Jerry Palm remembers when the all the people interested in Rating Percentage Index and NCAA brackets could "fit comfortably in a minivan," as he likes to joke.
That was before the basketball world was swept up by this statistical storm, which is brewing stronger these days.
"People just hunger for this information," said Palm, one of the earliest examples of a "bracketologist," now a well-known term. "My Twitter feed blows up this time of year."
Perhaps that's because the general public is starting to catch on. The NCAA has worked harder to publicize the method it uses to select the 68 teams for the men's basketball tournament -- a process many have perceived as a complicated one. They've opened a yearly mock draft to the media to help spread awareness, which has, as a byproduct, spread interest as well.
But when Palm first started calculating and updating strengths and weaknesses of teams and projecting that into bracket form, he said he and ESPN's Joe Lunardi were the only ones doing it. "Now I think there's like 80 people -- anybody with a computer and too much time on their hands is doing it," he said.
It has become such a fad that a burgeoning number of basketball fans is growing more dissatisfied with the modest tools -- mainly RPI, the formula the selection committee uses to weigh teams against each other. Palm, who works for CBSSports.com, had Minnesota at No. 76 as of Friday, with an RPI of 0.5612, a tumble from a couple of weeks ago when the Gophers were projected to make the NCAA field. Syracuse is No. 1, followed by Michigan State.
Palm said the two most common complaints he hears are opposites. Ether people are annoyed that a formula of any type is being used to help pick teams for the tournament, or they want something that takes into account more than the three components that determine RPI: team record, opponents' record and the record of their opponents' opponents.
But Palm says those pleas are both off base because the selection process needs some structure -- and just as important, the simplicity of that structure is key. Despite some flaws -- for example, RPI weighs home and road games too distinctly, Palm said, and it completely ignores any games played outside of Division I -- Palm thinks the original system is still probably the best one.
"They don't really need anything more sophisticated than that, and I don't really think anybody wants a computer making the decision," he said. "And the more sophisticated of a computer ranking system you use, the more people will expect you to use that to make the decisions. I think it's good enough for how it's being used. Because they want to measure how teams do against good competition. And you look at the top 50 RPI teams and the top 50 Sagarin [rankings] teams and the top 50 whatever ranking system you want to look at, you're going to see roughly the same 50 teams."