Giant Killers: South 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 higher; 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. Take a look at the upset chances for every game in every region, and check back with us on Wednesday when we look ahead to future rounds.

South Region

The South Region may lack upset depth, but it does feature our top Giant-Killing bet of the entire opening round.

BEST BET

No. 6 UCLA Bruins (53.1 Giant rating on a scale of 0 to 100 percent chance of beating a generic Giant Killer) versus No. 11 Minnesota Golden Gophers (57.5 Giant Killer rating on a scale of 0 to 100 percent chance of beating a generic Giant)

Upset chance: 74 percent

Never in the history of Giant Killers has a game materialized in this way. On Wednesday, our model ranked UCLA as the Giant most vulnerable to a potential upset. Then, on Friday, it pegged Minnesota as the second-best likely GK, behind only VCU, which earned a 5-seed and thus can't be a GK. The result is an upset rating the likes of which we've never seen, with the Golden Gophers expected to win three-quarters of the time. And that's based on stats UCLA accumulated with Jordan Adams, who is now lost for the season with a broken foot.

Even subjective analysts latched on this game immediately on Sunday, targeting Minnesota's rebounding prowess and UCLA's weakness inside. But they didn't know the extent to which that stat will matter; our model shows that offensive rebounding is a consistently strong trait among successful GKs and a standard weakness for slain Giants. The Gophers, of course, lead the nation in offensive boards (44.3 percent). UCLA is just 227th (30.2 percent). Good luck keeping Trevor Mbakwe and Rodney Williams off the glass.

Oddly, the Adams injury could inadvertently force UCLA to match up better with the Gophers. The Bruins went small after a slow start this season -- moving Kyle Anderson to the 4 -- to improve their suspect man-to-man defense and spread the floor on offense. Now, they have essentially no backcourt depth, and it might make sense to play two bigs most of the time with Anderson at the 3 and Shabazz Muhammad at the 2. But in the Pac-12 final, Ben Howland opted to stay small for the most part. We'll see if he changes his mind.

Losing Adams creates two other problems for UCLA, though. He had the highest steal rate on the team, another key factor for Giants. And he had the most made 3-pointers on a team that struggles from deep. That's rough for UCLA, especially considering the Gophers also force plenty of steals.

It is somewhat scary that this game has become such a popular upset pick. But our model is oblivious to such feeble human woes, and loves this matchup more than any other game on the board.

NOT COMPLETELY CRAZY

No. 5 Virginia Commonwealth Rams (83.0) versus No. 12 Akron Zips (27.8)

Upset chance: 28.9 percent

It's odd enough for VCU to be staring at this end of a GK matchup. But this game would have ended up with "Worth a Long Look" status had Akron not lost point guard Alex Abreu to suspension, a critical absence against VCU's press. Akron's stats, of course, reflect Abreu's time in the lineup, and they were already shaky with the ball (turnovers on 20.4 percent of possessions). That doesn't bode well against a team that leads the nation in steal percentage (17.1) in the way Dennis Rodman leads the nation in "questionable road trips."

So why do the Zips still have a decent chance to win? They are fierce offensive rebounders, grabbing their own misses 37.9 percent of the time (20th in the country). They launch a lot of 3-pointers, ranking 71st in 3PA/FGA. They force an above-average number of steals. And from a matchup perspective, VCU should struggle to deal with 7-foot Zeke Marshall inside.

Our model has lusted after the Rams for years, but it knows love is not blind. And it is willing to give the Zips a healthy chance to knock off the Rams. Are you willing to take that chance in your bracket, though?

No. 2 Georgetown Hoyas (74.9) versus No. 15 Florida Gulf Coast Eagles (11.1)

Upset chance: 13.4 percent

The Hoyas have lost to Giant Killers (Ohio, VCU and NC State) for three years running in the NCAA tourney. The chances of making it four this year aren't especially strong, at least in the round of 64, but the possibility is at least worth pondering.

Florida Gulf Coast is not a typical No. 15 seed, in that it beat Miami back in November. Granted, that's pretty much the Eagles' only meaningful win -- they got blown out by VCU and Duke and lost by 11 to Iowa State. But they force steals on 12.6 percent of possessions (23rd in the country), led by Bernard Thompson, who ranks sixth nationally in that category. And if you're going to bring one move to the dance, forcing steals ranks right up there with offensive rebounding and 3-point shooting for producing upsets. The Eagles are actually better than average on the offensive glass, too, so they are not to be taken lightly.

