No. 12 Stephen F. Austin beat No. 5 Virginia Commonwealth in overtime during the 2014 NCAA college basketball tournament. (Denis Poroy/Associated Press)

There are few certainties in March Madness, but one story line that plays out almost every year is a No. 12 seeded team upsetting a No. 5 seeded team during the tournament’s round of 64. In fact, last year was the first since 2007 that a 12-seed didn’t pull off the upset.

“There is tremendous parity in college basketball,” said Tom O’Connor, a five-year member of the NCAA selection committee. “A lot of the basketball players on each team play against each other in the summer and there is some familiarity. It makes for some great games. Anything can happen in the NCAA tournament.”

Parity aside, there is something weird happening in that particular 5-12 bracket. Dating back to 1985, No. 5 seeds have been upset 44 times in 124 games but based on a line that best fits the data, they should be winning closer to 72.5 percent of these matchups, not at a rate lower than what we see from No. 6 seeds. In fact, the rate is so low we can estimate it has just a 1.1 percent chance of occurring.


“The five-seed is where you lose natural geographic-area protection, because only the top-four line receives that protection,” explained Greg Shaheen, former senior vice president of the NCAA and organizer of the NCAA Division I men’s basketball championship. “Once you go to the five-[seed], you could be playing those games anywhere.”

When No. 5 Oklahoma State lost to No. 12 Oregon in 2013, the game was played in San Jose, Calif. No. 12 Harvard upset No. 5 Cincinnati in 2014, a game played in Spokane, Wash. Perhaps having to go out out west played a factor, but the selection process also plays a role.

“That’s also where you start to see some of the second and perhaps third teams from the bigger conferences as well as the emergence of some of the so-called ‘mid majors’ that can make their way north in the field,” Shaheen said. “That could make [the No. 5 seed] vulnerable because they’re good, but don’t have the same protections that a three- or a four-seed has. It’s the true middle of the field.”

No. 5 Temple, who finished first in the Atlantic-10 conference, found this out the hard way in 2012, when it was upset by the fourth best team in the Big East, No. 12 South Florida.

Despite the frequency at which a No. 12 seed has beaten a No. 5 seed, there is no orchestration behind the scenes to either encourage or discourage these type of matchups. In fact, the selection committee doesn’t even know what the bracket will look like until after it is completed.

“Wherever the five-seeds fall, wherever the 12-seeds fall, they fall,” said Shaheen. “The committee doesn’t build a 5-12 matchup, the process does. When the committee builds the bracket, they don’t know who is playing who. They are assigning the [Nos.] 1s, then the 2s, then the 3s, in sequence. They don’t get to who they are playing until much later in the process, and that is geographically and policy predisposed. They are more mindful of protecting the top-four lines. Once they get to the five through 12 [seeds], they basically know that anything could happen.”

The field of 68 has been set, and The Post's Adam Kilgore and Mike Hume look at some of the higher seeds that could be vulnerable to an early tournament upset. (Thomas Johnson,Randolph Smith/The Washington Post)

There also aren’t any conspiracies to seed the so-called “mid majors” in these slots. In fact, according to O’Connor, the term “mid major” doesn’t even come up.

“You are just looking at them on what they have accomplished during the regular season as a basketball team,” O’Connor said. “You are just looking at their complete resume during the season and how they should or should not be in the tournament. They’ve earned the right to be in the tournament because they are a good basketball team.”

No matter how they got there, the No. 5-No. 12 bracket has featured so many upsets that some coaches don’t mind being the underdog at that spot.

“It’s a great draw,” said Brad Underwood, coach of the Stephen F. Austin Lumberjacks, who beat No. 5 Virginia Commonwealth University as a No. 12 seed in 2014. “There become some beliefs from the 12 [seed] that they can win because it has happened so often. So I think there is some media that helps the 12, and maybe puts more pressure on a five. What traditionally happens is when you get a five, you get a younger power conference team or a third-place team in their conference. In our case we were a veteran team — tremendous character, tremendous toughness — and we’ve been through all the battles and it all just played into it.”

The Lumberjacks are the 13th most experienced team in the tournament, with an average of 2.15 years experience. Their bench also plays the 18th highest percentage of minutes (34.6 percent) among teams that made the field of 68. That type of consistency and cohesiveness in the lineup makes them an intriguing pick to pull off another upset or two this season. According to projections on FiveThirtyEight, Stephen F. Austin, a 14-seed, has a six percent chance of making the Sweet 16, the highest of any seed 13 or worse.


