A Few Words on Metrics
If you have a memory for such things, you may remember that 2015's bracket was one of the best ever produced, while 2016's was...not. That this year's bracket was relatively well-constructed* isn't itself a sign that we've returned to the metrics-friendly ways of 2015. Rather, this just happened to be a year where advanced stats, RPI, and the "eye test" all lined up absurdly well. That said, what is promising is the NCAA's desire to overhaul the numbers side of their current selection process. If the NCAA is genuine in its desire to replace the outdated RPI and all of its descendants**, then we should be able to eliminate a lot of the poor selections and orderings of previous brackets.
*Aside from criminal underseeding of Wichita and some weird ordering of the Big Ten teams, that is
**Virtually everyone that participates in the mock bracket exercise has said that straight-up RPI isn't used to evaluate teams, but similar things that are just as problematic (like record vs RPI top-50) are used, and those evaluations affect the proceedings just as much
In summation, most everyone agrees about the basic steps needed to improve the selection committee's toolbox. Once that happens, a majority of the problems with the selection process will be fixed or significantly improved. Even with perfect stats in hand, the philosophical debate of "best" vs "most deserving" will rage on, and continue to ultimately shape the decision process. After the stats summit, analytics godfather Ken Pomeroy weighed in with a viewpoint that may surprise: He is firmly on the side of the deservers. That said, he does make an important concession in his argument:
"I’d allow there could be a small role for predictive measures, though. A dominant team from a lesser conference is challenging to evaluate because it has fewer chances late in the season to pick up quality victories that their power-conference brethren."
While I certainly come down on the "best" side of the argument, the similarity between my views and Pomeroy's show that this isn't a binary dichotomy. Rather, "best" and "most deserving" lie on a spectrum, with dozens of sub-opinions working to form each individuals' summary position. In light of this, I wouldn't argue that everyone on the committee needs to have the same opinion on the matter. Rather, I would ask that the members take the time to fully understand their positions, the positions of others, and the surprisingly common values that make up these views. A diversity of opinions is often cited as a fundamental tenet of good teamwork, but that diversity doesn't work unless we examine and understand the true similarities and differences of one's worldview. In summation, I hope the "best" vs "most deserving" argument never dies, and that the ongoing debate around it continues to foster a better and better process of bracket creation.
Mid-Major Report
As usual, I have summarized the carnage of championship fortnight below. 2017 was a good year for the teams with the best chance of Cinderellaing as 1) all of the very best mid-majors made it to the tournament, and 2) virtually every conference tourney winner was at least good relative to the strength of their conference. The teams on the top seed lines are probably too good for a repeat of last year's high volume of first-round upsets, but the small schools did their part and should at least be pretty damn competitive - especially in the 5-12 and 4-13 matchups.
As always, I've only listed the mid-major conferences with either 1) legitimately good teams, or 2) one or two teams separated from the rest of the pack. For example, I don't really care that the top seed didn't win the Summit tournament, because there was a big clump of teams in the 150-200 range that didn't do much to separate themselves.
With that complete, onto the brackets!
EAST
If you base your evaluations strictly on pre-season potential, you might see the East as the toughest region. Villanova, Duke, Virginia, and Wisconsin were all seen as title contenders, and a few other teams weren't far behind. And since the regions are incredibly balanced this year, I wouldn't blame you for looking at the last two champions on the top two lines and calling it a day. That said, there's enough peculiarities and similarities to make me think this region could be more up for grabs than you might think. For one, Duke and Villanova have the same set of weaknesses: bad rim protection (both teams are sub-top-100 in defensive block rate), poor defensive rebounding, and lack of depth. Obviously, these teams compensate in other ways - Villanova might be the smartest offensive team I've ever seen and never commits fouls on defense, while Duke has probably the most raw talent in the field. But if one of the lower seeds can exploit these weaknesses, that team might just be able to collect a big pair of wins and make it to Phoenix.
Best First-Round Matchup: Virginia vs. UNC Wilmington
Virginia is probably the best 5-seed this year, so there's a chance they destroy everyone in their path for at least a couple of games. That said, I really want to see how a great defense deals with the nation's top two-point shooter (79%!), Devontae Cacok. The Seahawks are also great at holding onto the ball and rebounding their misses, so there's a chance they simply shoot the ball more than a struggling Virginia offense, which means anything can happen. Wilmington is not particularly deep (not that many mid-majors are), but Virginia is so bad at drawing fouls that this might nullify that weakness. I'll feel for Tony Bennett if is team has another disappointing March, but I also love chaos, so this is a coin flip for me.
Best Potential Matchup: Duke vs. Marquette or Villanova vs. Virginia Tech*
A couple of the best three-point shooting teams up against the top teams in the bracket in the round of 32? Yes, please. My most reliably favorite tournament games tend to be round of 32 matchups involving top teams, and while this tournament is full of exciting possibilities in that realm, these might be the two best.
*Look at Jay Wright and then look at Buzz Williams and then contemplate that these men are both at the top of the same profession
The Pick for Phoenix: Duke
I hate this pick, but I think Duke has enough offensive versatility to get through the slew of great defenses in front of them. SMU and Virginia are the best under-the-radar options to put everything together and go on a run, but when in doubt, go with the preseason #1.
MIDWEST
This is the weirdest bracket of the year. Kansas is probably the weakest one-seed, but they draw a few programs (ISU and Purdue) that have disappointed in previous years. Oregon and Creighton are missing important pieces to their teams but have shown enough without them to earn some trust. And the bottom quartet of the bracket is as strong as any. Also, Tom Izzo is lurking around as a nine-seed with 14 losses. Very, very bizarre.
