The 3-point line in men's basketball was moved from 20 feet, nine inches to the international distance of 22 feet, 1 ¾ inches this season to create more offensive spacing by forcing defenses to guard more of the court and to slow the trend of 3-point shots while still keeping them an important part of the game.
The 2019-20 college basketball season is roughly a month and a half old, which means there's enough data available to make some sound analysis about the effects, however noticeable or not, of the extended 3-point line.
I looked through 3-point data from all 353 Division I teams, plus historical data, to analyze the affects of the extended 3-point line and which ones are even noticeable.
First, let's look at some of my findings and then dig deeper into the impact of the new 3-point line.
- There are more higher-percentage-shooting teams and more lower-percentage-shooting teams than season. Basically, it's harder — or at least, less likely — to be an average 3-point shooting team this season. Isn't that weird? Actually, not really, there's an explanation for it.
- What if I told you that one of the best 3-point shooting teams in the country last season, which won multiple games in the 2019 NCAA Tournament and nearly pulled off a massive upset in the second weekend, basically lost its entire starting lineup and coaching staff, yet is making even more of its 3-pointers this season even though it's a harder shot? Keep reading to find out who it is.
- Oddly, roughly half of the 25 teams that have best-adjusted to the extended 3-point line are generally located in a similar region of the country.
A much larger variance across the sport
Last season, the gap between the best 3-point shooting team in the country (Lehigh at 42.3 percent) and the worst 3-point shooting team in the country (Jackson State at 27.9 percent) was 14.4 percentage points.
Through the games played Dec. 18 of this season, there are 36 teams with a worse 3-point percentage than Jackson State had last season, including reigning national champion Virginia (25.4%), currently ranked No. 9, and No. 6 Kentucky (27.5%).
The gap between this season's best 3-point shooting team, Saint Mary's, which is coming off of a 40-point, neutral-site win over Arizona State on Wednesday night, and the No. 353 3-point shooting team, Florida A&M, is more than 60 percent larger than the first-to-worst gap last season.
The difference between Saint Mary's (45.7 percent) and Florida A&M (22.1 percent) is 23.6 percentage points — a gap that's larger than the Rattlers' actual 3-point percentage.
If you doubled Florida A&M's 3-point shooting percentage, it'd still be lower than that of Saint Mary's.
MORE: The 'championship arc', forecasting Final Four contenders
While it's a long season with lots of basketball left to be played, most notably with upcoming conference games that should provide a tougher level of competition and small decreases in overall shooting percentages in top conferences, there are currently three teams that have made a higher percentage of their 3-point attempts this season than last season's national leader Lehigh did in 2019.
Saint Mary's (45.7%), Coastal Carolina (42.8%) and Marquette (42.4%) are currently ahead of the 2018-19 Mountain Hawks' national-best 42.3 3-point percentage.
Perhaps a better reference point is the number of teams that are shooting at least 40 percent from three.
Just five teams did last season: Lehigh, Wofford, South Dakota State, Grambling State and Fairleigh Dickinson.
As of Thursday morning, there are 15 teams north of 40 percent: Saint Mary's, Coastal Carolina, Marquette, Lafayette, South Dakota, Ohio State, Northeastern, Portland, Oregon, California, Virginia Tech, Richmond, Indiana State, Sam Houston Start and Georgia State.
So the number of 40-percent 3-point shooting teams has tripled so far — once again, that number will likely decrease over the course of conference play — but so far, there are actually more, better 3-point shooting teams this season compared to last, just as there are also more teams that are worse at outside shooting this season.
The scatter plot below shows the 3-point percentage of every Division I team in both the 2018-19 season and 2019-20 season, listed from the lowest percentage in each season to the highest.
The trend line on the scatter plot serves as a valuable tool and helps back up our point that there are more high-accuracy teams and more worse-shooting teams this season.
You'll see that around the 40-percent mark on the y-axis, which represents the current season, there start to be more points located above the trend line, showing a relative increase in 3-point percentage among the national leaders compared to last season. For example, this season's best 3-point shooting team, Saint Mary's, has a better 3-point percentage than the best 3-point shooting team last season, Lehigh, so the top point is above the trend line.
Likewise, you'll notice a number of points that are below the trend line at the bottom of the scatter plot, showing that this season's worst 3-point shooting teams have made a lower percentage of their threes than their 2018-19 peers (e.g. the team ranked 353rd in 3-point shooting this season has a lower 3-point percentage than the team that was ranked 353rd last season).
