MADISON, Wis. — Three letters that help determine whether a bubble team sneaks into the NCAA tournament field, is an at-large lock, or deserves a higher seed are at the same time dreaded and embraced by college basketball coaches across the land.
The Ratings Percentage Index, better known as the RPI, annually becomes the talk of the college hoops landscape in the final weeks leading up to the 68-team tournament draw announcement in mid-March. It is a significant, and sometimes controversial, mathematical equation used by tournament committee members to pick the very best teams that belong in college basketball's showcase tournament.
Among Big Ten coaches, its effectiveness is up for debate.
The RPI is based on three factors: Twenty-five percent of the formula measures a team's winning percentage, 50 percent measures the winning percentages of a team's opponents, and the other 25 percent measures the winning percentages of a team's opponents' opponents.
It does not take into account margin of victory or statistics other than win/loss results. And because of its design, who a team plays is often more important than whether a team actually wins.
Michigan State's Tom Izzo is among the coaches who consider the RPI to be a flawed system.
"Sometimes those RPIs get distorted by teams that go in the tank or injuries or different things that happen to your opponent," Izzo said. "If you get an injury, it counts different ways. But what about if your opponent does and a good team becomes a bad team?"
Izzo cited this year's Pittsburgh team as an example of the pitfalls of the RPI. Pittsburgh starting guard Tray Woodall missed two months of the season with a groin strain and abdominal tear. During his absence, the Panthers took a nosedive, beginning Big East play 0-7. Pitt won four consecutive games upon Woodall's return, but the RPI does not take into account that injury as part of Pittsburgh's overall resume.
"In this day in age with everybody transferring, it seems like more injuries than normal, or distractions because of suspensions and things, it's hard to schedule the right way," Izzo said. "That makes it difficult, too. Sometimes it's not your fault, but it works out hurting you in the long run. So the eyeball test still has to be part of it."
Coaches have found ways to skirt the RPI's measurements in recent years.
Games against Division II opponents don't count toward the RPI, and some coaches schedule early-season games against D-II schools instead of playing low-level Division I programs that could hurt an RPI. In 2009, Utah earned a No. 5 seed in the NCAA Tournament even though it lost at home to Division II Southwest Baptist.
Teams in mid-major conferences, particularly the Missouri Valley, tend to schedule tougher nonconference games because they reflect better in the RPI — even if those games turn into losses.
Of course, it is important to note that the RPI is not the only determining factor in selecting teams for the NCAA Tournament.
Other data sets include the Logistic Regressions/Markov Chain (LRMC), which uses information about home-court advantage and margin of victory. The Sagarin Ratings and Ken Pomeroy ratings also are factored in. Both use different mathematical equations to determine a team's ranking.
"I think it's a good measuring stick, but it's only one tool," Purdue coach Matt Painter said of the RPI. "It's just one tool among many tools in the entire process in how they come to their answer of who should be in and who shouldn't.
"A lot of people say they just solely go on the RPI, and that's not accurate."
In an effort to better educate college basketball fans and media members on the selection process, the NCAA recently posted information on its website that defines the committee's decision-making process. The posting includes explanations of the RPI, strength of schedule and records against top-50 and sub-150 RPI teams. All of those measurements, however, are based directly on the numbers provided by the RPI.
The RPI has been in existence since 1981, but before the 2005 season it was updated to account for differences in home, away and neutral-site games. A home victory now counts as six-tenths of a win, and a road victory counts as 1.4 wins. A home loss, however, counts as 1.4 losses, while a road loss counts as 0.6. A neutral game counts as 1.0 wins or 1.0 losses.
The idea behind the adjusted figures was that home teams generally win two-thirds of the time, which means a victory at home is considered less valuable than one on the road.
Count Ohio State coach Thad Matta among the legions of folks who don't fully grasp all the statistical data.
"I'm not exactly sure how the calculations work," said Matta, whose team is ranked No. 7 in the latest RPI. "But it appears to me over the test of time that it's been pretty accurate in terms of how teams finish up, who's at the top, who's at the bottom. It appears to be pretty effective in that regard. But I don't know how it's all calculated in terms of what goes into it."
Like Matta, Wisconsin coach Bo Ryan believes the RPI has withstood the test of time, even if it isn't the sole determining factor in selecting tournament teams.
"We've gone through as a conference trying to bring in people who have been on the committee to tell us exactly what it is and to what extent each factor is used," Ryan said, "whether it was strength of schedule or last 10 (games) or things like that. It's just a tool that a committee has to use."
The RPI is used to help determine the 37 at-large teams selected by the committee. The other 31 teams receive an automatic bid for winning a conference tournament or regular-season crown if there is no conference tournament.
The Big Ten is the strongest conference nationally in the RPI this season, with six teams — Michigan State, Ohio State, Michigan, Indiana, Wisconsin and Northwestern —ranking in the top 50. All but Northwestern are currently considered locks for the big dance.
Although the RPI might be a flawed system, its use by the selection committee demonstrates that it continues to be considered a valuable tool in determining NCAA Tournament teams. The most important factor, of course, is that a team wins as many games as it possibly can to stand out amid a sea of statistical information.
"I think it's good because it brings in very pertinent data," Indiana coach Tom Crean said. "The way that they gauge it, it's hard to argue with it. If you play a tough schedule and you're in a really good league like we are and you get wins and you get some road wins, those types of things all factor in.
"The biggest term you hear right now is body of work when people talk from the committee or there's talk about the committee. I think the RPI helps gauge that body of work."
Follow Jesse Temple on Twitter.
