PennStateHoops.com Discussion Forum

Basketball moneyball

http://www.slate.com/articles/sports/sports_nut/2013/02/nba_stats_gurus_can_t_work_together_anymore_that_s_a_problem.single.html

The last decade has seen tremendous progress in understanding the offensive side of the floor, but defense—where players must constantly rotate and cover for each other—presents a much knottier problem. Oliver believes that technology is providing the raw data to solve it, but all those NBA stat gurus working in isolation against each other aren’t close to cracking the code.

Where is that raw data coming from? Cameras that weigh about a pound and can fit in the palm of your hand. They’re provided by STATS, the global information behemoth, as part of its SportVU program, and they currently hang in the rafters belonging to 15 different NBA franchises, six per arena. They record everything: How far and how fast a player runs during the game, how many dribbles he takes when he has the ball, where he shoots from, the arc of his shot, whom he’s passing to, whom he’s not passing to, the spots where he get his rebounds, the spots where others get his rebounds. It’s endless. For each second of game play, the SportVU cameras capture the location on the court of the ball and each player 25 times, according to Brian Kopp, a VP at STATS. “You have 1 million data records per game.”

STATS acquired SportVU in 2008 from an Israeli company that had originally designed it for soccer. This is the system’s third year in the NBA since being recalibrated for basketball. STATS charges teams from $75,000 to $100,000 per season for SportVU, and the program has grown in that time from four initial teams to now half the league. The result is one of the largest and richest data sets not just in sports, but in the world.

Kirk Goldsberry, a visiting scholar at the Harvard Center for Geographic Analysis who also uses spatial mapping to analyze the NBA for Grantland and on his blog, Court Vision, is one of the few civilians who’s been granted access to any of the SportVU data. He’s working with another Harvard professor, statistician Luke Bornn, and four Harvard and MIT Ph.D. students in a semester-long project to break some of it down. “We look at that data and we say this isn’t just good data, this is the best space-time data,” Goldsberry says. “It’s just an incredible amount of information, regardless of whether it’s about NBA or anything else … There’s very few people who have ever seen any data like this.”

Bo Ryan would love this (provided he had a math guy on staff).

And a paper from the stats conference: The Dwight Effect: A New Ensemble of Interior Defense Analytics for the NBA (PDF)

[quote=“tjb, post:1, topic:3701”]http://www.slate.com/articles/sports/sports_nut/2013/02/nba_stats_gurus_can_t_work_together_anymore_that_s_a_problem.single.html

[quote]The last decade has seen tremendous progress in understanding the offensive side of the floor, but defense—where players must constantly rotate and cover for each other—presents a much knottier problem. Oliver believes that technology is providing the raw data to solve it, but all those NBA stat gurus working in isolation against each other aren’t close to cracking the code.

Where is that raw data coming from? Cameras that weigh about a pound and can fit in the palm of your hand. They’re provided by STATS, the global information behemoth, as part of its SportVU program, and they currently hang in the rafters belonging to 15 different NBA franchises, six per arena. They record everything: How far and how fast a player runs during the game, how many dribbles he takes when he has the ball, where he shoots from, the arc of his shot, whom he’s passing to, whom he’s not passing to, the spots where he get his rebounds, the spots where others get his rebounds. It’s endless. For each second of game play, the SportVU cameras capture the location on the court of the ball and each player 25 times, according to Brian Kopp, a VP at STATS. “You have 1 million data records per game.”

STATS acquired SportVU in 2008 from an Israeli company that had originally designed it for soccer. This is the system’s third year in the NBA since being recalibrated for basketball. STATS charges teams from $75,000 to $100,000 per season for SportVU, and the program has grown in that time from four initial teams to now half the league. The result is one of the largest and richest data sets not just in sports, but in the world.

Kirk Goldsberry, a visiting scholar at the Harvard Center for Geographic Analysis who also uses spatial mapping to analyze the NBA for Grantland and on his blog, Court Vision, is one of the few civilians who’s been granted access to any of the SportVU data. He’s working with another Harvard professor, statistician Luke Bornn, and four Harvard and MIT Ph.D. students in a semester-long project to break some of it down. “We look at that data and we say this isn’t just good data, this is the best space-time data,” Goldsberry says. “It’s just an incredible amount of information, regardless of whether it’s about NBA or anything else … There’s very few people who have ever seen any data like this.”[/quote]

Bo Ryan would love this (provided he had a math guy on staff).[/quote]

Kirk is a friend and fellow State College Area HS grad

DATA is good…it’s very good…

BUT decisions/conclusions should be made ONLY by the team’s coaching staff, who know their personnel, based on eyeballs and stats.

[quote=“tundra, post:4, topic:3701”]DATA is good…it’s very good…

BUT decisions/conclusions should be made ONLY by the team’s coaching staff, who know their personnel, based on eyeballs and stats.[/quote]

Nate Silver (geek) killed the pundits (eyeballs) in predicting the 2008 and 2012 elections.

I won my apartment’s tournament pool a few years back simply by going off of offensive and defensive efficiency statistics.

I called earlier in the year for Sasa to shoot more because of his offensive efficiency numbers.

I’m sorry but I’m afraid the tide of history is against you. I, for one, welcome our new geek overlords.

1 Like

Do they use SportVU to capture how many times they carry the ball during their killer crossover moves? :slight_smile:

1 Like
[quote="tundra, post:4, topic:3701"]DATA is good................it's very good............

BUT decisions/conclusions should be made ONLY by the team’s coaching staff, who know their personnel, based on eyeballs and stats.[/quote]

I’m sorry but I’m afraid the tide of history is against you. I, for one, welcome our new geek overlords.

