Threes, luck, and volume in the NCAA tournament


March 1993: Jud Heathcote announces that Tom Izzo will be his successor as head coach at Michigan State. (Lansing State Journal)

Michigan State was the final team to reach the 2019 Final Four, thanks to a Kenny Goins three with 39 seconds remaining against Duke. By virtue not only of Goins’ heroics but also the fact that, on the same afternoon, Auburn beat Kentucky in overtime, we now know that using one-and-dones in college basketball doesn’t work.

With that question settled once and for all (I’m kidding; apparently that needs to be indicated), let us turn our attention to the gathering of old geezers in Minneapolis.

If we think of said geezers as four offenses and four defenses, one thing to be said about the collective is that, with the possible exception of the Texas Tech offense, all of these units are used to seeing three-point attempts — both for and against — flying every which way in the tournament. Indeed, Ken Pomeroy noted last week at The Athletic that the NCAA tournament has become strikingly perimeter-oriented these last few years.

From the First Four through the Elite Eight, fully 40.6 percent of the shot attempts in the 2019 NCAA tournament have, by my reckoning, been launched from beyond the arc. Perhaps it should come as no surprise, then, to find that every Final Four team that did not have to face Carsen Edwards has arrived in Minneapolis carrying a really gaudy number for three-point defense.

The NCAA tournament can be a three-point defense lottery
Tournament games only

                 Opp. 3FG%
Texas Tech         23.4
Auburn             26.9
Michigan State     28.2
Virginia           39.2            

Conversely, there are no great two-point tournament defenses in the Twin Cities. Texas Tech of course carries that reputation, but Red Raider opponents have in fact connected a very normal 48 percent of the time inside the arc during the tournament. Meanwhile, Virginia’s interior defense has been pretty good the last four games (45 percent), short-handed and somewhat diminutive but blazing fast Auburn’s been about as you’d expect (57) and Michigan State is somewhere in between.

As for the impossibly low number Chris Beard’s defense is carrying for opponent three-point accuracy in the tournament, the first thing to be said about it is that it’s really not impossibly low. Holding opponents to just 23 percent outside the arc over the course of a mere four games happens every day of the week during the season. Anyway, if you’re looking for ostentatiously great three-point defense over the course of a four-game run to a Final Four, any anguished Indiana fan will tell you that Syracuse 2013 is still the gold standard there: 15 percent.

Who knows, maybe all these great three-point tournament defenses will at last regress toward the mean at the Final Four. It happened to Michigan last year. The Wolverines arrived at their national semifinal having held tournament opponents to 26 percent shooting beyond the arc and kept the mojo going for a bonus 40 minutes against Loyola Chicago before, finally, Jay Wright’s eager charges rained on the parade with their normalcy (10-of-27).

That said, the next 40 to 80 minutes of action does not give regression much runway with which to work. Besides, aberrant outcomes like crazy-bad opponent perimeter shooting or even, say, a No. 16 seed beating a No. 1, are part of why we love the single-elimination format.

Could we revisit the question of that format if three-point attempt rates keep charting a steep increase? What happens when, after the line (one presumes) is moved back but then attempt rates eventually resume their climb toward the 50 percent inflection point, we have a tournament wherein the defense has relatively little control over the success or failure of better than half of opponents’ tries?

Hold that thought until 2029 or so; meantime, don’t fret. Even back in the old days when there were way fewer three-point attempts in the tournament, circa 2015, we still had Final Fours made up mostly or even entirely of teams kissed by the hoops gods when it came to win-or-go-home perimeter D.

The NCAA tournament can be a three-point defense lottery (redux)
Average opponent three-point accuracies of Final Four teams
Pre-Final Four tournament games only

                 Opp. 3FG%
2015               28.8
2016               32.7
2017               32.7
2018               29.5
2019               29.7

Besides, Virginia’s still standing in 2019, somehow, even after greatest per-possession three-point shooting display by an opposing player in NCAA tournament history.

                            Game              Ind
                    3FGM    poss.    Min.     poss.     3FGM/poss.
J. Freyer, LMU 
vs. Michigan, 1990   11      108      32       86         0.127

C. Edwards, Purdue
vs. Virginia, 2019   10       62      44       60         0.166

The Cavaliers, by a hair, haven’t shot as accurately as their tournament opponents, but what Tony Bennett’s guys do bring to this party is shot volume in big historic quantities. O, the irony.

After all, the Purdue staff is on the record as tracking this shot volume stuff, but in the Elite Eight the Boilermakers were hoisted by their own petard. The Hoos gave the ball away on just eight percent of their possessions while rebounding 43 percent of their misses. On an evening when Virginia’s shooting accuracy was nothing special (and Purdue’s was, obviously, outstanding), it was a remarkable display of frequent scoring attempts, as though 2015 Wisconsin and 2017 North Carolina had formed a supergroup.

Normally, Final Four teams aren’t distinguished by their tournament shot volumes and, in particular, last year’s group was emphatically lackluster on that front. This year, on the other hand, all four teams did get to where they are in part by creating more scoring chances than their opponents. Actually, in the Cavaliers’ case, make that way more.

The NCAA tournament can be a shot volume showcase
Shot volume index, tournament games only

                      SVI     SVI     Margin
Virginia             102.6    91.2     +11.4
Texas Tech            93.9    89.7      +4.2
Auburn                98.8    94.9      +3.9     
Michigan State        99.1    96.0      +3.1           

Fluke or harbinger of a sea change as future tournament teams rapidly deploy to boost their shot volumes? Definitely a fluke, but an instructive one. Look at the Red Raiders, for example. Note the low shot volume, but also the ridiculously low volume for opponents.

Now, finally, I understand why the descriptions about less basketball in 40 minutes that are usually directed Virginia’s way always felt equally if not more apt for my experiences watching Tech games despite what on paper is a more “normal” pace. There’s just a really small number of shots for both teams in Red Raider games, regardless of the tempo.

So, yes, do remember the number 23 with Beard’s guys. Not only are tournament opponents hitting 23 percent of their threes, those same teams have also given the ball away on 23 percent of their possessions. To have a shot, Michigan State and, if it comes to that, Auburn or Virginia will want to come in higher and lower than 23 percent, respectively.

Speaking of the Hoos, I’ll be watching to see if they set a record for tapping the ball back for offensive boards the way Mamadi Diakite so famously did at the end of regulation against the Boilers. Diakite and Jack Salt were doing that on numerous occasions this season before The Rebound, and, if nothing else, it’s one way for teams to get second chances while still keeping their guards in proper Missouri-Valley-style-paranoid position for transition defense.

In fact, if Virginia wins it all, I suppose there’s a chance tapping the ball back will be the cool stylistic “it” thing for 2019, a la threes for Wright in 2016. Regardless, win or lose, rounding up torches and pitchforks to insist that a slow pace inevitably leads to March disasters now feels very 2018. That in itself is still another notable outcome produced by the 2019 NCAA tournament.