This article is part of our March Madness series.
For every great upset in the men’s NCAA Tournament, there is a favorite that found a way to lose a game it was supposed to win.
Sometimes it happens because the better team has an off day or didn’t prepare well enough for an underdog with something to prove. But other times, it’s the result of flaws in the higher-ranked squad’s style of play — or perhaps even evidence that it shouldn’t have been seeded so highly to begin with. In the latter kinds of cases, there are often clues strewn about that hint at a favored team’s impending doom.
With that in mind, we dug into the top-four seeds with clear red flags that could come back to haunt them before the tournament is over. Here are four types of highly seeded teams to be wary of when filling out your bracket this week:
Not every team that’s considered a contender on Selection Sunday is deserving of its status. That has perhaps never been more evident than this season, as this year’s bracket features two of the 10 most overseeded teams ever, according to Ken Pomeroy’s ratings.1 On Selection Sunday, Providence and Wisconsin joined the ranks of teams to earn top-four seeds despite relatively low efficiency marks.
|Season||Team||Adj. Eff. Rk||actual||Kenpom||Deviation||Exit|
But both overseeded teams kept on winning all season: Providence just won the Big East regular-season title, and Wisconsin boasted the fourth-most quadrant 1 wins during the regular season. So what’s masking each team’s weaknesses? The short answer is good fortune.
Pomeroy’s site defines “luck” as the difference between a team’s actual winning percentage and what we would expect from its underlying efficiency ratings — essentially measuring how fortunate a team has been by the disproportionate number of close-game coin flips in its favor. Among top-four seeds, here are the teams with the most favorable breaks in that regard:
|Team||Region||Seed||KenPom “Luck” 🍀||National Rank|
Unfortunately, deviations between a team’s actual and expected winning percentages seldom last long,2 which could spell trouble for a team whose resume is built on winning tight games. Perhaps no team embodies this category more than Providence, the No. 4 seed in the Midwest. The Friars were not only the luckiest team in the nation this season, but they were one of the luckiest teams in the history of Pomeroy’s metric. While they may not quite be on track to lose right away — we give the Friars a 67 percent chance of winning in the first round against South Dakota State3 — Providence would be a clear underdog in the second round, most likely against Iowa (a team with an inferior seed and record, but only because it had some of the worst luck of any tournament team).
Meanwhile, Wisconsin was the nation’s ninth-luckiest team during the regular season and conference tournaments. The Badgers had an .857 winning percentage in games decided by 5 points or fewer, which was tied for 10th-best in the country, but produced only a .706 winning percentage in all other contests — a number perhaps more indicative of their true talent. It probably will not matter in the first round against Colgate,4 but Wisconsin will have to hope its luck doesn’t go south in a potentially tough second-round matchup against the LSU-Iowa State winner.
Another often telltale sign of March Sadness for higher-seeded teams is a relative imbalance between offensive and defensive capabilities. The safest bets in the NCAA Tournament don’t tend to rely too heavily on one side of the ball — and you might want to pay particular attention to defense. Since the 2001-02 season, when KenPom started tracking adjusted efficiency stats, only five of the 76 Final Four teams have posted a regular-season defensive efficiency outside the top 50, while eight were outside the top 50 on offense. And when it comes to cutting down the nets, balance is a plus: Fourteen of the past 19 champions have featured pre-tournament rankings in the top 15 on both offense and defense.
We can also look at underperforming teams — ones that lost in the tournament before their seed line would suggest5 — and how this year’s top seeds echo past underachievers:
|Team▲▼||seed▲▼||Adj. off. rk▲▼||Adj. def. rk▲▼||Off.▲▼||Def.▲▼|
There’s one obvious candidate to be the next one-trick pony to fall short thanks to shoddy defense: the Purdue Boilermakers, who are perhaps the poster children for elite scoring paired with mediocre stopping. Led by the electric Jaden Ivey, the Boilermakers boast the third-most-efficient offense in the country, according to KenPom, featuring a devastating blend of athleticism, elite long-distance marksmanship and prowess on the offensive glass. But they also have an extremely porous defense, ranked No. 100 in adjusted efficiency, as they allow a disproportionate share of points on 3-pointers and struggle to force turnovers (ranked 346th in Division I).
If history is any indication, you likely can’t trust Purdue to make it to the Big Easy. No top-four seed with a defense as bad as Purdue’s has advanced to the Final Four since at least 2002, and only one made it with a defense ranked 75th or worse (Marquette in 2003). Providence makes an appearance here, too, with a defensive efficiency at No. 79 that’s more than 40 spots worse than the average underperforming No. 4 seed, but nobody has sacrificed defense for offense quite like Purdue this season. That’s disappointing news for a team that reached No. 1 in the AP Poll in December and that many thought could be the squad to capture the program’s elusive first title.
