ESPN’s Basketball Power Index (BPI) was originally designed to rate college basketball teams, not predict tournament results. In seasons past it was used as a résumé rating and predictive system to determine the best teams in the country.
BPI has performed about as well as comparable systems when projecting the tournament, but ESPN’s Sports Analytics Team enhanced the model by adding more predictive features into its tournament projections.
In addition to better account for a team’s pace- and opponent-adjusted scoring margin to rate teams, BPI’s game predictions now account for distance traveled, days' rest and altitude, which are factors found to be predictive of tournament success. The end result is that BPI should be even more accurate than in seasons past in predicting tournament results.
With that in mind, BPI’s tournament projections, which account for the strength of the teams and path to the title, allow us to uncover upset picks, sleepers and favorites for the 2016 NCAA tournament. Below are a few storylines using BPI’s projections.
Kansas is the tournament favorite
Kansas enters the 2016 NCAA tournament with a 17 percent chance to win the title, about four percentage points better than any other team in the field. The Jayhawks are in the midst of a 14-game winning streak that helped them capture their 12th straight Big 12 regular-season title and first Big 12 tournament championship since 2013.
During the streak, Kansas has been the most efficient team in the country and moved up significantly in BPI. In addition to its team strength, Kansas has a favorable path to the Elite 8 -- both in terms of opponents and game site -- and BPI projects the Jayhawks have a 71 percent chance to win their first three games.
Although Kansas is the BPI favorite to win the tournament, it’s worth noting that there is an 83 percent chance that some other team will cut down the nets. In a year defined by parity, don’t be surprised if this is one of the least “chalky” brackets in recent years.
Is this the year that a No. 16 seed finally knocks off No. 1 seed? Probably not, but BPI gives the 16 seeds a combined 6 percent chance to win a game.
How about a No. 15 seed beating a No. 2 seed? BPI projects there is nearly a one in five chance that a No. 15 seed wins a game in the first round for the eighth time in tournament history.
With the gap between the top teams and middle-tier teams this season smaller than in years past, don’t be surprised if mayhem ensues.
That may not mean that a No. 12 seed will reach the Final Four for the first time in tournament history, but it could mean that a higher-than-average number of double-digit seeds reaches the second weekend (BPI projects 1.8) or that a No. 4 seed reaches the Final Four (42 percent likely).
Overall, there is a 22 percent chance a team seeded fourth or lower wins the tournament. That may not seem high until you consider that a top-three seed has won 87 percent of tournaments since the tournament expanded in 1985.
BPI still expects the top seeds to do well, but there will likely be more upsets than what we saw in last year’s fairly predictable tournament.