The Analytical Transformation of the NBA...and soon the NHL?

[devilsrule33]

A fantastic article today on some really impressive technology that is being used in the NBA by a number of teams. And it looks like it could fit for NHL use very soon. Not for everyone, but I thought a few people here might find it interesting. 

 

New technology and statistics will change the way we understand basketball, even if they also create friction between coaches and front-office personnel trying to integrate new concepts into on-court play. The most important innovation in the NBA in recent years is a camera-tracking system, known as SportVU, that records every movement on the floor and spits it back at its front-office keepers as a byzantine series of geometric coordinates. Fifteen NBA teams have purchased the cameras, which cost about $100,000 per year, from STATS LLC; turning those X-Y coordinates into useful data is the main challenge those teams face.¹

Some teams are just starting with the cameras, while others that bought them right away are far ahead and asking very interesting questions. Those 15 teams have been very secretive in revealing how they've used the data, but one team that has made serious progress — the Toronto Raptors — opened up the black box in a series of meetings this month with Grantland.

The future of the NBA, at least in one place, looks like this:

That's Jason Kidd hitting a 3-pointer off a Carmelo Anthony pick-and-roll in the first quarter of Toronto's February 22 home win over the Knicks; the Knicks are in blue, passing the little yellow ball around, and the Toronto players are colored white. It looks simple, but the process of getting there took a bunch of people, including three Toronto front-office employees, more than a half-decade of work. In simple terms: The Raptors' analytics team wrote insanely complex code that turned all those X-Y coordinates from every second of every recorded game into playable video files. The code can recognize everything — when a pick-and-roll occurred, where it occurred, whether the pick actually hit a defender, and the position of all 10 players on the floor as the play unfolded. The team also factored in the individual skill set of every NBA player, so the program understands that Chris Paul is much more dangerous from midrange than Rajon Rondo, and that Roy Hibbert is taller than Al Horford.²

That last bit — the ability to recognize individual player skills — is crucial for the juiciest bit of what the Raptors have accomplished: those clear circles that sort of follow the Toronto players around and have the same jersey numbers. Those are ghost players, and they are doing what Toronto's coaching staff and analytics team believe the players should have done on this play — and on every other Toronto play the cameras have recorded.³ The system has factored in Toronto's actual scheme and the expected point value of every possession as play evolves.⁴ The team could use that expected value system to build an "ideal" NBA defense irrespective of the Toronto scheme, but doing so today would be pointless, since part of the team's job is to sell a sometimes skeptical coaching staff on the value of all these new numbers and computer programs, says Alex Rucker, the Raptors' director of analytics.

"You need that coaching perspective," Rucker says. "But we are still looking for where the rules are wrong — areas where there are systemic things that are wrong with what we do on the court. But any system needs to comply with what the coaches want, and what the players can do."

 

http://www.grantland.com/story/_/id/9068903/the-toronto-raptors-sportvu-cameras-nba-analytical-revolution

 

 

NHL next?
Thanks in part to the large amount of overlap with NBA and NHL buildings in the same city, the Toronto situation may change soon
if SportVU gets into the hockey world as Kopp expects.

"We were hoping to start this year, but because of the lockout we didn't. We did some initial testing, and the players themselves aren't very hard to track. There's a lot you can do with player
movement and positioning, and I know hockey is at least opening their eyes to analytics," Kopp said. "I know there's been this thought, 'Well how can you analyze it when it's so fluid?'

"But if you can actually measure the fluidity, then you're getting somewhere. Of all the sports that are going to be next, I think hockey would be up there."

 

http://www.cbc.ca/sports/basketball/opinion/2013/03/nba-teams-see-game-in-new-way-through-sportvu-system.html

[EdgeControl]

that example must have been taken from the last 5 mins of an NBA game, guys were actually playing some good defense!

[Zubie#8]

Thats pretty cool!

[Triumph]

NBA defense is better than college defense.

 

dr33, the thing that will block this in hockey are the GMs/league themselves.  They already had the beginnings of this sort of data when the Fox Glow Puck was used.  Given how terrible the league is at data keeping in general, and how the NHL GMs are a huge old boys club (seriously, you're either an ex-NHL player or the son of a GM/coach or both), it's unlikely that we see this in the next 10 years. 

