By William Ezekowitz
Preseason rankings may seem irrelevant when it comes to college basketball, but they are surprisingly important. For example, they are just as predictive of ending up in the Final Four as the current AP Poll. The reasoning behind this phenomenon is that the preseason AP and Coaches Polls take into account all of the changes that models have more difficulty accounting for (i.e. coaching changes, big recruiting classes, injuries, etc.). The conventional wisdom going into the season is actually fairly smart.
With this intuition, I decided to examine the teams who defied that wisdom throughout the season and went from unranked in the preseason to ranked in the last poll before the Tournament. Had the polls gotten something wrong? Or did these over-performing teams regress back to their preseason expectations in the Tournament? Using the past ten years of examples and seed win expectation data from FiveThirtyEight, I investigated.
Of the 97 teams who qualified for analysis, 32 outperformed their win expectation, while 65 underperformed. As a whole, a group expected to win 138.8 games won just 107. In fact, only one team, Kemba Walker’s 2011 UConn squad, even made the Final Four. A T-test of this difference revealed it to be statistically significant with a P-value of 0.0099.
Moreover, in none of the tested years did the sum of the teams’ actual wins exceed their expected wins. Put another way, even though individual teams may have outperformed once in a while, the group as a whole always underperformed.
So what does that mean for this year? Who are the teams whom we shouldn’t trust?
This year’s group is projected to win 11.1 games. The total of nine teams is the most since 2013, when, led by 2 seed Georgetown, who famously lost to Florida Gulf Coast, seven of the nine teams underperformed.
History has been repeating itself for ten years. Maybe a couple of these teams will beat the statistics. But on the whole, this group will underperform. Bet on it.
Butler is a 4 seed not a 5 seed as this article postulates.
You should really source Nate Silver’s 2011 NYT article, which studied exactly the same thing you did and reached the same conclusion…