Following popular upset picks while filling out your NCAA Tournament bracket has its pitfalls.

Predictive models like the ones over at FiveThirtyEight.com suggest most lower seeds have worse-than-even odds of advancing. That means that picking the same underdogs as many or most other players in your office pool gives you, at best, losing odds of merely keeping pace with everyone else.

Instead, you want to pick the upsets that are reasonably likely but undervalued by your competition. These games give you the best opportunities to actually pick up points by zigging when most other players zag.

With that in mind, here’s a look at the Post-Gazette’s third annual over and undervalued upset picks, which measures user trends in Yahoo! Sports’ bracket game against FiveThirtyEight.com’s metrics.

Advertise with us

Overvalued

● No. 7 Nevada vs. No. 10 Florida. Yahoo! users: Florida 52 percent. FiveThirtyEight odds: Nevada 58 percent.

Stay far away. The Gators are drawing more action than any team seeded 10th or lower, but they’re still well less than a 50/50 bet. For that reason, there’s very little upside here. Rather, Nevada is likely undervalued as a favorite. The Wolfpack aren’t in a power conference and they play their games late at night in the East. That reality is probably leading a lot of players to go with the known commodity in Florida, even if it did finish a very mediocre 19-15 and a lousy 2-9 against Associated Press Top 25 competition.

● No. 7 Wofford vs. No. 10 Seton Hall. Yahoo! users: 50/50. FiveThirtyEight odds: Wofford 63 percent. 

It’s not often you see a team from the Southern Conference seeded this high, so it’d be understandable if you’re hesitant to trust the Terriers. Don’t be. The analytics love this favorite. Kenpom.com ranks Wofford No. 19 in the entire tournament, suggesting even that seventh seed is a little low. And although they played a pretty soft schedule, they dominated it. Their points for/points against average spread is a spectacular 17.4 points. The Pirates, meanwhile, weren’t anything special in the Big East. The Terriers’ 0-3 record against Top 25 teams might make them a tough bet to go far, but they shouldn’t have any problem disappointing that half of players piling onto Seton Hall.

Advertise with us

● No. 5 Marquette vs. No. 12 Murray State. Yahoo! users: Marquette 61 percent. FiveThirtyEight odds: Marquette 64 percent.

The Racers’ action numbers compared to their odds aren’t dramatically out of whack. So if you have to have that trendy 12 over 5 upset somewhere on your bracket, you could certainly do worse. The problem is that a significant chunk of players seem to know that, too. The Racers’ Ja Morant is a potential lottery pick in the NBA draft, so this team really isn’t going to get a chance to fly under the radar. This diminishes the potential returns on eating those roughly 1-in-3 odds of a Murray State win. Pick this underdog only if you believe it’s truly the better team, and not for the sake of it.

Undervalued

● No. 6 Maryland vs. No. 11 Belmont.* Yahoo! users: Maryland 81 percent. FiveThirtyEight odds: Maryland 62 percent. 

The asterisk is a warning because the Bruins have to get past Temple in a play-in game Tuesday night to get a shot at the Terrapins. If they do, though, the numbers say they’ll be a strong 11 seed that many people seem to be overlooking because they’re too impatient to see who emerges from the matchup with the Owls. Maryland, meanwhile, is one of the youngest teams in the field, making it potentially vulnerable to a Belmont group led by experienced senior guards Dylan Windler and Kevin McClain. For that reason, you’ll want to pay attention. If the Bruins are still standing Wednesday, they could offer a ton of value compared to the upset picks above, simply because many people may have locked in their brackets with Maryland before the round of 64 is even set.

Advertise with us

● No. 4 Florida State vs. No. 13 Vermont. Yahoo! users: Florida State 90 percent. FiveThirtyEight odds: Florida State 79 percent.

Any time you can get a No. 13 seed with 1-in-5 odds to win, you’ve got to give it some thought. UC-Irvine falls in that bucket this year, too, but 17 percent of users are picking the Anteaters against Kansas State, meaning they have less potential to provide value than do the Catamounts, whom almost no one is picking. Beating Florida State won’t be easy. The Seminoles fell just one game short of the Final Four last year and have a winning record against ranked opponents this season. Vermont is cut out of the same classic Cinderella mold that Belmont is, though. It has an experienced rotation led by junior forward Anthony Lamb, who averages 21.4 points per game and boasts a field goal percentage of 52.1 percent. And hey. This Florida State team did slip up a couple of months ago at Pitt, which struggled for most of the rest of the year. The Catamounts aren’t a pick for the faint of heart, but they’ve got a puncher’s chance here.

● No. 8 Syracuse vs. No. 9 Baylor. Yahoo! users: Syracuse 66 percent. FiveThirtyEight odds: Syracuse 60 percent. 

There seems to be one No. 8 versus No. 9 game each year where the user numbers get out of whack and cause what should be a relatively even matchup to become a true scoring opportunity anyway. This year, the Bears are getting way less love than they probably deserve. To be sure, they have the worst FiveThirtyEight odds of any of the No. 9 seeds in the tournament. But they also have the second-best KenPom ranking in the group of four at No. 41, which puts them just six spots behind the Orange. ESPN’s Basketball Power Index has Syracuse as a 51.8 to 48.2 percent favorite, and Vegas has installed the Orange as just a slight two-point favorite. That’s a lot of data suggesting this is a more even matchup than popular opinion dictates. That’s why you should swim against the tide here, unless you really like Syracuse.

Adam Bittner: abittner@post-gazette.com and Twitter @fugimaster24.