The bigger issue is Georgetown's vulnerability. The Hoyas' various offensive stats are mostly middling, despite Otto Porter Jr.'s all-around brilliance, which is why they rank 62nd in the country in efficiency. They don't hit the offensive glass at all (30.6 percent, 209th in the country), they're a low-frequency 3-point shooting team and they turn the ball over too much (20.1 percent of possessions). Their defense, of course, keeps them in games against anyone, and they are especially strong on the perimeter, both in forcing steals and denying 3s. But it will be tough for the Hoyas to run away from anyone, which is why the Eagles can at least dream about an upset.

No. 4 Michigan Wolverines (83.8) versus No. 13 South Dakota State Jackrabbits (14.2)

Upset chance: 11.5 percent

Forgive us for getting nostalgic, but this game conjures up memories. One specific memory, in fact. Back in 2009, VCU was a trendy upset pick against UCLA. The Rams were two years removed from stunning Duke as an 11-seed and UCLA hadn't distinguished itself. But our model determined that the Bruins were exceptionally safe and warned readers to go against the grain and stick with the higher seed. UCLA prevailed.

So here comes South Dakota State, a sexy pick with a funky mascot and a stellar point guard in Nate Wolters. And there sits Michigan, which has fallen from No. 1 in the nation to No. 4 in this region in about a month's time. So it's not surprising to see the Jackrabbits' bandwagon filling up.

Thing is, that's an overreaction that isn't supported by our metrics. The Jackrabbits do some things very well, ranking in the top 10 in taking care of the ball and defensive rebound percentage. But neither of those stats correlates with GK success. The one that does, among their arsenal, is 3-point shooting. Wolters can stroke it. So can big man Jordan Dykstra and 6-foot-6 Chad White. Overall, SDSU ranked 10th in the country in 3-point percentage at 39.4, and 36.4 percent of their attempts came from beyond the arc.

However, the team struggles greatly in every other key Giant-Killing area. They don't force steals. They don't grab offensive rebounds. And they rank just 209th in defensive efficiency. If you don't think a John Beilein squad will be poised to take advantage of that, you need to read up on the man.

His Michigan squad is an enigma, in that it has top-notch talent, but has struggled down the stretch, losing six of its past 12 games. But our model loves the fact that the Wolverines turn the ball over less frequently than any team in the country, shoot it well (and fairly often) from deep and played a schedule that featured tough conference foes and plenty of GKs outside of the Big Ten.

So can the Jackrabbits win? Sure, they can. But is that likely? No, and considering the popularity of this upset, you could gain ground here by sticking with the chalk.

STAY AWAY

No. 1 Kansas Jayhawks (83.9) versus No. 16 Western Kentucky Hilltoppers (7.9)

Upset chance: 5.2 percent

Here at GK Central, we will always have love for this creature, who apparently can be booked for an event. In fact, we might invite Big Red over to unveil this year's All-GK team, given its strong WKU lineage with the great Orlando Mendez-Valdez.

Sadly, this year's Hilltoppers don't match their predecessors' on-court performance -- they rank 177th in offensive efficiency and 182nd on defense. At least their mediocrity is balanced. They do grab 34.8 percent of available offensive rebounds and shoot a lot of 3s (36.4 percent of their shots). But they turn the ball over 22.4 percent of the time, don't force steals and generally can't get stops consistently.

Kansas, against a better team, might be vulnerable, given the Jayhawks' overreliance on free throws, low-frequency 3-point shooting and struggles to force turnovers. Our model pegs their base power rating as seventh in the country (read: not worthy of a No. 1 seed). But this won't be the game to expose the Jayhawks' flaws.

No. 3 Florida Gators (89.3) versus Northwestern State Demons (8.5)

Upset chance: 3.3 percent

Florida has been the king of tempo-free stats all season, and our model isn't about to give up now just because of a few close losses. Strong Giant Killers would have trouble cracking the Gators' armor, given their top-five efficiency ratings at both ends, their prowess scoring from behind the arc and keeping teams from doing the same and their strength on both sides of the boards. And the Demons aren't a good Giant Killer.

Northwestern State is a horrific 3-point shooting team and even worse on the defensive glass (37th in the nation). The Demons do force 12.7 steals per game, but that's not much of an issue for Florida's guards. Maybe someone will poke holes in Florida's super-stat style, but that team is not Northwestern State.