After a topsy-turvy year, many believe the NCAA tournament will be particularly wide open this year. There will be a little extra madness.
Do the numbers back that up? Yes, but the degree depends on how we look at it.
Using our current BPI projections for the NCAA tournament compared to past projections on Selection Sunday (so not apples to apples but close enough) back to 2007-08, we can see how the current range of tournament outcomes for this year compares to the past.
If we're looking strictly at the favorite: we currently make that Kansas, with an 17% chance to win it all. That's a shorter favorite than we've had most seasons, but not the shortest.
In the 2016 tournament Michigan State was narrowly the tournament favorite at just 15% chance to win, though that was also a product of the teams around the Spartans in the bracket. BPI considered Virginia and Michigan State the two best teams in the country heading into that tournament -- and they were placed in the same region. No. 10 seed Syracuse actually ended up winning that region and Villanova won it all.
The following year BPI considered Gonzaga a full point per game worse than Villanova but made the Bulldogs favorites at 16% anyway. In this case, the favorite was actually buoyed by circumstance, as very weak No. 2 (Arizona) and No. 3 (Florida State) seeds filled out their region. Gonzaga lost the national championship game to North Carolina.
Currently we project Kansas to be in a better position than both of those teams, but far worse off than dominant favorites of years past such as Kentucky in 2015 (38%) or Virginia last season (33%).
But the favorite's chance to win does not depict the entire field. If we zoom out just a little and look at the cumulative chance of the top-three favorites -- in this year's case: Kansas, Duke and Gonzaga at a cumulative 49% -- that's the fourth-lowest championship chances since 2008 for the top trio. (Quick aside: BPI is unfazed by recency and therefore still relatively high on the Blue Devils, despite their slew of losses of late.)
But if we zoom out even more, that's where it starts to get really interesting. The current projections give an 80% chance that one of the top 10 most likely winners ends up taking the title -- a virtual tie for lowest among seasons in our dataset. That of course means there's a 1-in-5 chance a team from outside that group of 10 wins the title.
That's teams like Arizona, Maryland, Villanova , BYU, Oregon, Seton Hall, Michigan or West Virginia -- who all have at least 1% shot, per BPI.
Of course, tournament craziness can come from somewhere other than the national champion. So another question: will there be more first-round major upsets than there have been in recent past? Among Nos. 1 through 5 seeds since 2008, there have been an average of 3.6 first-round exits per year. If we look at the top 20 teams in this year's upcoming tournament based on potential seed, BPI expects ... 3.5 losses for those teams.
So don't expect more first-round madness in this season's tournament. But as for the championship? More than we've seen in quite some time, there are plenty of contenders with at least a shot to win it.