 

You can also look at the regular-season success of the No. 12 to get an indication if they can move on to the round of 32.

“If that team has already won 25 or 26 games, they are used to winning and that becomes contagious,” said University of Richmond Coach Chris Mooney, who helped guide his team’s win over No. 5 Vanderbilt in 2011. “Then you have a locker room full of guys who believe they will win no matter who they play and they are ready to go.”

Teams that fit the mold this year as a 12-seed include Arkansas-Little Rock (29-4), Chattanooga (29-5) and South Dakota State (26-7). Of those, Arkansas-Little Rock is a team to watch.

The Trojans rank 47th in the 2016 Pomeroy College Basketball Ratings and will play No. 5 Purdue, whose coach, Matt Painter, would be classified as being “snake-bit” in the NCAA tournament, per a study of NCAA tournament coaches by Peter Tiernan of BracketScience.com. Tiernan found that coaches in this class are the worst performers in the tournament, barely winning more than 55 percent of their games.

“Nobody talks about No. 13 vs. four-seeds or the No. 6 vs. 11s, it’s always the No. 5 versus 12s and for whatever reason over the course of time the 12s have won so now it has become a very big deal. Other than the No. 1 seeds, it’s the game that’s the most talked about in the bracket.”

Here are the chances we see an upset in each of this year’s games where a No. 5 faces off against a No. 12, based on Pomeroy’s college basketball ratings.

No. 5 Maryland vs. No. 12 South Dakota State

In order for Maryland to advance, they must take care of the ball. Their turnover percentage (19.1) on offense ranks 243rd while their defensive rate (16.5 percent) ranks 275th. Allowing the underdog to get extra possessions is a harbinger of an upset, making it tough to completely rule out a South Dakota State victory.

Chance of an upset: 30.6 percent

No. 5 Baylor vs. No. 12 Yale

Yale’s defense, which ranks 21st in the nation in terms of efficiency (95.0 points allowed per 100 possessions), is bolstered by one of the most experienced teams in the tournament (2.27 years). That could help put a damper on Baylor’s shooting, which is barely above average (eFG% of 51.7 compared to 49.9 percent league average).

Baylor Bears Shot Charts Details Shot Analytics

Chance of an upset: 43.1 percent

No. 5 Purdue vs. No. 12 Little Rock

Another case where turnovers might make a huge difference in an early-round matchup. Little Rock keeps their turnovers down on offense (15.9 turnover percentage) while maximizing them on defense (21.4 percent, 20th best in nation). Purdue, meanwhile, is one of the worst teams at creating defensive turnovers (13.8 percent). Only University of Central Florida, Delaware and George Mason — all non-tournament teams — were worse.

Chance of an upset: 32.0 percent

No. 5 Indiana vs. No. 12 Chattanooga

Indiana can shoot, and hit 41.5 percent of their shots from long range.

Indiana Hoosiers Shot Charts Details Shot Analytics

That’s going to put pressure on Chattanooga’s interior defense, which allows 50.4 percent shooting. Still, Indiana Coach Tom Crean has just one tournament campaign out of eight that has gone past the Elite Eight, so it’s tough to say whether or not he can guide his team far in this year’s bracket.

Chance of an upset: 20.6 percent

More 2016 NCAA tournament coverage:

