March Mathness: How to improve your bracket and avoid common mistakes
The probability of creating the perfect bracket is one in 9.2 quintillion
The probability of creating the perfect bracket is one in 9.2 quintillion
The probability of creating the perfect bracket is one in 9.2 quintillion
Whether you’re cramming team stats, crunching numbers or studying the odds, building the perfect bracket for the NCAA Division I men’s basketball tournament is nearly impossible – a one in 9.2 quintillion chance, to be exact.
“If you were to create one billion brackets per second and never repeat it, it would take you 300 million years to create the number of brackets that are possible for one March Madness,” said Dr. Tim Chartier.
Last year, ESPN collected 22.6 million completed brackets before the tournament began. That’s more than 27,555 brackets per minute from the moment the challenge opened.
The NCAA Division I men’s basketball tournament, known as March Madness, features 68 college teams in a single-elimination competition. The competition began in 1939 with just eight teams and expanded to a 64-team structure in 1985.
Chartier, a professor of mathematics and computer science at Davidson College, specializes in data and sports analytics, particularly ranking systems. He teaches students how to develop mathematical formulas and models to predict winners using bracketology and ranking methods.
Last year, Chartier picked No. 1 UConn, No. 1 Purdue, No. 2 Arizona, and No. 8 Nebraska as his Final Four. Half of his picks were correct. UConn and Purdue reached the championship game, with UConn ultimately winning the tournament, another prediction he got right.
Chartier also predicted that No. 1 seeds North Carolina and Houston would face difficult paths to the championship. Both teams lost in the Sweet 16.
Although creating a perfect bracket is unlikely to occur, Chartier shared some tips to help you build a better bracket than your friends, coworkers or group.
Focus on your Final Four selection
Chartier recommends prioritizing your Final Four picks first. Because most bracket contests award increasing amounts of points as the tournament advances, predicting the last four teams standing is the best path to boost your ranking.
And don't be too worried about missing that Cinderella team. Even if an unexpected team reaches the Elite Eight or Final Four, your percentile ranking won’t shift significantly.
Chartier said that's because so few brackets accurately predict these unlikely runs.
“People were not looking for them to win, so in some cases, it’s largely alumnus of the school picking that team because nobody else was really anticipating it,” Chartier said.
Last year, NC State shocked bracket makers by becoming the sixth No. 11 seed to reach the Final Four.
Seed pairing upsets
The 5-12 seed matchup is considered a prime opportunity for an upset in the men's tournament. Chartier and his students devote much of their time to analyzing which No. 12 seed is most likely to pull off an upset.
Last year, he correctly predicted that Grand Canyon, a No. 12 seed, would defeat their No. 5 opponent.
However, an analysis by the Hearst Television Data team found that the 6-11 pairing has seen more upsets than the 5-12 in the tournament's round of 64.
Of the last 156 games between the No. 6 and No. 11 seeds, the lower-seeded teams have won 61 times, six more upsets than the No. 12 defeating a No. 5.
The 1989 NCAA tournament was especially notable for No. 11 seeds, as all of them won their first-round matchups against No. 6 opponents.
Pay attention to late season performance
Chartier also recommends paying attention to how a team performs towards the end of the season. The end of the season is typically when teams begin to play their conference games. He considers these games crucial since teams are competing at their full ability against relatively evenly matched opponents.
Do your own research
While Chartier relies on mathematical equations and formulas to predict the winners, he still recommends doing your research. His methods don’t account for localized ranking, meaning he doesn’t detect when a strong team consistently struggles with a particular opponent.
“That’s part of what makes it fun, you can actually dig in yourself, create your own bracket and then just let things unfold and inherently you’re going to make some wrong decisions,” he said.
While No. 1 seeds have predominantly won the tournament, lower-ranked teams have made headlines with surprising victories. In 2014, UConn won the championship as a No. 7 seed.
The lowest-seeded team to ever win the tournament was No. 8 Villanova in 1985. No fifth seed has won the NCAA men’s basketball championship since the tournament expanded to 64 teams in 1985.
Most common mistake
And, the biggest mistake people make when filling out their bracket? Letting team allegiance cloud their judgment.
“I work with math students, and sometimes they feel so strongly about their allegiance to certain teams that they will actually change their underlying mathematics,” Chartier said.