This data speaks to the separation that the extended 3-point line has created between the haves and have nots of 3-point shooting. It makes sense, at least anecdotally. The 3-point line was moved back almost 17 inches in the offseason, which isn't far enough that it makes the three an impossible shot for competent outside shooters, but it's also going to make teams with 3-point shooting deficiencies even more noticeable.
The data backs that up as there are fewer "average" 3-point shooting teams.
Last season, the national average for 3-point percentage was roughly 34.4 percent and there were 156 teams that shot between 33.0 percent and 36.0 percent.
NOTHING BUT NET: The NET, explained: NCAA adopts new college basketball ranking
This season there are just 103 teams whose 3-point percentage falls in that range.
Below are histograms from the 2018-19 and 2019-20 seasons to support the notion that there are fewer average 3-point shooting teams. You'll notice longer "tails" on the second histogram as the 353 Division I teams are less concentrated by 3-point shooting percentage this season compared to last.
As we've established, there's a significant group of teams that are worse at shooting threes than the worst team last season and there are more teams that are making a higher percentage of their threes at the top of the sport.
So it makes sense that the fat part of the bell curve has gotten smaller as the extended 3-point line has made the 3-point capabilities or decencies more apparent for once-average 3-point shooting teams.
New 3-point line, coach and roster, but better results?
As a random aside, give credit to the Hokies, or whatever's in the water in Blacksburg, Virginia. Because in the offseason, Virginia Tech lost its head coach and its top five scorers, each of whom started at least 20 games last season, from a 2018-19 Hokies squad that ranked ninth nationally in 3-point percentage at 39.4 percent and they've actually gotten better from a statistical perspective.
FIRST YEAR FAME: Ranking the top 10 first-year head coaches
Shoutout to coach Mike Young, who previously coached at Wofford when the Terriers were the second-best 3-point shooting team nationally last season, and his new-look Hokies as they've improved to become a 40.6-percent 3-point shooting team this season, despite ranking 349th nationally in experience and 324th in minutes continuity.
That's the equivalent of playing Boggle, finding the word isthmus, shaking up the grid for the next round and immediately identifying the word onomatopoeia.
These schools have improved the most
Twenty-four schools have improved upon their 2018-19 3-point percentage by at least five percentage points, through the games played Dec. 18, led by Portland with an 8.5-percent improvement from 32.9 percent in 2019 to 41.5 percent this season.
The Pilots went 0-16 in West Coast Conference play last season and their offense ranked a distant 10th in the conference at No. 329 nationally but their improved 3-point percentage should allow them to attempt to close the gap with the rest of the conference.
You could make the case the WCC has adjusted to the extended 3-point line as well as almost any conference.
National 3-point leader Saint Mary's is tied for the second-best improvement nationally with an improvement of 7.9 percentage points and BYU has improved by 6.9 percentage points, so 30 percent of the WCC's member schools are in the top seven nationally in year-over-year 3-point improvement.
Also, give credit to the Pac-12: Stanford (+7.9%), Oregon State (+6.9%), Oregon (+5.8%), California (+5.7%), and Arizona (+5.2%). Maybe it's a West Coast thing.
Here are the 24 schools that have improved their 3-point percentage by at least five percentage points from last season to this season, through Dec. 18:
- Portland: 8.5%
- Saint Mary's: 7.9%
- South Dakota: 7.9%
- Stanford: 7.8%
- Ohio State: 7.4%
- Oregon State: 6.9%
- BYU: 6.9%
- Dayton: 6.3%
- McNeese State: 6.2%
- VCU: 6.2%
- Vanderbilt: 5.8%
- California: 5.7%
- Coastal Carolina: 5.7%
- Illinois State: 5.7%
- Oregon: 5.7%
- Richmond: 5.7%
- Toledo: 5.7%
- Portland State: 5.6%
- Air Force: 5.5%
- Nevada: 5.5%
- Wake Forest: 5.3%
- Arizona: 5.2%
- FIU: 5.1%
- Saint Peter's: 5.0%
Other than the seemingly West-of-the-Rockies bump the new 3-point line has provided, it's also interesting that many of the teams are at opposite ends of the spectrum — teams that were already NCAA Tournament teams and those that didn't play in any postseason outside of their conference tournaments.
STRONG RESUMES: These college basketball teams have strong March Madness resumes, according to the NCAA team sheets
Vanderbilt (11th nationally in improvement in 3-point percentage) finished 0-18 in the SEC last season, Wake Forest (21st nationally) finished 13th in the ACC, and Arizona, Stanford and Cal finished in the bottom five of the Pac-12.
