(CNN) — Here’s a good tip: The University of North Carolina is going to win the NCAA men’s basketball tournament.
A Georgia Tech professor’s computer model suggests UNC will win the NCAA men’s basketball tournament this year.
At least that’s the prediction of Georgia Tech professor Joel Sokol, whose statistical model correctly selected last year’s Final Four, championship game and tournament winner.
Be glad he’s not in your office betting pool.
Finding some sort of rationality in March Madness, which begins in earnest Thursday, has been an American pastime for decades. Tournaments are everywhere, and from sports television to the dinner table, everyone seems to have predictions about which team will take first place and why.
But in recent years, “bracketology”, to resolve the problem of simple elimination /topics/Basketball” class=”cnnInlineTopic”>basketball tournament is sometimes called, has increasingly become the scientific enterprise that its name suggests. It’s even a subject on which university professors and professional statisticians stake their reputations. 
Watch Sokol explain his predictions »
According to Kenneth Massey, whose website, at least 40 people or groups use statistical models to analyze college basketball and rank its teams online. masseyratings.comallows users to compare analyses.
Anecdotal evidence suggests that more people are participating in the game, perhaps because the raw data used to run the equations (team records, point margins, etc.) is easily accessible on the Internet.
All of these derived predictions complement the official computer model used by the NCAA, called the Scoring Percentage Index. But a selection committee of live people still ranks teams and places them in tournament categories.
At their core, computer models all work like question machines, said Jeff Sagarin, who has done computer evaluations for USA Today since 1985.
Different people offer different slices because they ask different questions.
Sagarin’s equations ask three questions: “Who did you play, where did you play, and what was the outcome of each specific play?” The computer keeps repeating these questions in an “infinite loop” until it finds a solid answer, he said.
Sagarin arranged the formula as such in part because he thinks home-court advantage is a big problem in college basketball.
Other models ask different questions or give questions different weights. Georgia Tech’s Sokol, for example, cares more about the margin of victory than where the game was played.
He got into NCAA basketball by making predictions after watching his team, /topics/Georgia_Tech_Yellow_Jackets” class=”cnnInlineTopic”>Georgia Techlost a blow in 2002 when Tennessee hit a buzzer-beating shot from half court.
Sports analysts made a big deal about the loss, he said, which struck Sokol as statistically stupid. For him, a close match might as well be a tie: a coin toss could predict the winner.
“This last move makes almost no sense,” he said.
It took a few years of testing for Sokol to begin to believe in his computer model, which was the subject of an academic paper in 2004. That year, Georgia Tech reached the Final Four, as Sokol and his colleagues predicted.
“They really made us believe in our own system,” he said.
Sagarin, who has one of the oldest and most respected prediction equations in the industry, said a year or two of success doesn’t mean a model will continue to correctly predict the tournament outcome.
He grew up as a sports junkie who was “too skinny” to succeed on the court or field. For a time, he used his extensive knowledge of statistics and college basketball to bet on the annual NCAA tournament.
Buzzer shots became the bane of his existence, and in 1983, after Houston lost to underdog North Carolina State on a last-second dunk in the tournament final, Sagarin gave up betting.
“Every year I try to make minor improvements to the computer model and do off-season testing, but it all goes out the window when guys hit three-pointers at the buzzer,” he said.
“There’s a lot of luck in life. Life is a series of coin flips, and human beings try to describe the logic behind it… when in fact it can be random.”
Dan Shanoff, who blogs about sports at danshanoff.comsaid that intuition is more important than statistics, but taking a look at the numbers can never hurt.
In the end, he says, it’s all chance.
“You think you know everything about college basketball and all of a sudden your mom finishes ahead of you (in a tournament pool), which happens to me every year with my mom,” he said.
Shanoff’s instincts tell him that Louisville will win the men’s tournament this year.
Georgia Tech’s Sokol puts Louisville in Final Four with /topics/University_of_North_Carolina_at_Chapel_Hill” class=”cnnInlineTopic”>UNCPittsburgh and the University of Memphis.
He said Michigan State and Oklahoma are ranked too high and should be upset, statistically speaking.
While all of this information may be useful to people betting on the Big Dance, Sheldon H. Jacobson, a computer science professor at the University of Illinois, said he is not filling out a bracket, despite the fact that he has developed a computer model for the tournament.
“I enjoy the matches and the tournament, and not filling out a draw gives me the freedom to do that,” he said.
All about /subjects/Basketball”>Basketball • /topics/Georgia_Tech_Yellow_Jackets”>Georgia Tech Yellow Jackets • /topics/University_of_North_Carolina_at_Chapel_Hill”>University of North Carolina at Chapel Hill