On the other side, meanwhile, Texas Tech stands out as having the worst-ranked offense relative to the average underperforming team at its seed. The Red Raiders’ defense needs no introduction — after program architect Chris Beard hitched his wagon to the hated Texas Longhorns, new coach Mark Adams molded a unit on par with the best defense Beard ever had in Lubbock. However, Adams has been unable to replicate the results on offense, as the Red Raiders have just the 74th-best effective field-goal percentage in Division I and turn it over on more than 20 percent of their possessions. Wisconsin makes another appearance here, thanks to a lack of emphasis on the offensive boards (250th in offensive rebounding percentage) and very poor 3-point shooting — but, unlike Texas Tech, it’s no defensive juggernaut.
One of the most counterintuitive findings in all of March Madness research is that preseason rankings are roughly as good at predicting NCAA tournament outcomes as team performance in the season itself. While this doesn’t mean that you can toss out the previous five months of results entirely when picking your bracket, it does mean that we have to keep a critical eye on teams with breakout seasons that wildly exceeded all expectations.
|Team||Region||Seed||AP Poll||Coach||Elo||Composite||Current Rk||Diff.|
This year, three teams that were unranked in the preseason AP or coaches polls have found their way to top-four seeds in the bracket: Arizona, Texas Tech and Wisconsin. The Wildcats are the biggest overachiever of the bunch, having climbed from No. 33 in a composite of polls and Elo ratings6 before the season to No. 3 in our team ratings after Selection Sunday — a 30-slot improvement that is nearly double that of any other highly seeded team. There were some reasons the Wildcats weren’t highly regarded early — they had fired coach Sean Miller in the spring (replacing him with rookie head coach Tommy Lloyd) and were coming off a self-imposed postseason ban related to a recruiting scandal — that may no longer apply. But Arizona is in a tough region with a difficult path to the Final Four as it is, and its status as a preseason longshot doesn’t make it any more enticing of a pick.
Behind the Wildcats, three other high seeds stand out for boosting their ranking by at least 10 slots since preseason: Texas Tech, Auburn and Kentucky. The Red Raiders aren’t total strangers to winning in March, having come within a few points of the championship in 2019, but their success this season under Adams went relatively unpredicted. And although the SEC’s No. 2 seeds were more highly regarded going into the year, with a couple of difference-making newcomers (Jabari Smith for Auburn and Oscar Tshiebwe for Kentucky) to help explain their ascents, they still improved rapidly enough to give some pause when pondering their tournament potential.
As a general rule, underdogs can make upsets more likely by introducing more randomness into a game — whether by slowing the game’s pace (thereby giving the favorite fewer possessions over which to flex its advantage in talent) or playing high-risk, high-reward strategies like shooting a lot of 3-pointers or trying to generate a lot of steals and turnovers on defense. It’s long been known that lower-seeded teams playing that kind of style have at least some extra chance of advancing in the tournament, compared with underdogs who play a more conventional style. However, it also stands to reason that favorites that are more susceptible to those tactics could be less likely to win than we would expect from their overall efficiency metrics.
To isolate who those teams might be, we created a composite ranking based on national ranks in three KenPom offensive categories: pace factor (from slowest to fastest), turnover rate (from highest to lowest) and the share of team points from 3-point shots (from highest to lowest). Sorting the sum of those ranks in ascending order — and, once again, looking at top-four seeds only — will give us the favored teams that tend to play the most upset-friendly style.
|Team||Region||Seed||Pace (Slowest)||Turnovers (Highest)||3P
No team rates highly in all three categories — it’s hard to be a top seed while turning the ball over constantly, for instance — but Villanova plays at a slow pace and lives or dies by the three as much as any contender. The Wildcats won two national championships in the last seven seasons playing a similar way under coach Jay Wright, of course, but this year’s version is significantly slower (No. 345 nationally) than those earlier teams, and it’s also more reliant on the 3-pointer than the championship 2015-16 squad was. Although it’s tough to see Nova having too much trouble before a potential Sweet 16 clash with Tennessee — another team on this list for its heavy reliance on 3-pointers — it’s still worth thinking about how the Wildcats’ style of play might make an upset more likely.
Broadly looking at the other categories, pace is another area in which Providence presents a red flag for bracket-pickers; the Friars have the second-slowest tempo of any top-four seed, trailing only Villanova. And Texas Tech stands out for the sheer frequency with which it turns the ball over, ranking 295th out of 358 teams in ball security — a sloppy tendency that could open the door for an upset in a potentially difficult second-round matchup against either Alabama, Notre Dame or Rutgers (if not sooner).
Check out our latest March Madness predictions.