College basketball experiencing a backlash to the Curry Effect

College basketball experiencing a backlash to the Curry Effect

By now, college basketball’s embrace of the three-point field goal has been well documented. More than 38 percent of attempted field goals in Division I this season was from beyond the arc, an astronomical 15 percent change from just five seasons ago, when the DI average for three-point field rate was just a third of all attempts.

What began in the NBA with the arrival of pace-and-space and the impact of players like Steph Curry has filtered down to the college game, and players who were in grade school when Curry’s Davidson shocked March Madness have grown up mirroring the guard’s perimeter exploits.

Take the small sample size of the past five final fours: since the 2015 tournament, which was the first season in which the spike of 3FGA percent became noticeable (and a several year acclimation period after the line moved back to 20’9”), 12 of the 20 finalists posted rates that topped the DI average.

It wasn’t limited to perimeter-laden offenses like Jay Wright’s at Villanova—teams like Wisconsin and Kansas, which had famously eschewed long-range attempts within the halfcourt offense, started to expand the boundaries of what their offense’s capabilities.

The Curry effect is in full swing: DI teams attempted nearly 39 percent of its attempts from deep in 2019, and as the percentage has continually risen since KenPom began tracking the stat in 2002 (a 2 to 3 percent change every season), college basketball will reach 40 percent by 2021.

It’s become almost normcore to talk about college basketball and three-pointers in the same breath—for example, a recent article on Wofford noted the phenomenon of the Curry Effect. And while the Terriers are indeed counting on its perimeter accuracy to upend March Madness, Fletcher Magee also suits up for the squad, and the guard has attempted more than 1,000 threes in his career and became the NCAA’s all-time in made threes earlier this season.

But what if the rest of college basketball is starting to experience the beginnings of a Curry Effect backlash?

Yes, the percentage of attempts is at an all-time high, but for the first time in several seasons, accuracy has dipped—to 34.4 percent (the previous lows, which hovered at 34 percent, were in 2014 and 2015).

According to data pulled by BartTorvik.com, there is an increasing trend of players attempting threes who frankly shouldn’t be taking those shots: of players in 2019 who attempted 10 or more threes, 42 percent converted at a rate of 35-plus percent (equaling at least 1.05 points per shot).

Conversely, nearly 60 percent attempted a three and connected on less than 35 percent, which is a disturbing amount of perimeter inefficiency.

College basketball’s Curry Effect was fueled for years by teams looking to stretch an opponents’ halfcourt defense, and create the necessary spacing to take advantage of angles and driving lanes (the implementation of the hand check rule before the 2013-14 season also helped this boon). But, this offensive freedom also meant that more and more players who likely shouldn’t be taking threes have suddenly been given the green light.

Some teams are taking advantage of this subtle shift: both Virginia Tech and Houston rank in the top 25 of KenPom’s defensive efficiency database, holding teams to well under 1 PPP, but each squad is among the worst in Division I in defensive 3FGA percent. Houston allows opponents to take 41.1 percent of its shots from deep, and the Hokies’ rate is even starker—more than 50 percent. The statistical juxtaposition is startling, but both Buzz Williams and Kelvin Sampson are capitalizing off bad shooters taking more threes.

Against Purdue in late November, the Big Ten squad converted 42 percent of its threes, but Virginia Tech focused on limiting looks for Carsen Edwards and Ryan Cline—easily the Boilermakers top two perimeter threats—and the guards made just 33 percent (7 of 21) in the Hokies win. Williams and his staff could live with Grady Eifert, Evan Boudreaux, Sasha Stefanovic attempting nine three-pointers, and even though the trio shot 66 percent, that was fine. Better they shoot 66 percent than Cline and Edwards heat up. Another example of this philosophy in action occurred during Virginia Tech’s win vs. Duke: sure, Zion Williamson’s unavailability hurt the Blue Devils, but so did Tre Jones and RJ Barrett shooting 2 of 10, and the majority of those attempts were wide open looks.

Similarly, when Houston played Connecticut in mid-February, Sidney Wilson and Tarin Smith attempted nearly 40 percent of the Huskies’ three-pointers, and the duo combine to convert sub-25 percent of their perimeter looks in 2019.

And perhaps this regression is also contributing to the escalation of teams seeking out attempts at the rim. Shots at the rim are the most efficient in basketball—teams score 1.20 PPS in 2019—and per BartTorvik.com, DI squads have attempted 35.7 percent of their shots around the basket.

This is the past decade’s third highest rate (again, the first was 2014, following the tightening of hand check rules), and begs the question: what’s more effective, a wide open three from a player shooting below 35 percent, or a rim attempt? Will teams attempt fewer threes next season? Or, rather, will teams simply become more restrictive of which players take those shots?

What’s interesting is that of the dozen teams projected as final four contenders, just three attempt a higher 3FGA percent than the DI average.

Forget Villanova’s combined 39 percent from deep in 2016 and 2018 or even North Carolina’s 37 percent from deep in 2017: this might just be the first national semifinal in some time that skews away from the perimeter.