Ahead of Saturday’s Final Four matchup between Villanova and Oklahoma, our prediction model had the Wildcats only slightly favored, with a 54 percent chance of winning, and nearly a quarter of the game was as tight as that probability suggests. When Oklahoma led 17-16 after eight minutes, our in-game win probabilities listed the odds as essentially the same as they’d been before tipoff. Fans across the country settled in for an exciting game, perhaps like the one Villanova had played against Kansas in the South Regional final.
But from that point on, things were anything but close. ’Nova rattled off a 12-0 run, touching off an eight-minute sequence in which it outscored Oklahoma 21 to 4. By halftime, the Wildcats led by 14 — but they weren’t done yet. After the Sooners briefly cut Villanova’s lead to single-digits with 16 minutes left in the game, the Wildcats scored 49 of the game’s next 63 points, including a 25-0 run that lasted approximately five and a half minutes on the scoreboard (but must have felt like an eternity to Sooner fans).
The result was a 44-point landslide win for Villanova, the most lopsided victory in Final Four history. That is a matter of historical fact. But using our Elo ratings (which estimate a team’s strength at a given moment), we can also say that it was probably the most impressive NCAA Tournament win in more than 53 years — and the 10th-most-impressive D-I basketball victory, period, since the 1949-50 season:
|DATE||WINNER||OPPONENT||GAME TYPE||SCORE||ELO-ADJUSTED POINT MARGIN|
|12/10/1994||Southern Utah||South Alabama||REG||140-72||+71|
|3/11/1963||Loyola (IL)||Tennessee Tech||NCAA||111-42||+70|
Because Elo measures the difference in relative quality between teams going into a game, it can be used (in conjunction with information about the location of the game) to create a predictive point spread. It can also be used to generate a hypothetical point spread that would have been expected from an average Division I team1 against the same opponent in a given game. So against Oklahoma on a neutral court Saturday, for instance, Villanova was expected to win by about 2.5 points; an average team would have been expected to lose by 21. That Villanova won by 44 implies that the Wildcats outperformed their own expectations by 41.5 points and those of an “average” team by about 65 points.
Suffice it to say that 65 points is an extraordinarily wide margin for a team to beat the D-I average by in a single game. The record since 1949-50 in any game between two D-I schools is 73, set by North Carolina when it trounced Manhattan College by 84 points in 1985. (Elo estimates that an average team, playing at home, would have beaten the 1200-rated Jaspers by about 11 points.) But that also took place in a forgettable non-conference game two days after Christmas.
To find an NCAA Tournament win more impressive than Villanova’s romp, you’d have to go back to 1963, when Loyola of Chicago exceeded average by 70 points with a 111-42 opening-round triumph over Tennessee Tech. And before the Wildcats’ win Saturday, no team had beaten average by 60 or more points in an NCAA Tournament game since 1971, when a previous incarnation of Villanova beat Penn 90-47.