Best First-Round Matchup: Michigan vs. Oklahoma State
There legitimately might not be a better matchup of offenses in the whole tournament than this first-round game. This oddity is partly on the committee for seeding these teams a scoche too low, but it's also on the generous defenses that let these teams down too often. The Cowboys do have one defensive strength (forcing turnovers), but Derrick Walton's sure handedness should cancel that out. What does swing my estimation slightly in favor of Oklahoma State is their ability to crash the offensive boards. That said, either team can simply shoot their way to victory, so this should be a fun one.
Best Potential Matchup: Kansas vs. Oregon
Virtually every bracket has multiple teams with awesome collections of guards and small forwards. After all, it's 2017 and this is where basketball is now (thankfully). But there might not be a better set of ball handlers, decision makers, and scorers than in this potential regional final. The idea of Frank Mason, Josh Jackson, and Traveon Graham going up against Dillon Brooks, Dylan Ennis, and Tyler Dorsey is pure basketball porn.
The Pick for Phoenix: Louisville
The weirdest bracket is also the hardest one to pick. Kansas is probably the most vulnerable one-seed, but the top half of the bracket isn't the most challenging. On the flip side, Louisville's going to have a tough road, but their defense is good enough to allow their feisty offense to stick around. I will almost certainly change this pick a dozen times before Thursday.
SOUTH
UNC-Kentucky should be enough to convince you of this bracket's worthiness. Which is good, because outside of a stupidly under-seeded Wichita team, this region underwhelms. UCLA has a lot of the weaknesses (lack of depth, bad defense) that typically doom upstart teams. Butler and Minnesota both have stronger resumes than actual team strength. The matchup of Arkansas and Seton Hall is easily the weakest in an otherwise loaded slate of 8-9 games. But none of that matters. You have North Carolina and Kentucky and Gregg Marshall's unparalleled petulance and that's enough.
Best First-Round Matchup: Minnesota vs. Middle Tennessee
Nothing in this region jumps off the page (although the KSU-Wake matchup in Dayton is pretty nice), so I'll take last year's Cinderella against a team that is nowhere near as good as the one they beat last year. Minnesota's shot-blocking proficiency (3rd in the nation) isn't a great matchup for Blue Raider team that gets a lot of their points around the rim, but if MTSU can capitalize on their one main advantage (rebounding), that could make the difference.
Best Potential Matchup: UCLA vs. Wake Forest
Yeah, I could have picked Kentucky-Wichita (and probably should have). But points are fun and this matchup would have all of them. The UCLA crew rightfully gets a lot of love, but John Collins just might be a better college player than any of them. Random note: I don't know if there is a better possible matchup of coaches in terms of how successful they were in college than Alford against Manning.
The Pick for Phoenix: North Carolina
Go with chalk here and get cute elsewhere. Yeah, UNC's guard depth isn't the best, but Justin Jackson has been amazing, and the team rebounds so well that it might not make a difference. Add in a relatively easy draw, and I expect Roy Williams and friends to get back to the Final Four. My love affair with De'Aaron Fox is almost enough to get me to swap in the Wildcats, but not quite.
WEST
Like every other region this year, the West has its pluses and its minuses. Gonzaga is actually a very strong one-seed (1st in KenPom, 1st in LRMC, 2nd in BPI), and West Virginia is the best four-seed by far. On the other hand, this bracket probably has the worst two, three, and six-seeds. Add in a bizarro Notre Dame team* and a Saint Mary's team that could be awesome or terrible (I have no idea), and you have yourself what is almost certainly the most unpredictable bracket.
*There is no outcome in this region that would be unexpected for the Irish. Win the whole damn thing? Sure! Give up 15 threes to Princeton and lose? Why not! They've won all of the games they should have won, and then played basically even with every other team and won about half of those. Mike Brey knows entertainment, if nothing else.
Best First-Round Matchup: Florida State vs. Florida Gulf Coast
Picking the dunkiest team to play a competitive game against the second-tallest team in the nation may seem counter-intuitive. But the Noles are only pretty good at actually acting like a tall team (22nd in block rate and 232nd in defensive rebounding), so it's not as ridiculous as it sounds. Realistically, I doubt FGCU has an answer for Dwayne Bacon and Xavier Rathan-Mayes, but who cares; There Will be Dunks.
Best Potential Matchup: Gonzaga vs. West Virginia
This would almost certainly be the best Sweet Sixteen matchup in the tournament, as it could go any number of ways. The "Press Virginia" mentality might overwhelm Gonzaga's backcourt, but the Zags' size advantage should play up in the half court offense. Both teams have great depth, so foul trouble shouldn't be an issue. If one player makes the difference, I'd be tempted to go with Jordan Mathews' gorgeous jumper. But the potential for a Zach Collins breakout is what makes Gonzaga so intriguing and so dangerous.
The Pick for Phoenix: Gonzaga
This pick is terrifying, but I'm sticking to my guns and doing it. You can't stop me.
FINAL FOUR
It's tempting to pick an all-Tobacco Road finale, but I'll stick to chalk and say North Carolina over Gonzaga. Deep frontcourts, skilled guards, great coaching, and relatively clear paths make for a easier pick than most years.
As for my dream bracket, it's quite simple: Notre Dame over Creighton. Earlier in the year, I would have slightly preferred the Jays' simply beautiful brand of basketball, but without Mo Watson it's not quite the same. I would love to send Vasturia and Beachem off with the best run in Notre Dame hoops history. Let's hope it happens.
