Valparaiso may have ended the season by winning 9 of their last 10 games, but given the margin-of-victory in their games and the quality of the competition, that record isn’t as impressive as it sounds. In the next table I construct the right measure of hot teams. I control for opponent and venue and recreate the Pomeroy Rankings focusing only on the Final 10 games of the season.

These numbers make more sense to me than the actual Pomeroy Rankings. Georgetown was hot late in the year, and these numbers show that. Florida was good, but they weren’t top overall team good. Indiana may not have won the Big Ten Tournament, but given the gauntlet they faced late in the season, their end of season performance was still impressive.  And Louisville is the top overall team, which fits perfectly with what we have all seen down the stretch.

I love Trey Burke, but given how Michigan ended the year, I simply cannot pick the Wolverines for a deep run. This table also scares me away from Miami. Yes the Hurricanes won the ACC tournament and their offense looked great. But keep in mind they only had to beat three bad defensive teams (North Carolina, NC State, and Boston College) to do it. Miami’s struggles at the end of the regular season suggest this team isn’t nearly as hot as some of the other elite teams.

My Bracket Picks

I have a method to my madness, and list some fun stats below, but I’m not going to sit here and say I have a magical formula for picking upsets or Final Four teams.

Champion: Duke

With Ryan Kelly in the lineup, Duke was dominant. Duke’s numbers don’t look all that great in the table that led off this column, but that still includes six games with Kelly out. The loss to Maryland in the ACC tournament certainly shook my confidence somewhat. But if we eliminated teams based on an occasional bad performance this year, there wouldn’t be any team in the Final Four. I honestly believe Louisville is the best team this year. But this hasn’t been the kind of year where the best team wins. Whenever we anoint anyone as the clear cut favorite, they fall on their face. And given that Duke is my second favorite choice overall, I feel like the Blue Devils provide the best tournament odds in a pool setting (except in ACC country).

Runner-up: Kansas

Here are the teams with the most wins against NCAA tournament teams seeded 13 or better. (I throw out the wins against the 14-16 seeds because many of those wins are less impressive.) Kansas is in a class by themselves:

Kansas 13, New Mexico 11, Louisville 10, Duke, Wisconsin, and Indiana 9

I also love Kansas’s NCAA tournament draw. A lot of people will be high on North Carolina based on what they did late in the year (see top table), but even with the smaller lineup utilized late in the year, North Carolina wasn’t able to beat elite teams like Duke and Miami. North Carolina only started to figure out how to beat the mid-level ACC teams. Against teams seeded 1-13 in the field, North Carolina is a pathetic 2-8 on the season. In fact, the Tar Heels have the worst Pythagorean rating of any major conference team in the field against that group. I simply don’t see how North Carolina can defend Kansas’ big people or hit enough jumpers against the stellar Kansas defense to have a chance. In fact, I see North Carolina getting tripped up in the first round by Villanova.

And as noted above, Michigan is in a big-time slump late in the year. Florida has struggled late in the year, particularly in any pressure situations. And while I respect John Thompson III as much as any coach in the nation, I also feel like Georgetown has over-achieved this season. Kansas is my clear favorite in the South and I like them to make it all the way to the title game. The rest of this is going to look predictable for awhile:

Final Four: Indiana, Ohio St., Elite Eight: Louisville, Gonzaga, Florida, Miami

Sweet Sixteen: St. Louis, Michigan St., Kansas St., New Mexico, Michigan, Georgetown, Syracuse, Butler

The toughest Sweet Sixteen choice for me was who to pick in the East. Marquette was clearly the least consistent team among the Top 4 seed lines. It is a no-brainer to pick someone else in that slot. But Butler isn’t an obvious pick based on how they struggled late in the year. Still, I’ve learned not to overthink the numbers with Brad Stevens. He will have the Bulldogs prepared. With Kansas St. I am playing a bit of a hunch. Bruce Weber knows how to prepare his players mentally for Wisconsin’s slow pace after having played the Badgers in so many previous seasons in the Big Ten, and the Badgers lack of fouling won't bother Kansas St. at all.

Round of 32 Upsets: Davidson, California, Belmont, Middle Tennessee, Villanova

Round of 32 Favorites: Oklahoma St., Wisconsin, VCU, Pittsburgh

Round of 32 Not Ice Cold: UCLA, Creighton, NC State, Illinois

Since Minnesota, Cincinnati, Temple, and Colorado all had rather unimpressive ends to their season (see table at start of this post), I can’t pick any of them to advance.

Round of 32 Coin Flips: Colorado St., Iowa St., San Diego St.

Notre Dame and Iowa St. both shoot a ton of jump shots. It just depends who makes them. Neither team plays good enough defense to slow the other team down. This will just come down to who is hot at the right time.