<
>

BPI's potential under- and over-seeded teams

Aaron Doster/USA TODAY Sports

Joe Lunardi released his latest Bracketology on Thursday and, although ESPN’s Basketball Power Index (BPI) and Lunardi agree on quite a bit, there are some marked differences worth examining.

Before diving into the seeds, it’s worth noting that BPI is not designed to predict the tournament field. That’s what Lunardi is for. BPI is a power rating that can be used to break down the current field to determine which teams can make a deep run, outperform their seed or pull off an upset.

How has BPI done in identifying over- or under-seeded teams in past tournaments?

Over the last four seasons, there were 24 teams that BPI believed were over-seeded by at least three spots. Those teams won 6.9 fewer games than expected, based on their seeds, and 10 did not make it out of the Round of 64. The 2011-12 Michigan Wolverines, 2012-13 Kansas State Wildcats and 2014-15 Providence Friars were all BPI over-seeded teams that did not make it out of the Round of 64.

Similarly, over that same time period, there were 26 teams that BPI believed were under-seeded by at least three spots. Those teams won 12.4 more games than expected, based on their seeds, and seven made it to the Sweet 16. The 2013-14 Kentucky Wildcats, 2012-13 Wichita State Shockers and 2011-12 Florida Gators were all BPI under-seeded teams that advanced to at least the Elite 8.

Teams that BPI values more than Bracketology

Taking the current 68-team field in Bracketology and reseeding the teams based on BPI yields four teams that are seeded at least three spots better in BPI than Bracketology. These four teams have the potential to outperform their seed and make a deeper-than-expected run in the tournament.

The Cincinnati Bearcats have the biggest projected seeding difference when it comes to BPI and Bracketology. Despite a 19-7 record, the Bearcats rank 25th in BPI due in part to a number of close losses and an above-average efficiency rating.

With four of their seven losses coming by exactly two points -- tied for the most such losses in the nation -- the Bearcats could easily be sitting atop the AAC standings. They rank 16th in the nation in net efficiency, including eighth in defensive efficiency, which is a better predictor of future performance than a team’s record alone.

The Connecticut Huskies, VCU Rams and Butler Bulldogs are three more teams that rank in the top 40 in BPI that could outperform their seed in the tournament.

Teams that BPI values less than Bracketology

If the current Bracketology field was reseeded using BPI, there are seven teams – the Dayton Flyers, Utah Utes, Oregon Ducks, Baylor Bears, Colorado Buffaloes, Providence and Monmouth Hawks – that BPI would have seeded three or more spots worse. These are the teams that have the potential to underperform their seeds and potentially make an earlier-than-expected exit.

Let’s focus on the two highest-seeded teams: No. 3 Oregon and No. 4 Dayton.

Even after dropping its last two games, Oregon has a solid record in a deep Pac-12 (20-6), which is a major reason why it ranks third in RPI. Unlike RPI, BPI accounts for how the Ducks won their games. They rank 52nd in scoring margin and 53rd in net efficiency, which generally does not bode well for a top-three seed. BPI still gives Oregon credit for winning 20 games against a difficult schedule, but it would seed the Ducks as a No. 6 seed instead of a No. 3 seed.

Like Oregon, Dayton (21-4) has an impressive résumé but may not be as strong as its record suggests. The Flyers rank 39th in scoring margin and 45th in net efficiency against BPI’s 77th-ranked schedule. If BPI were to reseed the field it would seed Dayton as a No. 8 seed instead of a No. 4 seed.

With 24 days remaining until Selection Sunday, there is plenty of time for teams to rise and fall in Bracketology and BPI. Nonetheless, these are the types of teams to keep an eye on as the season comes to an end because they have the potential to make or break a bracket.