Basketball pool picks not going so great? Here is why, says CU-Boulder math professor

It took 48 games for the first and second rounds of the 64-team NCAA men’s basketball tournament to be decided, and millions of fans are deliriously awaiting the Sweet 16 to commence today.

But is there anyone left among the tournament pickers still in the hunt for Warren Buffet’s $1 billion offer for filling out a perfect bracket? Dream on. According to University of Colorado Boulder Professor Mark Ablowitz, chair of the Department of Applied Mathematics, the odds of picking all 48 games correctly for the Sweet 16 from the original 64 teams by selecting randomly, like a coin flip, is approximately 1 in 280 trillion, or 1 in 280,000,000,000,000.

Sound tough? Really, really, really tough, says Ablowitz. But the odds of picking the ultimate winner from the full 63 games of the NCAA basketball tournament randomly are breathtaking: about 1 in 9.2 quintillion, or 1 in 9,223,000,000,000,000,000.

“These aren’t just numbers,” he said. “The real essence here is the issue of the behavior of large numbers. When numbers get exponentially larger and larger, things get out of hand very, very quickly,” said Ablowitz, who specializes in nonlinear optics, water waves and applications of complex analysis.

According to journalist Eddie Pells of The Associated Press (who no doubt had some help from a math expert), if every possible combination of NCAA tournament winners and losers in the field of 64 was filled out on its own sheet of paper, it would weigh about 184 trillion tons -- more than 500 million times as much as the Empire State Building.

Others have calculated that picking all 63 NCAA tournament games correctly would be about the same odds as shooting four hole-in-ones in a single round of golf, or winning three consecutive Powerball lotteries.

“I’m not a basketball maven,” Ablowitz said. “But if I had wanted to improve my personal odds, I would have looked at 10 high-profile sports websites that gave the odds for each game, then averaged those. Using 10 sources rather than one source smooths the numbers out a little, so you wouldn’t be putting all of your money in one basket.”

Ablowitz said sporting events are notoriously hard to predict, citing the 2014 Super Bowl, in which the Denver Broncos -- favored by 2.5 points over the Seattle Seahawks by Las Vegas oddsmakers -- were annihilated 43-8. And player injuries and illnesses can play a large role in athletic events, he said.

As far as the prediction skills of successful statisticians go, Ablowitz pointed to Nate Silver of FiveThirtyEight.com, a polling aggregation website, who correctly forecast the winner of the 2012 presidential race between Barack Obama and Mitt Romney in all 50 states and the District of Columbia, as well as 31 of 33 U.S. Senate races, in The New York Times. “But sports can be far more complex than presidential elections,” Ablowitz said.