Maybe it shouldn't be surprising that those schools have improved the most from three because when you only win a handful of conference games, or go winless in the case of Wake Forest, you likely have more room to improve than most teams.
But the list of the national leaders in improvements in 3-point percentage also includes Ohio State (No. 11 seed in the 2019 NCAA Tournament that advanced to the second round), Oregon (No. 12 seed that advanced to the Sweet 16), Nevada (No. 7 seed) and VCU (No. 8 seed) so already-good teams can have room to improve, too.
The Buckeyes and Ducks are both ranked in the top eight of the latest AP Top 25 poll.
Does 3-point attempt frequency affect percentage?
We graphed every Division I team's change in 3-point percentage from last season to this season, as well as each team's change in 3-point attempt percentage (meaning what percent of their total shot attempts were threes) from 2018-19 to 2019-20, and found that there isn't a sport-wide trend in changes in 3-point attempt percentage consistently affecting 3-point percentage.
That means that teams that, for example, are attempting a lower percentage of threes this season aren't collectively making more (or less) of them, just as teams that are attempting more threes this season aren't collectively making more or less than they did last season.
Sure, there are individual success stories like Ball State, whose 3-point attempt percentage has climbed by 13.9 percentage points this season (meaning for every 100 shots taken, almost 14 more of them are threes this season, compared to last) and the Cardinals are making 4.9 percent more of their threes this season.
James Madison, however, provides an easy counterpoint against a hypothetical sport-wide trend that taking significantly more threes will increase 3-point percentage. The Dukes have experienced an increase in 3-point attempt percentage of 13.8 percent but their 3-point percentage has decreased by 2.5 percent.
So you can't make a blanket statement that changes in 3-point attempts have a 1-to-1 correlation to 3-point percentage.
Similarly, South Dakota's 3-point attempt percentage has decreased by 8.8 percent while its 3-point percentage has increased by roughly eight percent. Cleveland State's 3-point attempts have dropped off by 17.4 percent from last season and the Vikings' 3-point percentage has also plummeted to the tune of 12.8 percentage points lower that 2019.
You can make some sport-wide analysis on changes in 3-point attempts and percentages if you break down the scatter plot shown above into quadrants. The most common change with the new 3-point line is that teams are taking and making fewer 3-pointers this season.
One-hundred and thirty-two teams, or roughly 37 percent of the sport, falls into this category.
Meanwhile, just 59 teams, or roughly 17 percent of Division I, are taking and making more threes.
|Changes||Number of Teams||Percent of Division I Teams|
|Decrease in 3PA, decrease in 3P%||132||37.3%|
|Increase in 3PA, decrease in 3P%||92||26.0%|
|Decrease in 3PA, increase in 3P%||69||19.5%|
|Increase in 3PA, increase in 3P%||59||16.7%|
Effects you (probably) won't notice
The extended 3-point line has affected some parts of the sport but sometimes to very minor degrees.
The average Division I 3-point percentage is 33.1% this season, which is the lowest in the history of kenpom.com, which dates back to the 2001-02 season. That's a decrease of 1.3 percentage points from last season, when the national average was 34.4 percent, but there was also a decrease of 0.7 percentage points from the 2018 season to 2019 and there's a good chance a lot of fans didn't notice that decline last season.
In the 2018 season, the national average was 35.1 percent and it was 35 percent in 2017.
It makes sense that 3-point percentages have dipped as the shot has become harder and anecdotally, you might remember games this season in which your favorite team or one of the players on your favorite team was unusually cold from deep, but the reality is that this sport-wide decline in 3-point percentage results in maybe one fewer made 3-pointer per game, on average.
Let's do some hypothetical math to figure out how much of a difference the decline in 3-point percentage makes on a 3-pointers-made-per-game basis:
- Based upon 2017-18 averages: If that team shoots 35 percent (as the national average was two seasons ago), that means it made seven threes, on average.
- Based upon 2018-19 averages: If it shoots 34.4 percent (as the national average was last season), it made 6.88 threes, on average.
- Based upon 2019-20 averages: If it shoots 33 percent (as the national average is this season), that's 6.6 made threes on 20 attempts, on average.
North Florida, the national leader in 3-point attempt percentage, takes 52.4 percent of its shots from three. The Ospreys are averaging roughly 26.5 3-point attempts per game, so there's a lot of teams that won't even attempt 20 threes in many of their games, as the hypothetical math outlined above.
