Stan van Gundy was the hero of the MIT Sloan Sports Conference.
The Sloan Sports Conference was a sports-nerd frenzy. It is perhaps the only place and time during the year where you can tap just about anyone on the shoulder and kickstart a conversation about PER vs. WAR. Just about every single person there was extremely knowledgeable about one sport or another. The conference is commonly believed to be all about the NBA, maybe due to Daryl Morey’s complete control over the proceedings- and the fact that his wide, beaming smile is stamped front and center on all of the promotional material. But there was quite a lot of MLB, NFL and NHL talk to be heard, and hockey and baseball research papers took home awards at the closing ceremonies. Still, though, basketball was the main event, and the preeminent speakers were all from the NBA. R.C. Buford, Kevin Pritchard and Adam Silver all impressed. But the real star of the sports conference was a certain mustachioed man, the former coach of the Orlando Magic, the one and only Stan Van Gundy.
ESPN should absolutely give a show to Stan Van Gundy. And I’m not talking about a radio show between the hours of 2 and 4 PM. Stan deserves primetime TV attention on the primary ESPN channel. He simply showers the world with wisdom. Van Gundy spoke tenderly on the subject of our perception of young players, saying about a young Lamar Odom: “When he came into the league, people said he was a bad guy because he smoked marijuana and skipped class. If not going to class and smoking pot made you a bad person, half of you (the audience) wouldn’t be here.” Later he very effectively explained why coaches are often suspicious of analytics, focusing on the fact that most players respond poorly (or not at all) to data charts and offensive efficiency ratings. That moment, in fact, was probably the most salient of the conference: a former coach describing the limitations of current advanced statistics as applicable solutions to a team’s problems. Because the most innovative presentation of the weekend focused not just on the numbers, but on how the numbers could be broken down into a palatable format for players and coaches. That was Kirk Goldsberry’s The Dwight Effect: A New Ensemble of Interior Defense Analytics for the NBA.
In Goldsberry’s research paper, he examined individual interior defenders in the NBA, discovering who was the best, who was the worst, and who fell in between. His findings were not particularly shocking (Dwight’s the best, David Lee is the worst), with a caveat for Cavs fans: Anderson Varejao is one of the worst interior defenders in the NBA, allowing over 50% shooting on plays around the rim which he defends. What was truly amazing, though, was the way in which he presented the information to the audience. Check this out. Goldsberry’s shooting charts, which you may be familiar with if you frequent Grantland, are exactly the type of visuals that could really help a coach impart to his players what they are doing wrong. (I would love to see Byron Scott sit down with Kyrie Irving and slap a printout in his face that shows anyone with a brain that he is an awful defender.) If I ran an NBA team, I would hire Van Gundy and Goldsberry, and then lock them in a room with a laptop, food, water and a hamster wheel for three months. Dan Gilbert could afford that, right? I hope so. (Also watch this for more info. And this for more laughs.)
P.S.
Zach Lowe was very nice about telling a never-ending line of Columbia and Harvard students that they (a) would not be offered a job by Grantland anytime soon, and (b) should not pursue journalism because it is a soul-crushing profession. He also chatted with me about the Cavs for a few minutes, and says that Cleveland fans should be very pleased with and excited by Tristan Thompson and Dion Waiters’ recent play. Lowe also mentioned that he thinks Kyrie Irving’s defense is a serious concern going forward- not an unfixable problem, but one that clouds his bright future considerably.
Tags: MIT, Sloan Sports Conference, Stan Van Gundy
kyrie defense was the best i ever saw him play tonight
Yeah, Kyrie played hard last night, which was nice to see. But he still got lost on screens several teams, leading to open jumpers. I could be wrong, but I believe that both the late Hayward and Burks threes were his fault.
Van Gundy is intriguing and there is definitely a market for him in the sports media. If they keep blackballing him on ESPN; perhaps the new FOX SPORTS 1 could take him on. I get so tired of the repetitive Lakers and Heat stories to the point of nausea on ESPN. And I’m not alone on that complaint!! Van Gundy will talk about all the NBA teams and that is refreshing and needed!! There is so much more excitement to report in the NBA that is being missed daily by ESPN. Also wouldn’t mind too terribly if he replaced Kenny Smith on NBATV. That dude is so annoying and arrogant!!!
Glad Zach Lowe has good things to say about the Cavs. He’s one of the smarter NBA writers.
