Commentary

Sowing the seeds for upsets

Updated: March 12, 2012, 6:12 PM ET
By AJ Mass | ESPN.com

Which team will be this year's VCU? How can we tell if a sleeper is about to awaken?

The most important round in your NCAA tournament bracket is the one that takes place on the first Thursday and Friday of the Big Dance. Because there are 32 games, there will always be a few upsets that crop up. If you pick the right lower seeds to advance -- and perhaps more importantly, ones that the rest of your pool fails to select -- while at the same time avoiding any incorrect predictions of demise of the eventual Final Four teams, you have a huge leg up on the rest of the competition.

Alas, basketball is not an exact science, and Valparaiso, Cleveland State and Austin Peay didn't exactly leap off the page when the brackets were printed in the years of their upsets. Few people, if any, put much faith in the George Mason and Butler teams that made their marches to the Final Four. Still, there must be a way to red-flag certain teams that may be primed for an unexpected thrashing, and those that are anxiously jumping at the chance to give one. And there is … it's all in the seeds.

Simply put, despite all the preparation that goes into the process, the seeding of the NCAA tournament often ends up looking like a hot mess. Many times, the selection committee severely underestimates or overestimates teams. This is especially true when it needs to wait until the very last minute just to figure out which teams will be in the field, which happened Sunday with St. Bonaventure playing in the Atlantic 10 title game, keeping the fate of many a bubble team in limbo well into the afternoon.

In addition, the committee has a history of favoring teams from the power conferences at the expense of those from the mid-majors and below. That's why it takes the Gonzagas of the world several years of making the tournament before they get assigned anything higher than a 7-seed, even though they probably deserved far higher much sooner. A team like Murray State, which is currently ranked ninth in the ESPN/USA Today Coaches Poll, still was given only a 6-seed. If the name on the front of the jerseys reads D-U-K-E for a team with the same resume, that team probably will be staring at no worse than a 2-seed.

But that's OK. It is exactly instances such as these -- when a 13-seed probably deserves to be an 8-seed, or when a 6-seed looks an awful lot like an 11-seed -- in which the upsets occur. Identifying these irregularities in seeding is the key to filling out a successful bracket.

So as I do every year at this time, I've taken all the teams that were invited to the NCAA tournament and ranked them according to a statistical formula that takes into account the very same categories that the committee says that it uses to select the at-large teams in the first place. These categories include things like strength of overall schedule, record in games played on the road and at neutral sites, strength of the out-of-conference schedule and wins against teams in the RPI top 50.

By comparing each team's strength in these areas in relation to the other teams in the tournament, we can assign each school a single value, which I call a team's Atomic Mass. The larger the number, the "heavier" a team is, the more it sinks to the bottom of the list and the closer to 16 we should expect its seed to be.

The teams listed in green below are our "money teams": teams our evaluation process has determined "should have been" seeded at least three spots higher than they were by the committee. Conversely, the teams listed in red are the "teams on red alert": teams that should have received seeds at least three spots lower than they were given by the committee.

If any upsets happen, they likely will be in the games that involve these teams. This doesn't mean we're actually going to pick all the money teams to win, nor will we necessarily go with all the teams on alert to lose. South Dakota State may well be more worthy of a 10-seed than a 14-seed, but I'm still not picking it to beat Baylor, for example.

Keep in mind that we're in no way suggesting that the committee should have seeded Iona ahead of Michigan. All we are saying is that the committee uses different criteria to decide that Team A deserves to go dancing over Team B than they do to determine where to place the teams in the bracket, and as a result certain "favorite" teams end up going head to head against teams they really aren't "better than," according to the very qualities that were responsible for getting them invited to the tournament in the first place. That's where Cinderella is born.

Last year's Dance saw only 10 of the 16 top seeds make the Sweet 16 and not a single No. 1 seed made it to the Final Four. Our system identified seven potential second-round upsets. For example, it successfully predicted 12-seed Richmond over Vanderbilt. But our pick of 13th-seeded Oakland to upset Texas fell short by only four points, and our call for a Utah State upset of 5-seed Kansas State fell five points short. Of course, neither the Longhorns nor K-State made it to the Sweet 16, so by taking a leap of faith and going against these favorites, our brackets were not hurt in the long term. Additionally, we predicted that Princeton would make John Calipari sweat, and the Tigers lost by only two points to Kentucky.

Now, in the interest of full disclosure, our system also really liked Old Dominion over Butler, and while we fell only two points short in getting that one right, the Bulldogs' second consecutive march to the title game certainly left our bracket out of the running for any potential glory. But hey, no system is perfect. And even if we fully stand behind the claim that we've correctly identified the most likely candidates for an early upset, there is no guarantee that all of them will occur, or that favorites not on our list won't go by the wayside. That's what makes the madness of March so marvelous.

