RPI a big factor in Pitt's seeding
Jerry Palm wants everybody to know something about him.
"I'm a geek," he said. "A real-live geek."
Palm made the proclamation earlier this week after detailing what takes place in his kingdom -- The Land of the Rating Percentage Index.
He produces college.rpi.com, a comprehensive Web site dedicated to college basketball-related statistics, most notably the RPI.
The RPI is a formula used by the NCAA Tournament committee to help determine which teams are worthy of the tournament's 34 at-large bids. The formula is based on a team's winning percentage (25 percent of the total), its opponents' winning percentage (50 percent) and the winning percentage of its opponents' opponents (25 percent).
Pitt's RPI was 19 entering the week, even though the Panthers are ranked eighth and have a 20-4 record (10-3 Big East.) Palm said Pitt is an example of a team that is highly regarded in the polls yet its ranking is in question because its strength of schedule is 61st overall.
"Pitt doesn't have any great, great wins; it doesn't have any bad losses," Palm said. "They're under .500 under against top 50 RPI teams. As profiles go, they look no better than a No. 4 seed to me in the NCAA Tournament. The only thing going for them might be that it's ingrained in the committee's mind that they're doing great things based on the polls. But really, they'll have a hard time getting into the RPI top 15 even if they win the Big East Tournament. They'd have to go undefeated to be a No. 1 seed based on their schedule."
Pitt coach Ben Howland responded.
"The NCAA committee uses the RPI as a guide, but it's not the end-all," Howland said. "You have to factor how teams play at the end of the year and so many of the other things that go into it. Those numbers are helpful, but there's more to it."
Palm correctly predicted all 34 at-large teams for last year's tournament, which reinforced a theory that the NCAA makes its decisions based primarily on the RPI, even if it refuses to acknowledge that fact.
Palm's web page will absorb millions of hits over the next several weeks from RPI-starved college hoops fans, who are hoping to see their teams' rating move skyward as the NCAA Tournament draws near.
"When I started this in '94, you could have fit the number of people who followed this in a mini-van," Palm said from his home in suburban Chicago, where his kingdom exists in his living-room-turned-office. "Now, it's unbelievable what's going on."
The fairness of the RPI often comes into question, but it is not Palm's job to defend the formula. He simply crunches the numbers and presents them to those who surf into his Web site.
He has little trouble putting it together because his background with computers is extensive, including a longtime stint as an analyst for a major bank in Chicago until last year. His focus these days is solely on his ever-growing site.
"The demand is there," said Palm, a husband and father of two young children. "I've been able to give it to them."
Palm has a computer program in place that comprises the RPI formula. He discovered it after it was detailed in a 1994 newspaper article and insists it took him little time to put it together once he acquired the information.
"You have to remember, it's what I did for a living," he said. "Plus, I'm a huge Purdue fan and I wanted to keep up with my team."
The toughest part for Palm is getting ahold of every Division I schedule prior to each season. There are 327 Division I teams this year, which means he had to obtain each schedule via Internet, phone calls or leg work. Once he had them, he had to type each into his system.
"Very time consuming," he said.
After the schedules are processed, it's pretty much free sailing for Palm, who plugs in the results of each game during the season and the computer produces updated ratings.
Through the years, Palm has heard numerous complaints about the RPI. There is the obvious omission of a category for road victories, which once was part of the equation. The flaw, according to detractors, is that those who play frequently at home enhance their ratings over those willing to hit the road.
There is also the contention that good teams from small conferences are punished because even if they're exceptionally competitive, they're going to lose ground to schools from major conferences because of their schedule strength. Butler, which finished 25-5 as a member of the lightly regarded Horizon League last season, lost out on an NCAA bid to bigger conference schools such as Boston College, a Big East member which entered the tournament with a 20-12 mark. BC got the nod, at least partially, because its top-35 RPI was better than Butler's 77.
Palm said he does not feel bad for teams that lose ground due to weak scheduling. Not did he identify Pitt, but also pointed to other Big East members such as Georgetown.
"Georgetown got left out last year because of scheduling," Palm said. "It's a Big East tradition to play a lesser schedule and sometimes, it can cost a team."