biggest themes echoed throughout the two-day conference was, “We now have the data we’ve always wanted. Now how do we use it?” Whether it’s how to apply information or discerning relevant material from background noise, that answer seems to be keeping stats gurus up at night.

[In other words, not all information is useful, and when too much information suddenly appears, everyone gets excited but few people have any idea what to do with it. Also: the “we” is the advanced metrics-interested people, not necessarily the players or the coaches. As Goldsberry has pointed out at Grantland, the “we” needs to be more inclusive and less exclusive if the data is going to be used effectively.]

4. Should we be taking advanced statistics with a grain of salt? For what it’s worth, Van Gundy takes them with a boulder. “I don’t trust most of it,” he said of analytics. “I read some of the stuff that people write on ESPN.com and stats on pick-and-roll defense and stuff that came off Synergy.com or somewhere else. I don’t know who is recording that information. … A lot of pick-and-rolls are designed to score and then there’s pick-and-rolls you’re running to get into something else. If you’re recording it and treating those two things the same then you don’t know what you’re doing.”

[Also worth noting that SVG is out of the NBA as well, and may be resentful of the analytics trend. There are so many gray areas within advanced stats that have trouble accounting for the non-plays. Van Gundy is referring to a pick-and-roll that is meant to lead into another action, rather than a pick-and-roll that is an attempt at the basket. It does seem like it would throw off the numbers whenever a site like Synergy is referenced. It makes me think of how certain point guards avoid taking shots in the lane against premier interior defenders like Hibbert or Sanders. Watching Rondo play Milwaukee last year, there were at least four plays where Rondo was clearly not going to attempt the same kind of penetration against Sanders as he would have against another defense. These factors are hard to measure, because they don’t show up in the box score. Intimidation, physicality and reputation were keys to the Celtics’ defensive success throughout KG’s time. The same is true for Chicago today. How are those measured in advanced metrics?]

5. Further fanning the flames: “There’s no substitute for watching film over and over and over again,” Van Gundy said. “The only numbers I trust are the ones my people keep.”

6. Upon reading Paul George has run a league-leading 130 miles this season, Van Gundy said, “Of what possible use is this information?” Tough to argue with that.

[Not useless at all. It tells us how much effort George puts into playing defense. It tells us that we should worry about the toll that might take on George’s career if it continues throughout the next few years. The reason it appears useless to Van Gundy is that he has no frame of reference for it.]

7. Brad Stevens has the strangest Dirk Nowitzki obsession. With his coaching staff preparing for their first-ever game against the Mavericks earlier this season, Stevens said he “became amazed by how enthusiastic Dirk was” at 35 years old. Stevens was genuinely blown away by how much Nowitzki talked during the game and how much fun he was having. “We’re in the middle of this long, arduous season and here’s a guy who has done it all over and over and he was having more fun than anyone else on the court.”

[This is interesting because it shows how without high expectations this year, Nowitzki is able to relax and play with more ease this season. One of the advanced-stats lovers easiest targets over the last few years is Monta Ellis, whose play on the lackluster Bucks, when teamed with the less-than-helpful Brandon Jennings at point guard, was often score-first and selfish. Instead, Ellis has thrived this year, when matched with Jose Calderon and Nowitzki. Ellis is one of the main reasons Dirk is having so much fun. This goes back to chemistry and how the pieces fit together. Those hard-to-measure ingredients that teams are constantly searching for. In a recent Dallas News piece, Mark Cuban and Rick Carlisle commented on the impact of Ellis:

“Monta’s made his life easier. We’ve seen the pick and roll where they [defenders] know he’s going to attack the basket and he throws it back to Dirk for an open jumper. It’s been a huge improvement. Monta leads the league in basket attacks, and it shows. If you leave Dirk, he’ll throw it to Dirk, and if you don’t leave Dirk, Monta will attack the rim. And that’s hard to stop.”

Added Carlisle: “They do a lot of things together and they’ve been very good for each other, which is what we hoped.”

8. Why does Tom Thibodeau play his starters so much? It doesn’t have anything to do with algorithms or advanced metrics. “He keeps his best players on the court,” explained Steve Kerr. “That helps you win.” Genius.

[It helps you win, for sure. It also makes the 82-game schedule that much more arduous and increases the likelihood that your team will run out of steam in the playoffs. The opposite of San Antonio’s philosophy. Thibodeau has had the players (Noah, Deng, Gibson, Butler) who play his style of tough, physical defense so well. The Bulls are overachieving despite injuries and trades again this year. Thibodeau deserves credit. However, he also has to learn to trust his bench and become strategic about not over-playing his starters. Not an easy task, when his team is scrapping for wins just to get home-court advantage.]