With that said, let's look to the 2012 bracket. This year, our careful examination of all of the second- and third-round matchups has given us the following possible upsets you might want to consider including on your sheets of integrity:

East region

Syracuse probably has drawn the most difficult of the 16-seeds in UNC Asheville, but since there's yet to be a No. 1 seed to go out in its first contest of the tourney, we're not about to make that call here by any means. Cincinnati, which did beat Syracuse in the Big East tournament, is a bit overrated by the committee, but so too is its opponent, Texas. As such, no upset will be predicted there. Montana should keep things close with Wisconsin, but the Badgers should be able to hold the Grizzlies off. As for actual lower-seed victories, we'll pick two games here:

No. 9 Southern Miss over No. 8 Kansas State: K-State has limped into the tournament, winning just half of its past 12 games. Southern Miss is a strong rebounding team that can compete on the boards with the Wildcats, and as long as LaShay Page makes his shots USM should advance in this one.

No. 12 Harvard over No. 5 Vanderbilt: Yes, Vanderbilt just beat seemingly invincible Kentucky in the SEC championship game. However, we're always a bit worried when a team is coming off such an emotional high. Add to the mix that there always seems to be a No. 12 that does indeed shock the world and the fact that Harvard is one of the top defensive teams in the NCAA, and we're going Crimson.

Midwest region

Although we do feel that both the Michigan Wolverines and the San Diego State Aztecs got a little too much benefit of the doubt in terms of their seeding, both should survive their initial contests. We also like Temple to hold off whichever team comes out of the first-round battle between Cal and USF, but if either does pull off the upset, we may have to start giving a little more weight to teams that gain a bit of momentum by playing in Dayton, Ohio, in future years. Other than that there is only one game that calls out to us for an upset here:

No. 14 Belmont over No. 3 Georgetown: Belmont opened the season with a one-point loss to Duke at Cameron Indoor Stadium. This is a team that averages 81.5 points per game. It is not going to be intimidated by the Hoyas by any stretch of the imagination. Plus, Georgetown is only 8-7 away from home this season. The Bruins are 14-6.

West region

The Atomic Mass really favors Memphis highly and we would not be surprised at all to see it in the Sweet 16. Sorry, Michigan State. It also indicates that Murray State is certainly legit enough to defeat Colorado State. Neither Florida nor Virginia impresses us much, but in that scenario we'll go with the higher seed and advance the Gators. If there is going to be one region that blows up, it's likely to be this one, led by the two upsets we're predicting in the upper half of the bracket.

No. 12 Long Beach State over No. 5 New Mexico: We'd feel stronger about this pick if we were completely confident that Larry Anderson, who missed the Big West tournament with a strained right knee, was at 100 percent. However, if he does suit up, having such a strong defender on the court could be the difference in what might be one of the most enjoyable contests of the second round. It's a shame the committee put these two teams up against each other, because both could have made deep runs had they been placed differently.

No. 13 Davidson over No. 4 Louisville: Louisville, another team that may be suffering from conference tournament hangover, has certainly one of the best defenses in the game today. But Davidson has five guys who average double-digit scoring and five players who shoot 34 percent or better from behind the 3-point line. There's not a defense in the world that can beat a team like that if it gets hot. Remember, this Davidson team beat Kansas in December and lost to Vanderbilt by only four.

South region

Mostly chalk for this region. Although New Mexico State and South Dakota State should be able to keep their games competitive well into the second half, we do think higher seeds Indiana and Baylor will ultimately prevail in those two games. Another disappointing matchup of two solid teams can be found in Wichita State and VCU; how we wish the committee didn't pit mid-majors against each other as much as they do. We're predicting only one "upset" here, and it's in the coin toss 8-9 affair.

No. 9 Connecticut over No. 8 Iowa State: The Cyclones went 6-8 away from home this season, while the Huskies were 4-2 on neutral turf. No, this is not the same squad that ran the table in last year's Dance, but some of the UConn starters saw firsthand what it takes to win this tournament. We'll put our faith in Jim Calhoun, at least until Kentucky is on the floor against his team.

So there you have it: six games in which the numbers say you should take a chance on the underdog, courtesy of the committee's rushed seeding process. By this time next week, we'll know for sure if the process produced results again in 2012, but either way we're going to have lots of fun seeing how this madness plays itself out.

AJ Mass is a fantasy baseball, football and college basketball analyst for ESPN.com. His book, "How Fantasy Sports Explains the World" is available for purchase here.You can e-mail him here.

Follow AJ Mass on Twitter:

ALSO SEE