The coaching staff will have a better understanding of the conditional probabilities that arise in various situations over the course of a game.

Here’s the Wired article mentioned in another thread:

http://www.wired.com/2014/10/faster-higher-stronger/

Chambers should get one of these systems authorized for the BJC. $100,000 is chump change in college athletics.

[quote=“tjb, post:8, topic:3701”]Here’s the Wired article mentioned in another thread:

http://www.wired.com/2014/10/faster-higher-stronger/

Chambers should get one of these systems authorized for the BJC. $100,000 is chump change in college athletics.[/quote]

The cost for colleges is less than $100K.

Duke, Louisville, and Marquette used the system last year. Duke was the first college team to install it last fall. Marquette got to use the system because they share an arena with the Milwaukee Bucks, who have the system installed. Louisville added the system after Pitino heard that Duke got it.

It was also used at the Big Ten Tournament.

I’ll volunteer to be the PSU sportsvu analyst. :slight_smile: who do i contact?

Question…does a human being have to take all of the video and actually input the data into some type of spreadsheet? Or does the programming software somehow actually ‘know’ the players (maybe by number or something else) and who has the ball, etc…and the data automatically somehow is put into usable type data. If it is a human, it must take somebody days to be able to watch everything and log everything into a usable form.

The camera knows.

wouldn’t really be worth 100k if somebody had to manually input all the data. :smiley:

1 Like

While this is not the case, NBA teams with this system did have to develop their own in house program for dealing with the huge amounts of data it spit out. I’ll try to find the article I read that described the Raptors use…

http://grantland.com/features/the-toronto-raptors-sportvu-cameras-nba-analytical-revolution/

The Atlhletic ($$$)* has a good story about San Francisco’s reliance on analytics, which is being credited with the program’s recent success.

Three assistants split the offensive, defensive and rebounding calculations from practice film. An hour-long five-on-five practice session equates to about 90 minutes’ worth of data entry for each assistant. Smith serves as the final arbiter, deciding perhaps if a defender used a hard hand on a closeout or a made a pedestrian effort.

None of this, in itself, is particularly unusual. Every team reviews practice film. Every staff grades its players’ performances. The difference at San Francisco is that the numbers have the final say, and everybody knows it. Those with the best stats play the most, regardless of recruiting ranking or reputation. Smith’s staff must fight its own biases toward more physically gifted players. Trust the numbers.

“It’s the buy-in that separates us from other people,” says assistant coach Todd Golden says. “It’s a true meritocracy.”

Players know all this before they enter the program. In what he calls “the nerdiest recruiting pitch of all time,” Smith often appeals to a player’s intellect when explaining his concept. “They break it down like a PowerPoint presentation on your recruiting visit,” says Ferrari, a senior whom Smith re-recruited to USF after Ferrari spent 2015-16 (the last season before Smith’s arrival) at a community college. “Your head is spinning a little bit. I was definitely a bit confused at first.”

That sales pitch might not appeal to everyone, especially highly rated prospects who believe they should play no matter what. Those guys probably weren’t coming to San Francisco anyway. Like the Oakland A’s of “Moneyball” hero Billy Beane, Smith seeks out the outliers — guys whom the numbers love more than talent evaluators.


* The Athletic had a special 40% off offer that ended yesterday - $2.99/month - but is still offering $3.49/month for the first year: http://theathletic.com/cbbtipoff40 . I am less enthusiastic about the publication than I was at the start - it has kind of lurched into a glorified fanzine, like Rivals or 247, in my view.

Just referenced the above in the coaching carousel thread - knew I had mentioned it before. I know - weird to post the same thing over and over again.

Onward:

Phillips says complimentary things about “Moneyball,” and his own book is more a correction than a refutation. He brings considerable historical knowledge to the task of establishing a point that does not actually seem all that controversial, which is that scouting involves measuring and scoring involves judging. “Facts don’t just appear,” as Phillips puts it. “They must always be made.”

But …

Lewis’s book has a lot of examples where scouts got it wrong but scorers got it right, so it’s regrettable that Phillips doesn’t provide much in the way of examples where the reverse is true. That may be because he wants to make a more philosophical point about the nature of data—that they’re always a hybrid of objectivity and instinct, analytics and intuition. He concludes that scouting has as much claim to being scientific as scoring does: “it serves as a well-developed and well-crafted set of heuristics for arriving at stable, generalizable, and reliable facts about the natural world.” He argues that since teams continue to use scouts, their experience must be irreplaceable.

:thinking:

And possibly of local interest:

It’s an adage of professional sports that lucky breaks, like bad calls, even out in the course of a season. (Actually, given a normal distribution of lucky breaks, there must be outliers. There is a luckiest person in the world. Probability theory requires it.) A major aspect of scoring [in the sense used here, “scoring” means “analytics”], therefore, is figuring out how to take luck out of the equation.

In his new book, “Sprawlball,” Kirk Goldsberry, a former San Antonio Spurs executive, describes the strategy bluntly: “With the exception of layups and dunks, two-point shots are simply dumb choices.”

When I grew up in basketball, people like Penn State coach Dick Harter played the opposite percentage: the entire game was about getting it inside for a layup, a foul, or both.

The three-point shot was the attempt to stop that. It was a really boring thing to watch.

It’s amazing to me that it’s taken this long for the pendulum to swing back.

More on “Sprawlball”