To read the rest of these notes and quotes, from Matt Dollinger, click here: http://nba.si.com/2014/03/02/mit-sloan-sports-analytics-conference-review-50-things/

***

From TrueHoop, Aaron McGuire and Timothy Varner recorded a partial transcript of the sit-down between new commissioner Adam Silver and Malcolm Gladwell, regarding the idea of tanking and the draft.

Today’s MIT Sloan Sports Analytics Conference featured a one-on-one interview of Adam Silver (AS) conducted by Malcolm Gladwell (MG). In their wide-ranging discussion, Gladwell asked Silver about the problem of tanking in the NBA. What follows is a transcript of that portion of their conversation.

MG: “All right, moving on. First serious question is about the draft. Lots and lots and lots of buzz about the draft, ideas of what might and might not be wrong about it. Let’s start at a high level. Satisfy my curiosity as a relative outsider about how weird it is that in the context of sports, we reward favor. We know the famous quotation by Marx about how communism was “from each according to ability, to each according to his need.” This precisely describes the draft and the luxury tax. The parallels of having a group of billionaire Republicans who get together to behave like Marxists is just … overwhelming!”

AS: “So … it’s a great point, and every cycle of collective bargaining we end up hiring economists to help us bargain as I said, as economists as well. I always get a kick out of hiring the economists. They first come in, we have a general meeting where they take our current CBA, our bylaws, our constitution, our guidelines, how we operate … they go away, they’re academic. They say, ‘We’ll study this and we’ll come back, we’ll present ideas for you.’ It always happens. They go in, they come back, they’re always really excited, they want to play something … and they say ‘You have it all wrong! You’ve created an incentive for teams to be bad!’ And that’s of course what the draft is, and why we created a draft lottery. And I always say, it’s the same issue. It’s the reason why historically, so many commissioners have been seen as untrustworthy for sports leagues. And that is that the case we present is that we aren’t true economic competitors. If we were true economic competitors, our teams would be trying to put each other out of business. In fact, they agreed to give to the worst performing team arguably the best player coming in. They agree to share all kinds of revenue – not just the direct revenue sharing transfers where the Lakers write a check for $50 million to their partners and say, ‘Hey, it’s important for us that you operate in a way that you can afford a competitive payroll.’ We share enormous amounts of collective revenue. So much so that the way our TV deal is set up, the more popular your team is, the greater number of local broadcasts we take away. And that’s a DIRECT cost to the team, especially as local deals go up. I’d say … if you want to call it a form of socialism, as some do, in terms of how sports leagues operate, I would say from an economic standpoint, we’re a single enterprise. We’re trying to create competition among teams. And that’s what makes our system. The lottery in particular, and our draft, we continue to tinker with it. By definition that’s why we have the draft lottery, we’re concerned about the disincentives to win that are built in the system.”

MG: “Do you think at the present time the disincentives towards excellence are too great in our lottery system?”

AS: “You know, I’m not sure. As we become more sophisticated, as conferences like this exist … I mean, even in the short time I’ve been here, I’ve been handed a lot of business cards already from people saying, ‘I have a better idea for how you do it.’ I’ve read in the last few days here, Mike Zarren’s wheel proposal from the Celtics … also an interesting idea. I’m open to taking a fresh look at it. We’ve experimented to changing things. One of the fundamental mistakes we made years ago when we were less sophisticated about analytics, where the worst-performing team would historically have a lottery of type. If you had the worst record, you’d have a 1-for-1 chance of getting the first pick. So what we did, we fixed that with the draft lottery to dramatically reduce the odds that they’d get the first pick. We kept building it over the years. But let’s say now that it’s a 25 percent chance that you get the first pick if you have the worst record. But the analytical people with the team are saying that even a 20 percent chance is better than a random chance, so I’m not really sure we’re at the optimal point. At the same time –- and I know there was some discussion yesterday about so-called tanking –- these are to the extent that teams believe they may have an incentive to perform poorly. I’m not sure analytics bear out that being the optimal way to operate in that situation. In terms of the word tanking, I think it’s used differently by different people. To me, tanking means a team goes out to intentionally lose the game. I think there’s genuine rebuilding in our system. Especially when you have a cap-type system and you have to plan for the future. Just like any business, there’s short-term and long-term results. I think given their desire to win, they make the decision that this player will not optimize our chance to win a championship, because this player will lead to two or three more wins this year. But our goal is to win a championship, and that player just isn’t going to help us meet that long-term goal.”