How is no one talking about how DeMarcus Cousins is a better defender than Dwight Howard on one of the worst teams in the NBA.
Rank
Defender
Shots Faced
% Close Range
% Mid-range
% 3-point range
Close FG%
Mid FG%
3-point FG%
1
Dwight Howard
409
48.2
32
19.1
45.7
38.2
43.6
2
DeMarcus Cousins
279
48.4
29.7
20.1
48.2
38.6
44.6
Cousins allows .3% more baskets near the rim, but THREE percent less from mid range.
He is the best defensive big in the NBA, why in the world would we not give up a defensive Sieve like Varejao and TZ+draft picks for him?
@Witmi
I’ll guess we don’t get him because
A) With the move coming up, the King/Sonics want to hold onto him.
B) He’s a petulant coach killer on the same plane as Iverson and Sprewell
C) Do you really think he’d grow up in Cleveland? It’s certainly debatable.
It’s a very high risk/very high reward type of move. Either he figures it out, or torpedoes Kyrie’s formative years with all the bullcrap he’s pulled so far in his career. Personally, I’m not willing to risk it. Could be just me, but I don’t think I’m alone on this one.
Sorry, but rating a big based on the shooting percentages of players within 5 feet of him is total shit. This basically means that if a guard blows by Dion or Kyrie for the layup, or if Zeller gets abused in the post, Andy is held responsible. If you look at defensive efficiency for actual plays defended, as tracked by Synergy, he ranks 15th in the league. It baffles me that a paper that bad, which is so obviously going to overrate average defenders on good defensive teams, and underrate good defenders on bad defensive teams, managed to rear its head at such a respected conference.
@Dani: The Hayward three near the end of the game was completely CJ’s fault. I know because I nearly threw my remote at the TV screaming at him that he’s the one guy on that team that you can’t turn your back on in that situation.
CJ was following Hayward around the perimeter, then turned his back on Hayward to ball-watch Millsap in the post. Hayward noticed that and floated from the top of the key over to the wing, and by the time CJ saw the pass heading out it was WAY too late to close out. If CJ just trusts that the post D is good and that the help will come from the man in the middle of the D (Kyrie) if needed, then we have rotators in place no matter where the pass goes. If he leaves his man then there is nobody in position to rotate over, while that’s not the case for anyone else (…and nobody else had to worry about guarding the team’s best 3-point shooter).
I get the obsession with advanced stats, but I think too much stock is often being placed in them. This isn’t baseball, as there are far too many moving parts and immeasurables in basketball that limit their uses. A player can have a major impact during a game and not even register in the box score by passing, boxing out and defending. The David Lee thing: did you know Golden State is one of the best teams in the NBA defending the 3 pointer? They lack size and often commit defenders toward the perimeter, so the most open looks come near the basket against the Warriors, healthy Andrew Bogut or not. For all we know, Mr. Lee could be a great interior defender with unfortunate stats — but we know this isn’t the case because we’ve watched him “play” help defense or stand awkwardly as an opposing guard lays it up in his face. He’s not a bad defender because of his opponent’s shooting chart; he’s a bad defender because of his ability and decision making, and the chart merely shows the statistical effects of this… I feel like people get this confused and use stats as the foundation of an idea instead of evidence. Numbers don’t lie, they just don’t tell the whole truth, especially in a team sport.
The players and coaches have a right to not be receptive of many advanced stats because numbers oversimplify and discredit the work they put in. A player learns how to release the ball a little higher to improve his jumper, or how to dip his hips lower when defending in the post — it means nothing to him to wave an offensive efficiency chart in front of his face. Sports are a world of kinesthetics and spatial awareness first and foremost — mathematics have a place in that world, just not at the forefront, as many sports writers might have you believe (NOT pointing fingers). I just think the obsession with stats and the eagerness to use them to understand what we see has exaggerated their usefulness. With all due respect to his body of work, here’s to hoping the Cavs don’t go out and hire themselves a John Hollinger.
Latest podcast with Brian Windhorst. Not really much on the Cavs in this one.
http://www.stationcaster.com/player_skinned.php?s=70&c=476&f=1106311
Nathan – I thought the same thing when I watched that video.
Pete – Rich tweeted a more succint version of that during the game last night. The Cavs need to have better awareness/trust on the defensive end. TT can often hold his own if he keeps bigs from getting too low.
On a positive note – Miles had a nice game overall imo. He hit some big shots in that 4th quarter.