The perfect bracket to win your March Madness pool

Five Cinderellas who could reach the Sweet 16

Print: NCAA tourney bracket (PDF) | Graphics: Tournament campaign buttons

The most likely upsets for the first round

{_json_={"game":{"gameId":"fc7bff57-b177-4c12-8912-69d0fcaf2e63","hideScore":"1","tags":["Pomeroy College Basketball Ratings","Ken Pomeroy","ncaa","march madness","ncaa tournament"],"title":"What are Pomeroy college basketball ratings?","byline":"Neil Greenberg","articleReferences":[],"createdTimestamp":1457899550285,"lastUpdatedTimestamp":1458657166079,"archive":false,"allowMoreThanOnce":false,"questionIds":["079b534c-06f4-4361-a8df-48b71e1164f2"],"createdBy":"greenbergna","lastUpdatedBy":"hunsickerb","live":true,"photoUploaded":false,"shareUrl":"http://wapo.st/1QRCEqw","embedToggles":[{"headline":false,"byline":false}],"captchaProtected":false,"questions":[{"questionId":"079b534c-06f4-4361-a8df-48b71e1164f2","excludeFromTrivia":false,"createdDate":1457899550000,"lastUpdated":1458657166000,"questionTitle":"What are Pomeroy college basketball ratings?","multipleSelectionAmount":0,"noteOptions":[{"hasComment":false,"element":{"type":"IMAGE","id":"-1.0","src":"<img src=\"https://games-cdn.washingtonpost.com/images/prod/quiz/015273f4-41d1-4b21-a43f-96f9b4a923e9/bc40166a-ad20-4d70-bc44-48ef41a01864-edited.jpg?ts=1457899531032\" alt=\"NCAA Men's Championship Game - Kansas v Kentucky; Jeff Gross\" class=\"note-image\" /><div class=\"context-image-credit\"></div>","original":"https://games-cdn.washingtonpost.com/images/prod/quiz/015273f4-41d1-4b21-a43f-96f9b4a923e9/bc40166a-ad20-4d70-bc44-48ef41a01864-edited.jpg?ts=1457899531032","edited":"https://games-cdn.washingtonpost.com/images/prod/quiz/015273f4-41d1-4b21-a43f-96f9b4a923e9/bc40166a-ad20-4d70-bc44-48ef41a01864-edited.jpg?ts=1457899531032","credit":"NCAA Men's Championship Game - Kansas v Kentucky; Jeff Gross"}},{"hasComment":false,"element":{"type":"HTML_TEXT","id":"-1.0","src":"<p>The Pomeroy College Basketball Ratings are a series of predictive ratings of men's college basketball teams published by Ken Pomeroy. The ratings are based on the Pythagorean expectation, derived from (AdjO^11.5)/(AdjO^11.5+AdjD^11.5).</p><p><b>Adjusted offensive efficiency</b> - An estimate of the offensive efficiency (points scored per 100 possessions) a team would have against the average D-I defense.</p><p><b>Adjusted defensive efficiency</b> - An estimate of the defensive efficiency (points allowed per 100 possessions) a team would have against the average D-I offense.</p>","name":null}}]}],"allowDuplicate":false,"status":"ACTIVE"},"apiRoot":"https://notes.washingtonpost.com/notes/games/webapi","embed":"<div class=\"context-note\">\n <link href=\"https://games-cdn.washingtonpost.com/shared/bootstrap.min.css\" rel=\"stylesheet\">\n <link href=\"https://games-cdn.washingtonpost.com/shared/notes/notes.css\" rel=\"stylesheet\">\n <div class=\"context-header\">\n <div class=\"note-headline\">\n <div class=\"note-title\">What are Pomeroy college basketball ratings?</div>\n <p class=\"note-byline\">by Neil Greenberg</p>\n <p class=\"note-timestamp\">Updated Mar 22, 2016 at 10:32 AM</p>\n </div>\n </div>\n <div class=\"context-body\">\n <div class=\"notes-inner-content context-block\">\n <img src=\"https://games-cdn.washingtonpost.com/images/prod/quiz/015273f4-41d1-4b21-a43f-96f9b4a923e9/bc40166a-ad20-4d70-bc44-48ef41a01864-edited.jpg?ts=1457899531032\" alt=\"NCAA Men's Championship Game - Kansas v Kentucky; Jeff Gross\" class=\"note-image\" /><div class=\"context-image-credit\"></div>\n <p>The Pomeroy College Basketball Ratings are a series of predictive ratings of men's college basketball teams published by Ken Pomeroy. The ratings are based on the Pythagorean expectation, derived from (AdjO^11.5)/(AdjO^11.5+AdjD^11.5).</p><p><b>Adjusted offensive efficiency</b> - An estimate of the offensive efficiency (points scored per 100 possessions) a team would have against the average D-I defense.