(* As we note every year, this might be more like a Patrick Ewing guarantee than a Joe Namath guarantee.)
The Perfect Bracket is a unique creation in that it incorporates more than just a series of projected game outcomes. It also utilizes the value of picking upsets to differentiate your picks from the others in the pool. Combine the two factors and you get what we believe to be the best possible picks to win your pool. And speaking of upsets, that’s a good place to begin our breakdown of this year’s bracket.
First, don’t stress too much over the first round and be conservative with picking upsets. Since 2011, the first year the tournament was expanded to 68 teams, no winner in the official bracket game of the NCAA tournament has had more than 28 of a possible 32 points in the first round, and most years they end up scoring a total of 25 points or fewer in the Round of 64. That translates to seven incorrect picks on average. Plus, since 2011, the higher seed has historically won 74 percent of first- and second-round games, so go with the chalk early, zeroing in on only a handful of upsets.
|Round (2011 to 2017)||Better seed wins …|
|Round of 64||74%|
|Round of 32||74%|
The low win rate of higher (meaning worse) seeds in the Elite Eight is a bit misleading, since oftentimes the seed differential is minimal (for example, a No. 3 seed beating a No. 2) and their true talent may not have been that great in the first place. In the Elite Eight, simply go with the stronger team and don’t worry about the seed. We’ll show you a little more on which teams are in fact stronger in a little bit.
Another principle rule of the Perfect Bracket: Don’t pick a No. 1 seed to be an upset victim in the first two rounds. Over the past seven tournaments, top seeds have a 51-5 overall record through the round of 32. That’s an 91 percent success rate. Don’t advance the lower-seeded teams too far, either. The No. 14, No. 15 and No. 16 seeded teams have gone 11-85 (11 percent win rate) against opponents seeded at least four spots higher than them.
So, what does perfection look like? Here’s a region-by-region breakdown of this year’s bracket. The percentage next to each team’s name is the chance that team has of advancing to that round, not winning that specific matchup. For example, Villanova has a 98 percent chance of beating No. 16 Radford/LIU Brooklyn in the round-of-64, and an 85 percent chance of reaching the Sweet 16 after beating the winner of No. 8 Virginia Tech/No. 9 Alabama.
As you’ll note below, the Wildcats do not advance to the Elite Eight in the Perfect Bracket. (Again, the 42 percent figure next to West Virginia in that round represents their odds of reaching that spot in the bracket, not their chances of upsetting Villanova). Picking the Mountaineers over the Wildcats, given a relatively strong upset projection, is also worth it because of the advantage you would receive over other brackets in your pool that have likely picked the favorite.
West Virginia generates extra possessions via offensive rebounds (37 percent offensive rebound rate, fourth highest in the nation) and turnovers on defense (23 percent, second). That will help slow down No. 1 Villanova and the nation’s highest adjusted offensive rating (127.4) according to Ken Pomeroy. The Mountaineers also excel in transition defense, allowing opponents to score just 41 percent of the time, the lowest rate among schools having to defend at least 500 possessions in transition this season — no small thing considering Villanova is the highest-scoring transition team in the country (1.2 points per possession).
Plus, the Wildcats rely heavily on the three-point shot, taking 47 percent of field goals from behind the arc. The league average is 37 percent. According to Ed Feng, creator of the sports analytics site ThePowerRank.com, teams that rely on three-point shooting tend to have a wider variance in their point outcomes. Feng’s research shows that since 2002 only one eventual champion took a significantly higher percentage of field goals from beyond the three-point line than the collegiate average. Of course, that was Villanova in 2016.
No. 2 Purdue, the fifth-best team according to Pomeroy, should have a relatively easy time making it through to the Final Four, as demonstrated by their 43-percent likelihood of reaching that round. They score the second-most points per 100 possessions (123) after adjusting for opponent with the fifth-best effective field goal rate (58 percent), fueled by their 42 percent shooting from behind the three-point line, the second-highest accuracy in the nation.
One first-round upset to watch for is No. 10 Butler over No. 7 Arkansas. The Bulldogs have good spot-up shooters in Kelan Martin (effective field goal rate of 53 percent), Sean McDermott (68 eFG%) and Henry Baddley (66 eFG%) with Martin and Kamar Baldwin also handling the ball during the pick-and-roll. Arkansas struggles against spot-up shooters (bottom 30 percent of the nation), giving Butler some easy points en route to the upset. According the Data-Assisted Victory Detector (DAViD) for the NCAA tournament (click here for the full explanation of the method), Butler has a score of 152, classifying the Bulldogs as a great value in the first round.
Xavier was rewarded with a No. 1 seed but you could argue it didn’t deserve one. Pomeroy had the Musketeers as the 14th best team in the country, outscoring opponents by 22 points per 100 possessions, the same as a pair of No. 5 seeds in West Virginia and Ohio State. The Musketeers defense doesn’t create many turnovers and struggles in transition (1.1 point per possession allowed, putting them in the bottom eight percent of teams) and against the ballhandler on the pick-and-roll (0.87 points per possession, bottom 12 percent) — two traits that could have them going home sooner than expected.
No. 2 North Carolina, meanwhile, has embraced small ball, using 6-foot-6 senior Theo Pinson at power forward and 6-8 junior Luke Maye at center. Pinson has shown good court awareness out of the post, often finding Joel Berry II, Kenny Williams and Cam Johnson on the perimeter. Those three combine for over 17 three-point attempts per game with 36 percent accuracy, making them a nightmare to defend by even the best of teams.