I have to say that there may be some truth to what Nathan says. One thing I’m noticing is that the Synergy data does not back up Goldsberry’s paper at ALL. If you look at the Synergy data, it ranks AV as the 15th best defender in the NBA on a per possession basis, noting that he gives up .71 points per possession, including a scintillating .58 PPP on isolations and a .64 PPP on the pick and roll man. Postups? A solid .83, with a ridiculous 14.6% turnover rate. David Lee isn’t nearly as bad as the paper contends either, giving up an average of .83 points per possession.
Why don’t these numbers jive? Is Nathan right that bigs are getting a lot buckets against them that aren’t their fault? Schemes and the guards’ ability to direct penetrators to the right angle for help, is certainly part of the equation. Teams that are collectively bad are going to make Goldsberry’s percentages much higher. Is it that Goldsberry’s method is better at measuring help defense while Synergy’s is better at measuring individual defense?
The paper is interesting but I think there needs to be a lot of exploration of methodology and assumptions. Many of Goldsberry’s assumptions don’t seem to be born out in 82games ratings or regularized adjusted plus minus when it comes to players like Varajao, who Goldsberry panned, but who had a 2.4 RAPM on defense which is good.
Much of this methodology is interesting because it shows what teams put their bigs on an island, and what bigs help teams change the “shape” of the defense, but it hides a lot of things that players do well, like AV and David Lee being good pick and roll defenders.
A lot of Goldsberry’s data is just too abstracted to tell someone what is really going on with each player. According to Goldsberry’s thesis, good help defenders either decrease the percentage of an opponents shot around the basket (like Larry Sanders), or intimidate another team from even shooting around the basket (like Dwight Howard), but he doesn’t correlate that to what the offenses’ chances to score on a given possession are… Are players getting themselves so far out of position that they’re giving up offensive rebounds? Goldsberry admits a lot of his limitations at the end of the paper though.
Cousins is a good defender, to be sure, but he is a block hunter. Note his TERRIBLE stats against spot up shooters. Also note his fairly low turnover rates.
David Thorpe ranked Dion as the top 21 year-old rookie:
1. Dion Waiters, Cavaliers
Waiters has All-Star talent. He isn’t quite the athlete that Dwyane Wade is, or the shooter J.R. Smith is, but he has a lot of both of their talents with some Eric Gordon mixed in. If Waiters reaches his potential, he is someone who will be dynamic off the dribble, attacking relentlessly and earning six-plus free throws a game, at minimum. His explosion step is impressive and his body can play through contact on his way to the rim, like Wade.
Waiters just has to learn when and where to shoot, focusing on making shots and not just taking them. His slashing ability and good feel for finding teammates means he can be an excellent playmaker, and partnering with Kyrie Irving is giving him a great glimpse of how a player can balance scoring and playmaking for others. In Cleveland, he may ultimately be forced into a Jason Terry/Manu Ginobili-type of role, which would suit his talents perfectly.
1. Dion Waiters, Cavaliers
Waiters has All-Star talent. He isn’t quite the athlete that Dwyane Wade is, or the shooter J.R. Smith is, but he has a lot of both of their talents with some Eric Gordon mixed in. If Waiters reaches his potential, he is someone who will be dynamic off the dribble, attacking relentlessly and earning six-plus free throws a game, at minimum. His explosion step is impressive and his body can play through contact on his way to the rim, like Wade.
Waiters just has to learn when and where to shoot, focusing on making shots and not just taking them. His slashing ability and good feel for finding teammates means he can be an excellent playmaker, and partnering with Kyrie Irving is giving him a great glimpse of how a player can balance scoring and playmaking for others. In Cleveland, he may ultimately be forced into a Jason Terry/Manu Ginobili-type of role, which would suit his talents perfectly.
Thoughts on preemptively shutting down Kyrie like Scott mentioned last night?
Best to air on the side of caution if he really has something wrong with it. A bit unfortunate because I really want to see him and Dion get better playing together.
Totally agree. It’s always a scrub in practice who hurts someone. I vaguely remember in 1994 or so when Danny Manning tore his ACL on Joe Klein’s foot during practice. Banging knee’s with Kyrie might have been Omri’s only memorable story that we’ll remember in 10 years.
If the hyperextended knee could lead to something major like a meniscus or ACL tear than I’m all for them shutting him down. At this point Kyrie seems to want to play through the pain.