NCAA Weekly, 14.1: The biggest story of the 2022-23 season? Shot volume

at the top of the sport, a return to tradition (sort of)

For most of the last decade I’ve found myself watching games somewhat differently thanks to the work of John Gasaway. He’s very good at getting someone to see how statistics within the game can be interesting and useful without being boring about it. (This is also a good summary of the goal of this site.) But what Gasaway first designed in 2016 and refined later on fascinates me:

It’s hardly news that programs like UK and UNC are excellent at offensive rebounding, but what’s changing is the willingness and ability of elite teams to bring their turnover rates down to the same level as all those plucky yet disciplined underdogs they’re always trampling underfoot. Taking care of the ball and rebounding your misses is a good way to safely navigate your way around an off night on the perimeter.

The original equation was Shot Volume = 100 + OREB% - TO%. The new one is more complicated but a little more accurate (presumably) and involves the idea of 100 possessions where there are no free throw attempts. The point is this: shot efficiency, AKA actually making the shots, is great. That’s the first step. The differentiator, especially on nights when your team isn’t shooting well, is shot volume. Can you find a way to make up for poor shooting? If so, you’re going to be harder to put down than a team that doesn’t.

Largely the original formula (deemed Easy Shot Volume Solver by Gasaway) has held up well since its unveiling. No champion in the last 20 years has had a Shot Volume offensively below 112.3 (2015-16 Villanova), and the average national champion during that time sat at 118.3. This was important beyond the obvious reasons. Considering the national average generally hovered around 112 and is now down to 110, you could make the argument that these teams were consistently getting 6-8 shots more per game than the average team they played. That’s remarkable.

However, that’s just the offensive side. There’s not been a ton of work done on defensive shot volume. So, given all the data I’ve accumulated over the years, I figured it would be worth a spin. Let’s get the disappointment out of the way first: defense doesn’t correlate much at all to NCAA Tournament wins, and offensive shot volume remains the king. However, in the last few years, a simple equation of Offensive Shot Volume minus one’s Defensive Shot Volume has proven to be equally as correlated to NCAA Tournament wins. How much shots you create when you have the ball still matters, but how few shots you allow when you don’t also has an impact.

This matters to 2022-23 specifically, which has quietly become The Year Shot Volume Left Its Flop Era. (Or whatever it is children say.) The top teams in men’s college basketball this season are better than they’ve been at any other point in the 68-team era in one specific aspect: shot prevention. Also, they’re better than ever before in our little Overall Shot Volume metric.

Offensive shot volume sits at a perfect 12-year average of 117.3 among the upper echelon of the sport; nothing shocking or unusual is really afoot there. What is unusual is how well teams like Houston, Tennessee, and UCLA are forcing turnovers and preventing second chances when they don’t have the ball. Among the current top seven teams nationally on Bart Torvik’s site, six (Alabama excluded) make up the top six Shot Volume Overall teams in America. It’s a surprising level of dominance in a year where the best offenses are generally paired with bad defenses.

What could this mean for March? I think it’s best to avoid seeing this as a full trend and instead look at case-by-case scenarios. In my experience, this equation tells you more about who not to pick in March than it does who you should pick. And even in this very volume-heavy season, there’s some teams that stand out in negative and positive ways.

Among future high seeds, Alabama and Kansas stand out…in not-so-fun ways

Depending on what site you go to, there looks to be roughly six serious contenders remaining for the 1-seed line. Alabama, Kansas, and Houston are locks or near-locks, while the final spot will be a battle between Purdue, UCLA, and Texas. Baylor certainly could get there, but for the sake of this argument, we’re leaving them out because I simply cannot believe in their defense. Sorry!

Anyway, that’s a six-team field that remains. Four of these teams shape up quite nicely via the traditional Shot Volume numbers. Houston plays in a weak AAC, but with a 123 OffSV and a +17.3 OvrSV, they’re tracking for the highest Shot Volume rating by a non-Power Six team in my 20-year (‘01/’02-pres.) database and the second-highest overall number behind 2016-17 North Carolina. Purdue offers a bad turnover margin of -1.9 per 100 possessions, but their sheer rebounding dominance and shot prevention on defense is enough to give them good-looking numbers. UCLA is second-highest this year at a +14.8 OvrSV and quietly has a +9 turnover margin per 100 possessions. Texas is more flawed than these three but rates out well in shot prevention.

That leaves two, and the two teams may well be the top two overall seeds. Alabama, despite…everything, is the odds-on favorite to be the #1 overall seed. They’re obviously a great team, and they have a lot of good they bring to the table. What’s most important, though, are the turnovers. Alabama grades out well as a rebounding team, but with a -3.2 turnover margin per 100 possessions, they’re on track to have the second-worst turnover margin ever for a 1 seed, only in front of 2001-02 Cincinnati. Of the 14 1 seeds to enter the NCAA Tournament with a negative margin since 2002, only six even made the Elite Eight and three made the Final Four.

Kansas, meanwhile, has the reverse problem: rebounding. The Jayhawks are barely above water at a +1.1 rebound margin per 100 chances. That would be the seventh-worst rate for a 1 seed since 2002. Their overall shot volume in general just isn’t impressive, though; it sits at a 112.3 OffSV, which would be the second-worst of all-time for a 1 seed. The good news for Kansas: the current second-worst is 2018 Kansas, who made the Final Four. The bad news: that’s the only 1 seed below a 114 OffSV to have made the Final Four in the last 20 years.

