Giant Killers: Midwest upset picks

For each of the four regions, we'll break down the six games with Giant Killers implications (reminder: A Giant Killer is a team that beats an opponent seeded at least five spots better; check out our full methodology for more details).

We have also sorted the games into four categories -- Best Bets, Worth a Long Look, Not Completely Crazy and Stay Away. It's important to remember that we're dealing with probabilities here. No upset is a given -- if it were, it wouldn't be an upset in the first place. And even a game with a 10 percent chance of going in favor of the underdog means that one out of every 10 times, the underdog should win.

Instead, we can give you the likelihood of an upset, based on rigorous statistical analysis that merges a team's base power rating with our GK "Secret Sauce" -- the statistical similarity between potential Giant Killers and historical Davids, as well as Giants and their predecessors who were slain. By coming up with a rating for every team, we are then able to compute an upset percentage for each matchup.

What you choose to do with that information is up to you, and should probably depend on how many points you receive for picking an upset in your pool. But keep checking back with us as we hit the Midwest and West Monday, the East and South on Tuesday and look ahead to future rounds on Wednesday.

Midwest Region

The Midwest Region is stacked. You've read about it, listened to talking heads scream about it and maybe even tweeted about it. And it's true. As we mentioned last night, seven of the top 19 teams in the BPI somehow landed in this region.

But even with so many talented squads at the top, upset opportunities abound, starting with our model's second-likeliest pick for the opening round.

BEST BET

No. 6 Memphis Tigers (59.4 Vulnerability Rating on a 100-point scale) vs. No. 11 Saint Mary's Gaels (51.1 Giant Killer Rating on a 100-point scale)

Upset chance: 67.5 percent

There's something about Saint Mary's. The Gaels may not have shown it in three losses to Gonzaga, but they carry all sorts of traits that are historical markers of Giant Killers (to say nothing of guard Matthew Dellavedova, who actually is a historical Giant Killer).

Chief among them are both outstanding offensive rebounding (37.2 percent of available misses, 27th in the country) and high-frequency, accurate 3-point shooting. Those qualities contribute to the country's 11th-most efficient offense, which in turn leads to another key factor our model recognizes: The Gaels are simply better than their seed. In fact, their base power rating is 21st in the country -- 22 spots ahead of Memphis. Chew on that for a second, then stamp them onto the next line of your bracket.

If that's not enough to convince you, maybe this will: The Tigers don't just have a low power rating, they also have some disturbing similarities with historically slain Giants. In particular, they take a lot of free throws and they don't make them (66.6 percent shooting). Plus, guards Joe Jackson and Geron Johnson throw the ball all over the place, leading to turnovers on 20.8 percent of the Tigers' possessions.

WORTH A LONG LOOK

No. 5 Oklahoma State Cowboys (80.9) vs. No. 12 Oregon Ducks (20.0)

Upset chance: 31 percent

By now, you've familiar with Oregon's surprising slide to a 12-seed. And if you go by the BPI, the complaints are valid: The Ducks rank 33rd by that advanced metric. But the Ducks still have a chance to show they're all they're quacked up to be thanks to a solid GK rating. Oregon forces steals on 12.1 percent of possessions (33rd in the country) and grab offensive boards on 36 percent of their misses (46th). Our model would like the Ducks even more if they let it fly a little more: Just 21 percent of their points come from downtown.

What's odd about this matchup is that Oklahoma State has strengths that counter many of the Ducks' best weapons. The Cowboys don't turn it over much (Marcus Smart gives it away some, but Markel Brown has a steady handle and reserve guard Phil Forte is second in the country in turnover rate) and they are better than average on the offensive glass. The Cowboys are too reliant on free throws, though (a stat our model has consistently targeted as a critical weakness for a high seed) and they don't shoot well (or often) from 3-point range. The Cowboys rank only 56th in offensive efficiency, which means Oregon should be able to hang around. And in March, we know that can mean for an underdog.

NOT COMPLETELY CRAZY

No. 3 Michigan State Spartans (87.1) vs. Valparaiso Crusaders (24.7)

Upset chance: 19.3 percent

Bryce Drew! Bryce Drew! Have you heard about Bryce Drew? Sadly, our model awards no points for coaches who have hit buzzer-beaters in the tourney. But this Valpo squad has a bit of magic to it regardless. Did you know that the Crusaders are fourth in the country in effective field goal percentage? That 6-foot-7 senior Ryan Broekhoff grabs 7.3 rebounds per game and makes 2.6 3s a night on 43.2 percent shooting? That Valpo keeps folks off the offensive glass (72.7 percent defensive rebounding rate)?

But the biggest news about the Crusaders, from a GK perspective, is their 3-point shooting. They hit 37.5 percent of their trifectas and 39.8 percent of their attempts come from deep. Defensively, they have issues, though, especially because they don't force many steals, which would help against the turnover-prone Spartans. That's the key place to get Michigan State, which is typically excellent on the glass and forces plenty of steals at the other end. But the Spartans don't chase foes off the arc -- opponents take 33.8 percent of their shots from deep. So Valpo will at least get enough looks to have a proverbial shooter's chance at an upset.

No. 4 Saint Louis Billikens (71.3) vs. New Mexico State Aggies (7.3)

Upset chance: 18.7 percent

The Billikens are hot and are generating deserved Final Four buzz. But they're one of those squads that is a better team than a Giant. Why? Because their particular weaknesses are the trademark of those who get burned. The biggest concern, by far, is a 28.1 offensive rebounding rate, which is just 287th in the country. Safe Giants tend to punish GKs in the paint; Rob Loe and Cody Ellis would rather shoot 3s than hit the glass.

New Mexico State is actually built to exploit Saint Louis in that area. The Aggies are 38th in the country with a 36.4 percent offensive rebounding rate. But they have no other weapons of note. They don't force steals, are one of the worst 3-point shooting teams in the country (17.6 percent of their total points, 340th) and turn the ball over constantly on offense.

If our model had hands, it wouldn't slap you across the face for picking the Aggies. But it certainly wouldn't slap you a high-five, either.

STAY AWAY

No. 2 Duke Blue Devils (68.3) vs. Albany Great Danes (5.8)

Upset chance: 6.6 percent

As we wrote over the weekend, Duke's Giant rating is significantly better with Ryan Kelly (82.2). But regardless of which number you use, this is a game to avoid.

It's not that Duke is bulletproof. We continue to harp on Duke's poor offensive rebounding, which is exacerbated by a tiny perimeter. And the Blue Devils don't force turnovers at a typical rate for a Mike Krzyzewski team, ranking just slightly above average.

But Albany does little that should cause Duke to worry. The Great Danes don't excel in any of three areas that are so critical for GK success -- 3-point shooting, steals and offensive rebounding. They also don't guard the arc very well (a disaster against Duke's bevy of 3-point shooters) and turn the ball over on 21.2 percent of possessions. In other words, they don't bring much dog to this fight.

No. 1 Louisville Cardinals (95.4) vs. No. 16 North Carolina A&T Aggies (2.5)

Upset chance: 1.0 percent

The second-safest Giant against the 317th-ranked offense (NC A&T) delivers exactly what you'd expect: Essentially no chance for an upset. We'd write a little more about why this is so lopsided, but you have other regions to study, gutters to clean and teeth to floss. Move right along.