January 31, 2017

Scouting by the numbers

By Judy Richter

When it comes to preparing for a game, the more that Stanford women’s basketball coaches and players know about the opponent, the better.

That’s why they traditionally have studied video of an opponent’s games to discern what defensive and offensive sets it uses and what each player’s tendencies are. They also study their own games to help their players.

Now they have even more information to study thanks to sports analytics, a statistical tool that “teams can use to their advantage,” said Peter Lee, who provides the numbers to the Cardinal team a week before each game.

Lee, a second-year graduate student in statistics, is a member of the Stanford Sports Analytics Club and began working with the team last year.

“I’ve always really loved basketball,” he said in a recent interview. After spending much of his youth in Los Angeles, he went to Korea, where he played high school basketball.

He earned a bachelor’s degree in economics at Vassar College in upstate New York. His first job involved working as an economic theory consultant, giving him a chance to learn how to code.

Using a database called Synergy Sports Technology, he creates charts about the team and players and compares them with opponents.

Four factors for success

In his view, there are four main factors in a team’s success:
  1. How well it shoots the ball
  2. Turnovers
  3. Rebounds
  4. Free throws
“Free throws are the most efficient shots” because there’s no one to guard the shooter, he said.

Long-range 2-pointers are the most inefficient, especially when compared with 3-pointers. The logic is that if the shooter takes a just a few steps back to get beyond the arc, the score goes up 3 rather than 2 points. Over the long haul, that can make a big difference.

He looks at every opponent on the basis of those factors. He tries to discover what types of shots individual players prefer so that Stanford defenders can make them uncomfortable.

Although some people like to look at points per game as a measure of success, he prefers points per possession.

And although it’s helpful to view video, it doesn’t tell the whole story. Peter’s information helps to reveal team and player tendencies through several games.

Stanford defense ‘one of the best in the nation’

Peter also scouts the Cardinal. “Our defense is one of the best in the nation,” he said.

In addition, he measures the productivity of various Stanford lineups, starting with the starters. Then he narrows it down to smaller combinations. For example, what happens when senior Briana Roberson and sophomore Marta Sniezek, both of whom usually play the point, are on the floor together?

He attends practices and home games to get a first-hand view.

His report for the Oregon State game in Maples on Jan. 4 totaled five pages and used numbers from the current season. (OSU won that game 72-69 in double overtime.)

Observations about Oregon State

The report starts with key observations about OSU, such as “Scores 0.90 points per possession, which ranks 25th out of 349 NCAA Women’s Division I teams (93rd percentile nationally, 4th in the Pac 12).”

It mentions “elite offensive rebounding rate,” slow-paced offense, “great shooting numbers,” “good care of the ball” and “good short and long jumpers and good post-up numbers.”

“Who should we leave open?” the report says. “Obviously, we don’t ever want to leave a perimeter player unguarded with the ball for five seconds, but there will be situations when we’ll need to help off certain players to either double or provide help defense.”

He lists three tiers of players, starting with those who can be left open and ending with those who should almost never be left unguarded.

Then there are various graphs, such as “Transition/Half Court Breakdown,” “Man/Zone Breakdown” and offensive statistics.

He measures shot distribution and efficiency in various situations.

Observations about Stanford

He concludes his report with similar observations and graphs for Stanford. His key observations say:

“Allows 0.69 points per possession, which ranks 48th of 349 NCAA Women’s Division I teams (86th percentile nationally, 4th in Pac 12); “Elite at guarding the hoop and very good at defending jump shots; “Allows very little transition defense to the opposition, doubly impressive also considering that they have a very good offensive rebounding rate.”

Peter’s main contact person with the team is associate head coach Kate Paye.

Kate comments on reports

The information in his team and individual analytics reports “helps us identify and clarify individual and team opponent’s strengths and weaknesses and helps us develop our game plan and teach it more effectively,” Kate said in an email.

“Being able to give our team numbers that demonstrate why we are doing things or need to do certain things against a particular opponent really helps (our) players … assimilate and buy into our game plans,” she said.

“Peter also does self-scout work where he does analytics on our individual players and our team. This helps us identify individual and team strengths and weaknesses and how we can help our team improve,” she concluded.

No comments: