The biggest breakthroughs in college basketball today aren't playing out on the court or in the weight room. They are happening on spreadsheets, like the ones that Jason Richards keeps open on his office computer at the University of Pittsburgh's Petersen Events Center.
There, he catalogs hours of action -- dunks, 3-pointers, blocks, steals -- and enters the data into a series of equations that will help men's basketball coach Jamie Dixon make decisions.
For example, when senior Talib Zanna grabs an offensive rebound, he may not score any points for the team. But he'll earn two points toward his player efficiency measurement -- just as much value as a slam dunk or a pair of free throws -- a relatively new concept in college basketball.
Analytics helped to transform Major League Baseball -- think the movie "Moneyball" -- in the past 20 years and have crept into National Basketball Association boardrooms in the past decade. But it wasn't until the past few years that the science has been embraced in college basketball.
Mr. Richards, Pitt's coordinator of video and analytics, and Mr. Dixon are among the leaders in the emerging study of analytics in the college sport.
"The game has changed," Mr. Dixon said. "Well, it's changed, but it hasn't changed. What teams are doing, you've got to keep up with the times."
The sport has always been littered with numbers that aim to measure a player's value. But by analytical standards, those statistics are rudimentary. Traditional statistics -- such as points scored, assists, rebounds, steals and blocked shots -- offer only a surface-level analysis of a player's contribution and can even misrepresent a player's true value.
And with 347 teams in the NCAA's Division I level -- and a maximum of 13 scholarship players on each of those teams -- the data can be overwhelming.
Athletic departments are spending thousands of dollars on computer programs to help better analyze data. Jobs once held by basketball experts are now being offered to graduates with finance, statistics and economics degrees, all in an effort to succeed in a crowded college basketball industry that generates $1.1 billion annually, according to the U.S. Department of Education.
The Butler University men's team under Brad Stevens, now head coach of the Boston Celtics, helped launch the analytical movement last year when it hired Drew Cannon, then 22, to fill the first statistics-only staff position in college basketball.
Mr. Cannon, who majored in statistics at Duke University, now works in the Celtics' front office.
Sports Illustrated described him as Butler's "secret weapon," and many college coaches have tried to emulate the success of the Indianapolis school and Mr. Stevens, who holds an economics degree.
Pitt, which opens its season tonight, expanded its video coordinator role during the offseason to include analytics, and Mr. Dixon chose Mr. Richards, who had worked on Pitt's staff for the past three years, to handle the new responsibilities.
Mr. Richards had become interested in statistics in 2008 while playing in the Miami Heat organization. A third ligament injury to his left knee forced him to retire in 2010.
He studied history at Davidson College in North Carolina, where he and current Golden State Warriors star Steph Curry played on a team that advanced to the Elite Eight in the 2008 NCAA tournament. After joining Pitt's staff in 2010, Mr. Richards acquired his MBA at Pitt's Katz Graduate School of Business. His parents played at Pitt and his sister, Lindsay, is an assistant on the women's team.
Mr. Dixon has always been interested in numbers and statistics, long before he began coaching at Pitt in 1999. He was a finance major at Texas Christian University and later earned a master's degree in economics as a graduate assistant at the University of California, Santa Barbara. His final project in his graduate-level econometrics class at UCSB in 1992 was a statistical analysis of NBA teams.
He and Mr. Richards read countless articles about basketball analytics, looking to see if those methods are something they can apply at Pitt.
This summer, Mr. Dixon asked Mr. Richards to review every offensive possession from all 31 games of Pitt's 2012-13 season. Each possession was placed in a category -- after a steal, after an offensive rebound, after a defensive rebound, after an opponent scored and after a dead ball -- and Mr. Richards documented the results of each of those possessions.
The findings supported much of what the Pitt coaching staff expected: They are most successful on offense after steals, which usually lead to a fast-break opportunity, and least successful after a dead-ball timeout, when the defense can get set up.
But it also revealed a valuable piece of information: Pitt's success rate after an opponent's basket is not much higher than its success rate after a dead-ball timeout. With that piece of information, Mr. Dixon determined his offense was taking too long dribbling up the court, giving defenses too much time to prepare.
"You can always watch the game and understand it," Mr. Richards said. "But once you get the numbers, you understand just how efficient you are as a team."
The system -- aptly titled Pitt Analytics -- is unique to Pitt, as far as Mr. Dixon and Mr. Richards know, but many of the analytics systems the team uses -- a plus-minus rating system, other measures of player efficiency -- are borrowed from others.
Danny Lawson, director of basketball operations at Duquesne University, uses a plus-minus rating developed by Miami Heat president Pat Riley to help set lineups at given points in the game. He, like Robert Morris head coach Andy Toole, regularly reads KenPom.com, a Salt Lake City-based college basketball analytics website that has turned into a must-read for college coaches.
"There's more data available now than there was 10 years ago -- lots more data," said Ken Pomeroy, founder of KenPom.com. "You finally have people out there making good use of it."
Michael Sanserino: firstname.lastname@example.org, 412-263-1969 and Twitter @msanserino.