Among potential 2-4 seeds, Consider Tennessee (and UConn)

Yes, I recognize the ridiculousness of that statement, given that this is a Tennessee website at heart. But! Tennessee quietly generates a ton of shot attempts and does a great job at preventing them on the other end. Haslametrics rates Tennessee out as 8th-best in America offensively at field goal attempts per 100 possessions, while they’re 22nd-best on defense. They’re one of two teams in the entire nation in the top 25 of both stats; it’s no wonder that the Easy Shot Volume Solver Volume 2 rates them out as a top-three team nationally. (This is also why Tennessee wins as many games as they do despite nasty shooting.)

Tennessee’s query: teams with a >5% OREB% margin, a +3 turnover margin, and a Shot Volume between 116-122. 21 2-4 seeds came back; 14 (67%) made the Sweet Sixteen, while nine (43%) made the Elite Eight. (Normal rates for 2-4 seeds: 55% and 16%.)

Meanwhile, UConn is a true throwback to the 2000s. The Huskies rebound an astounding 39.1% of their missed shots, the top rate in all of basketball, and are at least treading water with turnovers despite some offensive struggles. Like all teams, UConn has a serious flaw - foul trouble - but Connecticut is quietly just as good as Purdue at pumping up shot after shot on offense. It’s why UConn actually has a winning record (7-5) in games they’ve posted a sub-50% eFG% in.

UConn’s query: an OREB% of 37% or better, an offensive Shot Volume of 118+, and an OvrSV between 10 and 20. The results: 21 teams, 12 Sweet Sixteen, but notably, eight Elite Eight teams and five Final Four ones.

Among potential 2-4 seeds, Consider Not Considering Kansas State and Marquette (and Indiana)

This section is less desirable to write. All three of these teams are really fun stories that I’ve enjoyed following, and all three possess the ability to make me look very, very foolish. But unfortunately, this is as much a history website as it is a statistics and opinion website. History does not favor these three.

For one, all three have a sub-112 OffSV, which has traditionally belied an early exit. Since 2002, there’s been 22 2-4 seeds with an offensive Shot Volume number below that range. Exactly one has made the Elite Eight or further: 2019 Texas Tech, who merely had the best defense in modern college basketball history. In fact, only six (27%) have even reached the Sweet Sixteen, which is a huge deal when you consider 2-4 seeds should expect to make the second weekend simply by being a 2-4 seed.

Each team has their own Achilles heel, though. Kansas State stands out by having a weak OvrSV of +2.6 (Top 25 average: +7.6), but their biggest problem is having the highest TO% of any Top 25 team at 19.7%. Starting at 2014 (when the TO% across CBB finally dipped below 20% and never looked back), there’s been 12 2-4 seeds who’ve entered March with an offensive TO% of 19% or worse. None got past the Sweet Sixteen, and only four even got that far.

Marquette’s problem is a very obvious one: they’re terrible at rebounding. It’s by choice, because Marquette’s roster is very much scoring-first and doesn’t have much in the way of height or post fortitude for rebounding purposes. That said, it’s still a huge problem. Marquette sits well below water at -4.6 per 100 rebounding chances. That’s the worst rebounding rate on record for a 2-4 seed since 2011-12 Michigan, who bowed out in the Round of 64. Even filtering Marquette into a group that includes teams with similarly great turnover margins/low OvrSV scores didn’t help; teams with a 108-112 OvrSV and a negative rebounding margin made the Elite Eight or further once in 19 tries, with 13 bowing out before the Sweet Sixteen.

Indiana’s the most problematic case. Not only is their OffSV weak at 111.5, their OvrSV of just +0.7 is tied with Creighton as the lowest of any Top 25 team this season. I ran a query for teams like Indiana in my database: OffSV between 108 and 113, OvrSV between -2 and +2, a negative turnover margin, and a seed between 3-5 (Indiana’s expected range). The results: 14 teams, two Sweet Sixteen bids.

Other research goodies

• Teams with an offensive Shot Volume below 110 are 1-for-169 in making the Final Four. The only exception is 2017-18 Loyola Chicago, who won their first three games by a total of four points and were an outlier in many regards. (They’re also just 3-for-169 in making the Elite Eight, and no team seeded 1-7 has gotten past the Sweet Sixteen.) Top 50 teams that meet this unfortunate metric: Creighton (108.9), Boise State (109.4), Liberty (109.1), Penn State (104.8).

• Top 4 seeds with an OvrSV of 13 or higher: 24-for-31 in at least making the Sweet Sixteen, 19-for-31 in making the Elite Eight. That seems notable in this particularly topsy-turvy year. Teams that look >50% to be top 4 seeds and have a 13+ OvrSV or better: Houston (17.3!), UCLA (14.8), Tennessee (14.7), Purdue (13.5), and Connecticut (14.2). Statistically, one of these five won’t get to the Sweet Sixteen and two the Elite Eight, but that’s still three of your eventual Elite Eight teams. Maybe. We’ll see.

• Being a top-end seed with a bad turnover margin: generally not good. 1-4 seeds with a turnover margin of -1 or worse per 100 possessions are 6-for-57 in making the Final Four and 12-for-57 in getting to the Elite Eight. Generally, those who have overachieved have either had an extremely positive shooting split (average of +9.7% on eFG%) and have owned the offensive boards (average of +6.8 per 100 chances). Some higher-end teams with negative turnover margins: Alabama (-3.2), Purdue (-1.9), Arizona (-1.3), Xavier (-1), and Indiana (-1). Purdue at least has the offensive rebounding to overcome it, potentially, while Alabama has a very positive eFG% margin. We’ll have to wait and see.

• No, there’s no magic stats for 5-8 seeds that go far. Except… The best strategy is simply to assume that any 9-11 (or further) seed that makes the Final Four is, indeed, a giant outlier. That’s basically always true. Only once in the KenPom era have any of them been a top 25 team entering the tournament: 2006 George Mason. Otherwise, it’s all purely random.