</p><p><b>Adjusted defensive efficiency</b> - An estimate of the defensive efficiency (points allowed per 100 possessions) a team would have against the average D-I offense.</p>\n </div>\n </div>\n</div>"}, game={gameId=fc7bff57-b177-4c12-8912-69d0fcaf2e63, hideScore=1, lastUpdatedBy=hunsickerb, lastUpdatedTimestamp=1458657166079, allowDuplicate=false, createdTimestamp=1457899550285, questions=[{excludeFromTrivia=false, lastUpdated=1458657166000, questionId=079b534c-06f4-4361-a8df-48b71e1164f2, createdDate=1457899550000, questionTitle=What are Pomeroy college basketball ratings?, noteOptions=[{hasComment=false, element={original=https://games-cdn.washingtonpost.com/images/prod/quiz/015273f4-41d1-4b21-a43f-96f9b4a923e9/bc40166a-ad20-4d70-bc44-48ef41a01864-edited.jpg?ts=1457899531032, src=<img src="https://games-cdn.washingtonpost.com/images/prod/quiz/015273f4-41d1-4b21-a43f-96f9b4a923e9/bc40166a-ad20-4d70-bc44-48ef41a01864-edited.jpg?ts=1457899531032" alt="NCAA Men's Championship Game - Kansas v Kentucky; Jeff Gross" class="note-image" /><div class="context-image-credit"></div>, edited=https://games-cdn.washingtonpost.com/images/prod/quiz/015273f4-41d1-4b21-a43f-96f9b4a923e9/bc40166a-ad20-4d70-bc44-48ef41a01864-edited.jpg?ts=1457899531032, id=-1.0, credit=NCAA Men's Championship Game - Kansas v Kentucky; Jeff Gross, type=IMAGE}}, {hasComment=false, element={src=<p>The Pomeroy College Basketball Ratings are a series of predictive ratings of men's college basketball teams published by Ken Pomeroy. The ratings are based on the Pythagorean expectation, derived from (AdjO^11.5)/(AdjO^11.5+AdjD^11.5).</p><p><b>Adjusted offensive efficiency</b> - An estimate of the offensive efficiency (points scored per 100 possessions) a team would have against the average D-I defense.</p><p><b>Adjusted defensive efficiency</b> - An estimate of the defensive efficiency (points allowed per 100 possessions) a team would have against the average D-I offense.</p>, name=null, id=-1.0, type=HTML_TEXT}}], multipleSelectionAmount=0}], archive=false, questionIds=[079b534c-06f4-4361-a8df-48b71e1164f2], title=What are Pomeroy college basketball ratings?, tags=[Pomeroy College Basketball Ratings, Ken Pomeroy, ncaa, march madness, ncaa tournament], articleReferences=[], createdBy=greenbergna, allowMoreThanOnce=false, shareUrl=http://wapo.st/1QRCEqw, captchaProtected=false, embedToggles=[{headline=false, byline=false}], byline=Neil Greenberg, live=true, photoUploaded=false, status=ACTIVE}, _service_=com.washingtonpost.platform.castor.contents.JgeContentService, apiRoot=https://notes.washingtonpost.com/notes/games/webapi, embed=<div class="context-note"> <link href="https://games-cdn.washingtonpost.com/shared/bootstrap.min.css" rel="stylesheet"> <link href="https://games-cdn.washingtonpost.com/shared/notes/notes.css" rel="stylesheet"> <div class="context-header"> <div class="note-headline"> <div class="note-title">What are Pomeroy college basketball ratings?</div> <p class="note-byline">by Neil Greenberg</p> <p class="note-timestamp">Updated Mar 22, 2016 at 10:32 AM</p> </div> </div> <div class="context-body"> <div class="notes-inner-content context-block"> <img src="https://games-cdn.washingtonpost.com/images/prod/quiz/015273f4-41d1-4b21-a43f-96f9b4a923e9/bc40166a-ad20-4d70-bc44-48ef41a01864-edited.jpg?ts=1457899531032" alt="NCAA Men's Championship Game - Kansas v Kentucky; Jeff Gross" class="note-image" /><div class="context-image-credit"></div> <p>The Pomeroy College Basketball Ratings are a series of predictive ratings of men's college basketball teams published by Ken Pomeroy. The ratings are based on the Pythagorean expectation, derived from (AdjO^11.5)/(AdjO^11.5+AdjD^11.5).</p><p><b>Adjusted offensive efficiency</b> - An estimate of the offensive efficiency (points scored per 100 possessions) a team would have against the average D-I defense.</p><p><b>Adjusted defensive efficiency</b> - An estimate of the defensive efficiency (points allowed per 100 possessions) a team would have against the average D-I offense.</p> </div> </div> </div>, _config_={_jge=note, id=fc7bff57-b177-4c12-8912-69d0fcaf2e63}}