Picking favorites in the NCAA tournament is the smart play, but it's the upsets and Cinderella teams that make it memorable ... and help you get the edge you need to win your bracket pool.
As we decipher which upset picks look like the smartest plays in Round 1, we are going to focus on the true underdogs — a No. 9 seed beating the No. 8 seed in the first round is technically an upset, but those winners at least three seeds below the favorite they are facing is where the true value lies.
One upset we won't be picking is a No. 16 seed over a No. 1 seed. Since 1985, when the tournament expanded to 64 teams, the top four teams are 128-0 in the first round. It is tempting to knock off one of the No. 5 seeds early, since they, historically, have been the most likely low seeds to go home sooner than expected, but the No. 10 and No. 11 seeds offer more value this year.
The teams best built to topple favorites are the ones that have a high scoring margin while creating extra possessions through turnovers and offensive rebounds. Here are four upsets you can bet your bracket on, with odds based on point differentials using Jeff Sagarin's Predictor metric, one of the best indicators of future wins this season.
No. 12 Middle Tennessee over No. 5 Minnesota
Odds of an upset: 40.3 percent
Middle Tennessee won its second straight Conference USA Tournament championship Saturday night and are one of the best transition teams in the country, scoring 1.18 points per possession off the break (top 4 percent in the nation).
Junior guard Giddy Potts can not only run the team in transition, he has a lethal no-dribble jumper (41 for 90, 68.3 effective field goal percentage) and a deft touch around the basket (1.4 points per possession, not including post-ups).
The Blue Raiders also boasted a plus-108 turnover differential, 15th highest in the country.
No. 10 Marquette over No. 7 South Carolina
Odds of an upset: 48.3 percent
Marquette doesn't have an experienced roster — only Katin Reinhardt, a graduate transfer, has played in the tournament before — but third-year coach Steve Wojciechowski has built his team around strong shooters, producing the sixth-highest effective field goal percentage in the country. Two of the teams ahead of Marquette are No. 1 seeds (Gonzaga and Villanova) while the other two top seeds, Kansas and North Carolina, are far behind.
Freshman guard Markus Howard is perhaps the team's best shooter (65.8 eFG percent), hitting 54.9 percent of his 4.8 three-point shots per game leading up to the tournament. Junior guard Andrew Rowsey, a transfer from UNC-Asheville who sat out the 2015-16 season due to NCAA rules, can also shoot, but is at his best handling the ball on the pick and roll (1.06 points per possession, top 5 percent in the nation).
No. 10 Oklahoma State over No. 7 Michigan
Odds of an upset: 48.3 percent
Oklahoma State led the nation in offense (124.8 points per 100 possessions) with Jawun Evans, Jeffrey Carroll and Phil Forte all averaging over 10 points per game (19.0, 17.4 and 13.3 points, respectively). Carroll is also aggressive on the offensive glass, grabbing almost three offensive rebounds per game and scoring 63 percent of the time on those putbacks. Michigan, meanwhile, is one of the worst offensive rebounding teams this season (25.3 percent, 294th out of 351 teams), though that may help keep the Cowboys out of transition.
The Cowboys aren't strong defensively (102.7 points allowed per 100 possessions) but their 20.8 defensive turnover rate is 46th in the country, allowing them to create extra possessions for their high-octane offense.
No. 11 Rhode Island over No. 6 Creighton
Odds of an upset: 35.9 percent
Rhode Island can score (110.7 points per 100 possessions, 64th in the country), rebound (33.1 offensive rebound percentage, 60th) and limit turnovers (16.6 percent, 52nd). It can defend, too, and hold opponents to the ninth lowest effective field goal percentage this season (44.7 percent), with a mere 29 percent success rate allowed from beyond the three-point line. The Rams also limit the amount of three-pointers their opponents take, keeping it to less than a third of all field goal attempts (29.8 percent, 18th in the country). That will be a problem for Creighton, who relies on its 40.1 percent shooting from long rage.