MG: “Let’s talk about this. You mentioned about the wheel. For those of you who aren’t familiar, this is the notion where you assign draft positions in a fixed order so that each team would have the first pick once every 30 years. Suppose you wave a wand and that’s in place tomorrow in the NBA. Can you tell me about the impact you think that would have on the NBA?”

AS: “So, my point here came up with this proposal. Over the course of 30 years, you move throughout mathematically in terms of your draft pick. This goes to show some issues. When Mike first showed it to me, I thought, ‘Wow, that solves our problems.’ Teams can plan for the future, they have absolutely no incentive to do anything but win the maximum number of games per season, they know where the draft pick is coming from. And I said, ‘Let’s socialize this a little bit, let’s send it out to other GMs and get other team reactions.’ And what came back … maybe it’s an obvious issue, but it’s one that surprised me was when teams said, ‘Hold on a second.’ There’s a belief that certain markets have advantages. That players may choose to be on the coast, be in a larger market as opposed to a smaller market. I’m not entirely sure that’s the case, but that’s the perspective. The concern by some of the teams was that if a player going into college or coming out of high school … Say that he knows the hometown team here, the Celtics, has the No. 1 pick in two years. I’m going to wait those two years to come out, because I can game the system as a player. I can choose to be a Celtic. Maybe another player wants to be in Orlando and he waits till he’s a junior to come out. Because if I’m in Orlando … In return, appropriately so.” [MG interruption]

MG: [laughing] “I’ve heard that objection. And that objection strikes me as really lame. I mean, it proposes that the player can actively predict where they’re going to be taken in the draft, that’s No. 1. And two, that the teams themselves wouldn’t react to this very thing simply by trading their draft picks. If the Celtics really want to take Joe Blow, and he’s No. 1, they can simply swap their picks.”

AS: “That’s what we heard about from the teams. As I said, I liked the idea initially. We’ll still study it. Basketball is so unique. It’s both a team and an individual sport. In terms of you and I, or a group of basketball experts here, and if we put the cards on the table. ‘There are 450 players in the league. We want NFL-style parity.’ I think the issue here is that if you don’t have one of the 12 or 15 players in the league, maybe it’s a smaller number, you have a very small chance of winning a championship. Because an individual player can be so dominant in this league. And then there are these players that come around once every few years, once in a generation … But I think, I remember reading about LeBron in ninth grade. We saw him coming along. Could we predict how many championships he’d win? I don’t know. But there was no question that he was a game-changer for any team that could get him. So I think then there’s that sense among teams. It’s once every 30 years, guys come across, and suddenly the player says, ‘Don’t bother, I’m not gonna play for you. I’ll wait a year or two and choose my team.’ Look, it’s definitely not the perfect system we have right now. And based on the discussion we had about it, it’s one of those topics …”

MG: “I’d like to talk about the fact that when you first heard that idea, you were quite taken to it. So there’s some element of dissatisfaction in your own mind with the current system?”

AS: “Yeah. In part, and again … Because I’m not sure the analytics support this notion that you can game the system and lose this many games to have this specific record, it requires a lot of precision. In terms of how good someone’s going to be, where he’ll come out in the lottery. But I’d say in terms of the business of the NBA, the fact that there’s so much chatter about the notion of tanking … Of course that concerns me from a business standpoint.”

Advertisements
Tagged , , , , , , , , , , , , , , , ,

3 thoughts on “SI’s Compendium of Sloan Sports Analytics Conference Quotes With Commentary

  1. Ben says:

    “We now have the data we’ve always wanted. Now how do we use it?” Whether it’s how to apply information or discerning relevant material from background noise, that answer seems to be keeping stats gurus up at night.

    In other words, not all information is useful, and when too much information suddenly appears, everyone gets excited but few people have any idea what to do with it.

    I’m not sure that I agree that this is what he’s saying here. Saying “few people have any idea what to do with it” feels to me like people are clueless, rather than saying it takes time to sort through an enormous amount of data to figure out how things are connected and what is most important.

  2. Ben says:

    Absolutely agree with the comment on the miles George has run.

  3. jonahph says:

    I didn’t meant to imply that people are clueless, but that the novelty of the new information sometimes creates a buzz, whereas the conclusions about what is and what isn’t important (determining the signal vs. the noise) are often subjective, and, like you said, take time to sort through.

    Even then, it might remain subjective (one team values players that rarely over-dribble while another team values players that run more miles). Interesting to consider how protective teams in MLB and NBA have become in terms of how they value the new information. They believe they have the secret recipe, which may or may not be true, but remains a mystery to everyone on the outside.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: