Making sense of the first round of Madness
This was the payoff. This was the reward waiting at the end of the murkiest college basketball season in recent memory, one in which the No. 1 ranking changed hands six times, where the consensus was that there were no great teams, and where even one of the consensus title contenders turned out to be far more vulnerable than anyone realized.
After four-and-a-half months that yielded little clarity, 68 teams were thrown into a single-elimination bracket, and it took just two days for it to devolve into total madness. A tournament-record 10 double-digit seeds won first-round games, and Friday’s victors included teams seeded Nos. 15 (Middle Tennessee State), 14 (Stephen F. Austin), 13 (Hawaii), 11 (Northern Iowa) and 10 (VCU, Syracuse). The lone letdown was Holy Cross getting routed by 39 points by Oregon in the West Region’s 16-vs.-1 game. The point of a daily NCAA tournament recap is to make sense of things, but making sense of March 18, 2016, is not possible. All we can do is cherish the carnage:
1. Put Northern Iowa in Oklahoma City, and magical things happen. Six years after Ali Farokhmanesh shot down No. 1 seed Kansas in the second round at Chesapeake Energy Arena, the No. 11 Panthers found a new hero—and created an even better highlight. With his buzzer-beating, half-court game-winner against No. 6 Texas in Friday’s nightcap, Northern Iowa senior guard Paul Jesperson trumped Little Rock’s Josh Hagins for Shot of the Tournament ...
... and had shades of Arkansas’s U.S. Reed beating Louisville in the 1981 tourney (h/t @davevcu):
Jesperson’s shot was gorgeous—despite his use of the glass—but don’t let the formula of 12-loss No. 11 seed + upset of a major-conference opponent + alltime NCAA tournament buzzer-beater fool you into thinking Northern Iowa advanced on magic alone. The Panthers have the most prominent case of split-personality disorder in the entire bracket. When they face teams ranked outside the top 40 in efficiency on kenpom.com, Northern Iowa plays at a level that would make them the 76th-best team in the country—the equivalent of a UCLA or Northwestern, neither of which is in this bracket. But when the Panthers face a top-40 opponent, they transform into a team that would rank 11th in the nation in efficiency. Just to be clear—not an 11-seed, but rather the 11th-best team in the country, with wins over Texas, North Carolina, Iowa State, Stephen F. Austin and Wichita State (twice). Take a look at the splits, which I’ve adjusted for competition and location:
There really isn’t much separating Versus-Legit-Competition Northern Iowa from the Nos. 1 (Oregon), 2 (Oklahoma) and 3 (Texas A&M) seeds left in the West Region. The Aggies, who have Northern Iowa up next, might want to consider cloaking themselves in the jerseys of a mid-to-low-level Missouri Valley Conference team on Sunday, in hopes of coaxing out the Bad Panthers.
2. The best measure of Friday’s absurdity is that a halfcourt buzzer-beater was far from the most seismic event. The team responsible for that was No. 15 seed Middle Tennessee State, which came into the day ranked No. 123 in kenpom.com, with its highest-quality wins coming over Old Dominion—and the Monarchs’ only postseason bid is to a new tournament called the Vegas 16. Middle Tennessee picked up a slightly higher-quality win in St. Louis: it upset No. 2 Michigan State 90–81, knocking out a team that was Vegas’s pre-tournament co-favorite along with Kansas and North Carolina.
There was absolutely no way to see this coming. Middle Tennessee played two NCAA tournament teams in the regular season, and in those two games, its offensive efficiency was 76.4 (in a loss to VCU) and 83.4 (in a loss to South Dakota State). If you’re a newcomer to tempo-free stats, scoring 80-ish points per 100 possessions is abysmal. So what was the Blue Raiders’ efficiency against Michigan State, which had a top-25 defense? An unreal 132.3, the highest of any Spartans’ opponent this season. A Tom Izzo team that was widely forecasted for Houston got hit by a bolt of lightning on Friday, as Middle Tennessee shot 57.9% from three and 55.0% on its twos.
No. 15 seeds have won nine games in the expanded NCAA tournament era, and what the Blue Raiders did offensively is the second-best performance* of that nine, with only Norfolk State’s 2012 upset of Missouri featuring more improbably accurate shooting:
(* Pre-2000 efficiencies estimated due to lack of offensive-rebounding and/or turnover data)
Place Middle Tennessee’s offensive performance in context of the other No. 15s from 2016 and ... wow. The Blue Raiders were 30.5 points per 100 possessions better than the second-best No. 15, and 49.6 PP100P better than the worst No. 15:
3. An analytic case can be made for Norfolk State-over-Missouri 201 as the biggest upset in modern NCAA tournament history, and I’m obviously a believer in analytics, but I’m convinced that Middle Tennessee-over-Michigan State is bigger. The quality gap between Norfolk State and Missouri may have been more significant, but few analysts had faith in Mizzou as a potential title-winner.
The Tigers’ then-coach, Frank Haith, had zero NCAA tournament head-coaching experience, and they entered that tourney ranked 107th in adjusted defensive efficiency. The lowest any eventual title team has ranked in that metric, coming into the tournament, is 57th (Duke, in 2015), and the median ranking for eventual title teams since 2002 is 12.5. All of which meant that 2012 Mizzou was viewed as ripe for getting upset—just not upset in the first round.
Michigan State, on the other hand, had the nation’s best all-around player in senior Denzel Valentine, the most respected tournament coach in Izzo, and it ranked 23rd in adjusted defensive efficiency when the brackets were announced. The Spartans were perceived as being as safe a Final Four pick as the No. 1 overall seed in the field, Kansas. And now they’re gone—inexplicably gone after just 40 minutes of postseason basketball. The only thing that will trump Middle Tennessee-over-Michigan State is when a 16 finally knocks off a No. 1—and the only other first-weekend upsets I can recall with similar bracket-busting impact are Kentucky over Wichita State in 2014 and Farokhmanesh’s Northern Iowa over Kansas in 2010.
4. With Michigan State out of the field, we need to re-assess the paths to Houston for the No. 1 seeds. On Selection Sunday, Virginia—due to its potential Elite Eight game against the Spartans—was viewed as having the most difficult draw. Now, the Cavaliers seem to have the most favorable road to the Final Four. Even if their region goes according to chalk, their highest-ranked potential opponent in efficiency is Iowa State, at No. 18, and the Cyclones are one of just four teams remaining with a defense ranked outside kenpom.com’s top 100 (the others are Butler, Notre Dame and St. Joseph’s).
Here’s how I rank the No. 1s’ remaining roads from easiest to most difficult, with their chalkiest possible opponents in parentheses:
1. Virginia / Midwest (No. 9 Butler, No. 4 Iowa State, No. 3 Utah)
2. Oregon / West (No. 8 St. Joseph’s, No. 4 Duke, No. 2 Oklahoma)
3. Kansas / South (No. 9 UConn, No. 5 Maryland, No. 2 Villanova)
4. North Carolina / East (No. 9 Providence, No. 4 Kentucky, No. 2 Xavier)
The Tar Heels have the toughest road mainly because Kentucky is still in the way. The Wildcats have even more talent than UNC, including possibly the best backcourt in the bracket. They’re a No. 4 seed that analytically resembles a No. 2, and they’re peaking in March, which could make them even scarier ... but in this tournament, it might not mean a damn thing.