Over the last week, I’ve added two new features for Insiders. You can find them under the “TIPS+” section. The first was about key performance indicators for seed classes, and the most recent is about how to spot the most likely Cinderellas—and victims—in the dance.
I touch on the basics of upset spotting, provide three simple guidelines for identifying giant killers, then go round by round, offering conditions that would significantly improve your odds of picking upsets.
One issue we settle in the article is what constitutes an upset. Not every game in which a lower-seeded team knocks off a higher seed is an upset. Nobody’s going to fit a glass slipper on a nine seed that beats an eight seed in round one. It’s only when you get a gap of at least four seed positions between opponents that a game has upset potential.
Surprisingly, two-thirds of tourney games meet this condition. Of the 1,764 games that have been played in the last 28 years, 1,170 of them have pitted longshots against favorites—and the underdog has won about one in five times. That’s an average of 8.6 upsets per tourney, or roughly one in every seven games. This chart shows the round-by-round results of upset games in the 64-team era.
Because of the way the brackets are set up, most of the upset match-ups occur in the first three rounds of the tourney. Of the 1,568 games played in these rounds, 1,125, or 72%, have been upset match-ups. Picking these upsets correctly—or at least minimizing the number of victims you advance—is essential to building a winning bracket.
While first-round upsets grab most of the attention, the upsets in the second and third rounds do the most damage to your bracket. Only 22 of the 122 opening-round upset victims are teams seeded one, two or three. The top three seeds are a dominating 314-22 (.935) against first-round darkhorses. Only about one in 16 top-three seeded teams will fall victim to an upset in round one. Last year was an anomaly. Two 15 seeds, Norfolk State and Lehigh, stunned two seeds Missouri and Duke. Before last year, 15 seeds were just four for 108 in beating two seeds.
High seeds are much more prone to upsets in the second round. The top three seeds are just 193-60 against Cinderellas. That’s a solid .763 winning percentage, but it’s nowhere near the lockdown .935 rate of the opening round. Instead of only one in 16 teams losing, the top three seeds lose about one in every four games—and more than two per year. They also comprise 79% (60 of 76) of the second-round upsets. Last year, only one top-three seed got slayed in the second round. Third-seeded Georgetown got edged by NC State, 66-63.
The Sweet 16 isn’t as treacherous for the top three seeds as round two, but it isn’t a first-round cakewalk either. Of the 32 third-round upsets of the modern era, 21 have involved teams seeded one, two or three. But they do have a .819 winning rate against longshots (95-21). Not surprisingly, top seeds have the least trouble with Cinderellas (50-7, .877), followed by two seeds (30-7, .811), then three seeds (15-7, .682). Last year, the only high seed to be slain by a darkhorse was three seed Marquette, which lost seven seed Florida.