Maye, meanwhile, averages 17.2 points and 10.1 rebounds per game. And while he doesn’t take as many three-point shots as his teammates, he makes them at a rate of 44 percent.
The one small surprise in this region is the projected “upset” of No. 5 Ohio State over No. 4 Gonzaga. Gonzaga should have no trouble getting past No. 13 UNC Greensboro, but Ohio State should be a tougher matchup. Based on their adjusted net ratings, Gonzaga would be a 2.5-point favorite on a neutral court, giving them a 57 percent win probability. That’s low for a team being picked in 70 percent of brackets on ESPN to move on to the Sweet 16, making Ohio State a decent value in this slot.
The Jayhawks finished the season ranked No. 9 in Pomeroy’s rankings, which, if used to seed the teams, would make them a No. 3 seed. According to research by Jeff Feyerer, since 2004, there have been 41 teams that have been “overseeded” in this way; just three — 2015 Duke, 2014 Wisconsin and 2009 Michigan State — have made it to the Final Four.
Duke might be too aggressive on the offensive boards, even if the Blue Devils are successful (39 percent, highest in the nation), making them vulnerable to plays in transition on the other end. Their first-round opponent, Iona, scores 1.1 points per possession in transition, putting them in the top 25 percent of college teams. While no one should be picking that upset, potential second-round opponents No. 10 Oklahoma and No. 7 Rhode Island rank in the 81st and 93rd percentile, respectively, for transition offense and Michigan State, a potential Sweet 16 opponent, ranks in the 85th percentile and can match Duke’s rebounding prowess. One of those teams is bound to take advantage of Duke’s weakness and keep them from advancing far in the Midwest. In the Perfect Bracket, that team is Michigan State.
New Mexico State, a 12 seed, could be a dark horse. The Aggies went 28-5, with wins over tournament teams Miami (Fla.) and Davidson, plus won the WAC tournament. Sure, they will be looking for their first NCAA tournament win since 1993 and have lost 10 straight March Madness games since, but their defense ranks 14th in the country thanks to limiting opponents to an effective field goal rate of 45.5 percent. Two of their defenders, junior Sidy Ndir and sophomore A.J. Harris, allow 0.75 and 0.71 points per possession, respectively, putting them in the top 15 percent of all college players.
This region should go according to seed. No. 1 Virginia and No. 2 Cincinnati were two of the best teams all season and don’t have much opposition in the region. And yes, that’s true even after accounting for the DeAndre Hunter injury.
The Bearcats should have no trouble with No. 15 Georgia State, but could face a scare against No. 7 Nevada if the Wolf Pack gets by No. 10 Texas. Nevada can score (10th best offensive efficiency per Pomeroy), shoot the three (40 percent) and take care of the ball (14 percent turnover rate, third best). But even if Cincinnati prevails, as the Perfect Bracket predicts, Virginia’s smothering defense and above-average offense will be too much for them to overcome.
No. 4 Arizona could stir up trouble with freshman Deandre Ayton peaking at the right time. The 7-foot 1, 261-pound center scored 32 points with 18 rebounds in the Pac-12 championship game and is lethal around the basket, scoring on 72 percent of his shot attempts. He also averages 1.26 points per possession in the post and 1.4 points per possession on cuts to the basket.
Final Four and beyond
A perfect Final Four is nice to have, but not essential. Less than half a percent of people that have filled out a bracket in the past six years have gone 4-for-4. You’re much more likely to get two right, preferably the same two teams that will be up against one another in the title game.
Virginia has the best chance to make the Final Four, appearing 52 percent of the time as the representative of the South. Purdue and North Carolina are next, followed by Michigan State and Duke.
The most important pick in any bracket is the national champion. In most scoring systems, choosing the correct national champion is worth the same number of points as going 32-for-32 in the first round, and, according to data from the past seven years of the official bracket game of the NCAA tournament, every one of the past seven winners had the participants and winner of the national championship game right. Luckily, winnowing down the 68 teams in the tournament to a select few who should be the national champion is relatively easy.
Over the past 16 years, every national champion except one, Connecticut, a No. 7 seed in 2014, was a No. 1, 2 or 3 seed. Winners have also played in one of the top five strongest conferences per the Simple Rating System, a schedule-adjusted margin of victory rating that is expressed in points per game, with an SRS of zero indicating an average team. And all but three of the past 16 winners have had their own, individual SRS rank in the top four nationally.
The qualifying conferences this year include the Big 12, Big East, ACC, SEC and Big Ten, highlighting Villanova (25.7 SRS), Duke (25.0), Purdue (24.2), Virginia (23.9) and Michigan State (23.5) as our potential national champions. Among those, Virginia (23 percent), Villanova (17 percent) and Duke (16 percent) have the best chance to win in these simulations.
Since Virginia is the only team of the three to not have any serious blemishes on their resume, the Cavaliers are our team to win it all. Their finals opponent, Michigan State, lands there ahead of Duke due to pool-related value. Currently 10 percent of brackets are on the Blue Devils to win it all (compared to seven percent for Michigan State), meaning a Spartans win would give you a potentially decisive edge late in the tournament.
More on the NCAA tournament: