http://edition.cnn.com/2014/11/04/business/revolutionizing-basketball-with-data/ Excellent article breaks down the work of Kirk Goldsberry and his quest to map every single shot taken in the NBA with CourtVision and SportVU, which use a host of statistical models and technologies like missile guidance cameras to create the most in-depth performance analysis maps. Here's an excerpt relating to Moreyball and the Rockets:
I think Morey was well aware of the Dwight effect before this guy's presentation. He was trying to acquire Dwight for a while after all. Just goes to show how ahead of the curve Morey is with this analytics stuff compared to other GMs.
I honestly think his biggest contribution to the game is actually his manipulation of and finding loops in the salary cap & CBA to find an edge. The analytics are great for on the court usage. Yes, 3 points are better than 2, etc. However the numbers attached to on court performance are based on the human factor. However with the CBA, the salary cap, and NBA trade rules... the numbers or words on paper never lie. There isn't a human element to put a finger on. The rules are rules, and as long as the NBA front office will approve it, Morey will do it to gain an edge. The NBA is full of copycats as we saw with Cuban and the Parsons contract. However Cuban might have bypassed maybe one of the most important aspects of NBA contracts.... tradeability. I think as long as folks continue to focus on Morey's persistence to be cutting edge in analytics it works great to deflect attention away from Morey's true skill that really sets him apart from other GM's.... managing the salary cap.
Yeah.....Morey probably already knew about this. Think about it this way..........we already had one of the top rim protectors in the league signed, but Morey still felt it necessary to go after the absolute best defensive center in the league. Neither being rejected before or Dwight's injury changed that. I'd be interested to see if this has changed since he joined the Rockets. Are opponents still wary of going into the paint against Dwight and settling for mid-range jumpshots?
I mean...a lot of teams were trying to acquire Dwight. He was one of the biggest free agents. Analytics or no analytics, it's pretty easy to recognize his talent level.
I think you nailed it. I can't remember what podcast it was on but Zach Lowe essentially said the same thing. He said there are only a handful of guys in the world who know the ins and outs of the CBA and having one is a gigantic advantage to your team.
I think he is just willing to work hard and willing to take risks that might make him look bad. Other GMs don't make 100 phone calls just to get a marginal talent improvement that might come from brokering a three team trade. They also don't trade one player for another at the same position because it is so easy to find the winner.
Perhaps this is why last season I searched relentlessly for statistical signs that Harden was a bad defensive player. I used every available free or moderately priced source of data I could get my hands on. I had databases with tens of thousands of lines of ascii text. Everything I dug up pointed to Harden being an above average defensive shooting guard. But then there were those YouTube videos. Perhaps Harden kept the ball from being passed to his defensive assignment, and then gave up much to early of the play? I am still searching for an answer.
If you remember that Battier article in the NYT, a lot of what Goldsberrey is saying was already known (player tendencies, etc.) The novel thing that he brought to the NBA was an effective visualization of the data. Whether or not teams were able to use it more effectively, we do know that this new method made it infinitely more accessible to the general public. For example, we knew before hand that Josh Smith was an incredibly inefficient chucker with poor shot selection. Goldsberry's graphic reinforced that further in a way that was more aesthetically pleasing and accessible for the public.
Doesn't take a genius to see that players avoided taking on Dwight in the paint. Dwight passed the eye test way before there was any advanced analytics to back it up. Being able to quantify an observation and have the data back you up reduces uncertainty in your assessment of a player, but it was hardly necessary to assess Dwight's eliteness.
Those videos are really unfair. In an 82 game season harden averages about 35 minutes a game meaning on average last season he played a total of 2900 minutes give or take. That video was 11 minutes. I could make an 11 minute video of lebron's defense and make him look like a defensive sieve too. I think that video was one of those things that lazy sports fans flocked to to attack James because they did not care enough to form an opinion by themselves.
People also don't really know that Morey's methodology for trading stems back from his days when he was in Sales. Not enough is made of this. I wish I could find the podcast interview he did where his trading style was brought up and he referred to his "pipeline" and hinted at farming techniques. Morey and his team really more than anyone else in the league except maybe now Philly (actually Sam Hinkie was probably the driving force of this more than Morey) really understand what motivates their buyers (in this case what motivates other teams to make moves), and they do an amazing job farming these buyers over time. They make sure no conversation ever has a closed door, and they realize that the more conversations they can have, the more they will uncover in order to ultimately find a "PAIN POINT" in their front office, which in turn leads to Morey coming back with a solution. Yes this does lead to the Rockets players names being brought up around the league, and it seems like Morey is trying to sell off every player on his roster, but nothing could be further from the truth. The FACT is, if Morey is talking to another team about a player, pick or whatever, its almost surely because Morey has found out that the other GM ultimately sees value in what he is discussing. In the past with NBA and even sports transactions, the ideals are backwards. You call up teams when you need to move a player. This is a huge problem in the world of professional negotiations. The best time to be making calls to teams is when you can give the other team something they want. Not when everyone knows you are needing to get rid of something. Farming these trades and treating them with the same proven sales methodology that works for successful business folks out in the corporate world is really what has made one of the biggest impacts in Morey's tenure with the Rockets. Only now are seeing that Hinkie was really the driving force behind that, and now that he's the boss man, we see now that Philly is going to take that method and go crazy with it.
man...Hakeem's or even Deke's numbers at his prime would be off the charts! I'm curious as to how they compare.
Yeah, it was only four months AFTER Morey heard Goldsberry that he signed Dwight. Don't worry about Morey trying to acquire Dwight a year earlier...
Stats do not show the full defensive picture. Youtube videos do not show the full defensive picture. But this doesn't mean he was in between; there is no middle ground with Parsons and Harden last year. They both sucked, and that is why the Rockets were a middling defensive team even with a 3 time DPOY and 2nd team defensive PG. Really, the only real answer is to go through the whole game; however, applying defensive team rankings context to a specific player can give a general idea.
I'm quite certain The Dwight Effect was a HUGE part of the pitch the Rockets gave Dwight when they met him in L.A. Supposedly, he's embraced the statistics that Morey gives him, and he tries to implement them into his game plan.
dobro1229 with some great posts in this thread! This is precisely Morey's strength over other other GMs. The number of transaction avenues this office must be working at any given moment shows incredible diligence and hard work. No wonder Les Alexander is perfectly happy with the way they make their decisions.
