Big Data Revolutionizing Sports Stats

Tiny cameras, big data crunching servers deliver millions of stats during NBA games, UEFA soccer matches.

Big Data analysis of Tony Parker dribbles

Tony Parker of the San Antonio Spurs (with ball) dribbled 839 times in a 107-96 win over the Oklahoma City Thunder on Feb. 4, the most dribbles by a player in a game that used SportsVU this season.

As network television cameras capture Miami Heat superstar LeBron James slamming a monster dunk or San Antonio point guard Tony Parker swishing a 3-pointer during the National Basketball Association finals high above the Spurs’ home court tiny cameras are working unseen. Their job? To record massive amounts of sports data from the rafters of AT&T Center.

This new data is revolutionizing the way coaches and their staffs use real-time statistics to adjust games on-the-fly or during the season. Data such as how many times a player drives to his right versus his left, who is shooting from what spot, the time of possession for each player and team. How many passes, how many dribbles, how many rebounds and from which quadrant of the court.

The beauty of sports lies in the unpredictable — the upset, a Cinderella team making the Final Four, the bicycle kick in soccer. But the surge in “big data,” even in sports, is another example of technology continuing to change the world in all corners and all industries.

Court Vision from Big Data

Half of the NBA’s 30 teams now use a relatively new sports data collection service called SportVU, which collects upwards of 1 million data records per 48-minute basketball game.

Teams use the data to track player stats and game flow a lot more in depth than a paid intern with a clipboard keeping a stat sheet on the sidelines.

“The cameras we’ve installed are using algorithms to capture 1 million data records and we need these processors to be powerful and quick,” said Brian Kopp of STATS (the company that owns the SportVU technology). “The (Intel Xeon) HP workstations we use needed to be not huge but powerful.”

SportVU cameras are synched with complex algorithms extracting positioning data for all objects on the court, capturing 25 pictures per second. Each picture is time-stamped and automatically processed by a computer, which connects the data to the play-by-play feed and delivers a report within 90 seconds of a play.

Big data analysis of Kevin Durant shot attempts

A SportVU heat map of the Oklahoma City Thunder's Kevin Durant’s open shot attempts during a recent playoff game.

Almost instantly, coaches and statisticians have this information at their disposal on their laptops or tablets.

Big Data on the Soccer Pitch

SportVU cameras are also installed in most arenas where UEFA (Union of European Football Associations) Champions League matches are held all over the world. Servers are busy taking collected data and creating heat maps to show where most of the game’s action takes place, how many miles players are running, their speed with and without the ball, and the nearest defender to their shots on goal.

Coaches use this data to adjust game plans and meet with individual players to talk about ways to improve their performances. For instance, maybe a player will have three more seconds per touch to get a pass off if he stands 5 feet away from a defender versus 3 feet. Or perhaps coaches can change offensive attacks based on the heat map of where the defense stands most of the game.

From Missile Tracking to Sports Stats

Kopp says the SportVU technology was originally developed by the Israel military to track missiles using advanced optical recognition.

After his military career, developer Miky Tamir took his SportVU technology public and started a company in Israel tracking soccer games. STATS, based in Chicago with offices all over the world, bought SportVU in 2008.

Looking ahead, Kopp said the company is trying to get rafter cameras hung in National Hockey League arenas for next season, but no teams have yet signed on the dotted line.

As for college basketball in the United States, SportVU covered Big 10 Conference arenas last year and Kopp hopes STATS can secure a contract to provide data for the NCAA’s March Madness in 2014.

Big Data Revolutionizing Sports Stats

Tiny cameras, big data crunching servers deliver millions of stats during NBA games, UEFA soccer matches.

Big Data analysis of Tony Parker dribbles

Tony Parker of the San Antonio Spurs (with ball) dribbled 839 times in a 107-96 win over the Oklahoma City Thunder on Feb. 4, the most dribbles by a player in a game that used SportsVU this season.

As network television cameras capture Miami Heat superstar LeBron James slamming a monster dunk or San Antonio point guard Tony Parker swishing a 3-pointer during the National Basketball Association finals high above the Spurs’ home court tiny cameras are working unseen. Their job? To record massive amounts of sports data from the rafters of AT&T Center.

This new data is revolutionizing the way coaches and their staffs use real-time statistics to adjust games on-the-fly or during the season. Data such as how many times a player drives to his right versus his left, who is shooting from what spot, the time of possession for each player and team. How many passes, how many dribbles, how many rebounds and from which quadrant of the court.

The beauty of sports lies in the unpredictable — the upset, a Cinderella team making the Final Four, the bicycle kick in soccer. But the surge in “big data,” even in sports, is another example of technology continuing to change the world in all corners and all industries.

Court Vision from Big Data

Half of the NBA’s 30 teams now use a relatively new sports data collection service called SportVU, which collects upwards of 1 million data records per 48-minute basketball game.

Teams use the data to track player stats and game flow a lot more in depth than a paid intern with a clipboard keeping a stat sheet on the sidelines.

“The cameras we’ve installed are using algorithms to capture 1 million data records and we need these processors to be powerful and quick,” said Brian Kopp of STATS (the company that owns the SportVU technology). “The (Intel Xeon) HP workstations we use needed to be not huge but powerful.”

SportVU cameras are synched with complex algorithms extracting positioning data for all objects on the court, capturing 25 pictures per second. Each picture is time-stamped and automatically processed by a computer, which connects the data to the play-by-play feed and delivers a report within 90 seconds of a play.

Big data analysis of Kevin Durant shot attempts

A SportVU heat map of the Oklahoma City Thunder's Kevin Durant’s open shot attempts during a recent playoff game.

Almost instantly, coaches and statisticians have this information at their disposal on their laptops or tablets.

Big Data on the Soccer Pitch

SportVU cameras are also installed in most arenas where UEFA (Union of European Football Associations) Champions League matches are held all over the world. Servers are busy taking collected data and creating heat maps to show where most of the game’s action takes place, how many miles players are running, their speed with and without the ball, and the nearest defender to their shots on goal.

Coaches use this data to adjust game plans and meet with individual players to talk about ways to improve their performances. For instance, maybe a player will have three more seconds per touch to get a pass off if he stands 5 feet away from a defender versus 3 feet. Or perhaps coaches can change offensive attacks based on the heat map of where the defense stands most of the game.

From Missile Tracking to Sports Stats

Kopp says the SportVU technology was originally developed by the Israel military to track missiles using advanced optical recognition.

After his military career, developer Miky Tamir took his SportVU technology public and started a company in Israel tracking soccer games. STATS, based in Chicago with offices all over the world, bought SportVU in 2008.

Looking ahead, Kopp said the company is trying to get rafter cameras hung in National Hockey League arenas for next season, but no teams have yet signed on the dotted line.

As for college basketball in the United States, SportVU covered Big 10 Conference arenas last year and Kopp hopes STATS can secure a contract to provide data for the NCAA’s March Madness in 2014.