[DevilMinder]

I think once there is a competitive advantage realized the adoption of analytical tech into the sport will be rapid. Overviews of player movement and set plays that are pretty much printed out for coaches would come into play in any series. Of course the case could be made that a good coaching staff could pick that up on its own.

 

I was looking for NHL scoring heat maps and came across this blog post, thought it was interesting.

 

 

http://blogs.thescore.com/nhl/2013/03/12/why-advanced-statistics-approaches-hockey-completely-wrong-but-still-gets-things-right/


A piece on the corsi system being misused.

[Marshall]

Very interesting.

[Sarge18]

One of the most compelling thing this could create for hockey is the justification of patterns in both players and teams.  Take Ovechkin for example.  His offensive game is oft criticized for being 1 dimensional in how he enters the offensive zone and what he tries to accomplish once there.  If you could complile a seasons zone of tracking for his zone entries with and without the puck, and create a probability of what he may do, you have a very powerful coaching tool, both to help a player improve, and how to stop a player or team.

 

This all could be extrapolated to a team and the system that they play, what is a teams tendency when coming out of the corner, how to they want to cycle, how do teams breakout.  It takes what we all see with out eye tests and quantifies the occurrences to tell us if what we saw was an aberration or a trend.  The distinction can make all the difference in how you approach a team.

 

I know you can accomplish similar things in game tape review right now.  But this is real time data, allowing a team to make mid period or intermission adjustments much faster than they would be able to do otherwise.

[njdevsftw]

NBA defense is better than college defense.

 

dr33, the thing that will block this in hockey are the GMs/league themselves.  They already had the beginnings of this sort of data when the Fox Glow Puck was used.  Given how terrible the league is at data keeping in general, and how the NHL GMs are a huge old boys club (seriously, you're either an ex-NHL player or the son of a GM/coach or both), it's unlikely that we see this in the next 10 years. 

 

 

On a sidenote: I miss the Fox Glow Puck. 

[devilsrule33]

NBA defense is better than college defense.

 

dr33, the thing that will block this in hockey are the GMs/league themselves.  They already had the beginnings of this sort of data when the Fox Glow Puck was used.  Given how terrible the league is at data keeping in general, and how the NHL GMs are a huge old boys club (seriously, you're either an ex-NHL player or the son of a GM/coach or both), it's unlikely that we see this in the next 10 years. 

 

No doubt. Brian Burke, your typical thinking hockey GM, said as much at the Sloan Analytics Conference this year. I can not imagine NJ ever looking at such things as long as Lou is in charge. It is getting so lazy for people to say I'll trust my eyes over data when anyone making decisions is obviously using both tools. The NBA is ahead of the curve with GMs investing heavily in analytic tools, but the issue is getting many of the coaches to listen to this information.

 

The article shows the Raptors are having trouble themselves getting their coach to do that. Actually, some people think the Raptors allowed this article to be written to put external pressure on the team's coaches. There is clearly a divide there right now. And why allow this internal information out there?

[brodeur14301991]

It sort of already exists, here is a great research paper on the matter http://www.sloansportsconference.com/wp-content/uploads/2013/Total%20Hockey%20Rating%20(THoR)%20A%20comprehensive%20statistical%20rating%20of%20National%20Hockey%20League%20forwards%20and%20defensemen%20based%20upon%20all%20on-ice%20events.pdf

Edited March 20, 2013 by brodeur14301991

[Pepperkorn]

One of the most compelling thing this could create for hockey is the justification of patterns in both players and teams.  Take Ovechkin for example.  His offensive game is oft criticized for being 1 dimensional in how he enters the offensive zone and what he tries to accomplish once there.  If you could complile a seasons zone of tracking for his zone entries with and without the puck, and create a probability of what he may do, you have a very powerful coaching tool, both to help a player improve, and how to stop a player or team.

 

This all could be extrapolated to a team and the system that they play, what is a teams tendency when coming out of the corner, how to they want to cycle, how do teams breakout.  It takes what we all see with out eye tests and quantifies the occurrences to tell us if what we saw was an aberration or a trend.  The distinction can make all the difference in how you approach a team.

 

I know you can accomplish similar things in game tape review right now.  But this is real time data, allowing a team to make mid period or intermission adjustments much faster than they would be able to do otherwise.

Agree - this is just what I was thinking.  It's a step further in getting some meaningful data - you have context.  It IS not unlike watching tape so I don't see why it'd be such a bad thought.