Perhaps in a self-aware strategy, Florida A&M, which is last in the country in 3-point percentage at 22.1 percent, is the most 3-point-averse team nationally. Just 20.9 percent of the Rattlers' shots are threes for an average of roughly 11.9 threes taken per game.
All of that goes to say that on the whole, from a 10,000-foot view of the sport, a one-to-two-percent decline in 3-point percentage translates to two teams combining to make maybe one fewer 3-pointer than they would've last season.
UNDEFEATED TRACKER: Which teams remain unbeaten?
Sure, you might notice more significant results in specific cases on the extreme ends of the spectrum, like Saint Mary's or Florida A&M, but if you turn on the TV on a Tuesday night to watch two teams play in a game in which you have no rooting interest, you may not notice any significant changes in the number of threes attempted or made.
Similarly, 3-point attempt percentage has also declined this season but across all 353 Division I teams, the average percent change is minor. The national average for 3-point attempt percentage is 37.7 percent this season, a decrease from last season's 38.7 percent average.
That means that for every 100 shots attempted, one fewer of them is a 3-pointer this season compared to last.
It shouldn't be a surprise that 3-point attempts are slightly down year over year. That was part of the impetus for moving the 3-point line back — to curb the reliance on 3-point attempts and to make defenses cover a larger area of the court.
It's not as if college basketball teams are now collectively risk-averse with the extended 3-point line.
The national average on assist rate (meaning what percent of shots were assisted) is 51.6 percent this season, which is the lowest in the 19-year history of kenpom.com. That's a decrease of 0.3 percent from last season.
You could make simultaneous cases that the extended 3-point line may have triggered this low-mark for assist rate while also arguing that one fewer made basket being assisted out of every 300, on average, has had no tangible effect on the game.
For reference, the national average for assist rate was higher than 55 percent from the 2002 season through the 2007 season but in recent years it has been closer to 52 percent. So there was already a gradual downward trend in assist rate (once again, not necessarily a noticeable one to the casual observer if you're watching a game on your couch on a Tuesday night) but this season will potentially finish with a record-low number in that metric.
Why might that be?
Once again, the 3-point line was extended, in part, to make defenses cover a larger percentage of the court and to improve offensive spacing, but that also means that offensive players are arguably more spread out than they've ever been in the history of college basketball. Just as the 3-point line is fractionally longer this season, so too might be cross-court passes as guards and wings position themselves behind the new line.
Also, the extended 3-point line means that there's more space inside the arc, specifically in the mid-range area, which is often the home to less efficient shots compared to those at the rim (due to the high shooting percentage associated with layups and dunks) or behind the arc (due to the payoff of three points versus two).
So if you try to reverse engineer an explanation as to why assist percentage has taken a small dip, you can argue that coaches try to design plays that lead to high-percentage shots (once again, at the rim or behind the arc) but if there's more space to take mid-range shots, the shots taken in the mid-range aren't necessarily the ones that were designed when a play was called, meaning the players who take those shots may have tried to beat their defenders one-on-one off the dribble and taken shots that are theoretically less likely to be assisted.
NET EXPLAINED: How the NET rankings work
Other small but potentially related statistical trends this season include:
- The national average for block percentage is 8.9%, which is the lowest since 2009 (8.8%). In the years in between, it was between 9.2 percent and 9.7 percent, so it's a small change but you could make the case that since defenses have to cover a larger part of the court, it's harder for help-side defenders or rim-protectors to challenge shots in the additional offensive space that is now being utilized more regularly due to the extended 3-point line.
- The national average for offensive rebound percentage is 28.2%, which is the lowest in the history of kenpom.com. Once again, if offensive spacing has been increased due to the extended 3-point line and perimeter players are often operating at least one or two feet further away from the rim than they were last season, it makes sense that offensive rebounds are harder to grab. That's a decrease of just 0.2 percentage points from last season and 0.5 percentage points compared to two seasons ago. The national average for offensive rebounding percentage has been as high as 35.6 percent in 2006.
- The national average for non-steal turnover percentage (meaning what percent of turnovers aren't steals but rather, unforced turnovers, whether it's an errant pass or a shot-clock violation) is up to 10.5 percent, which is the highest since 2012 (also 10.5 percent). In each of the last two seasons, the national average was 9.7% so similarly to many other observations we've noted, it's a small blip on the radar but probably not a coincidence for the national average to have increased this season. With increased offensive spacing, you could argue that there are more, longer passes that players, especially guards and wings, have to make. That could lead to more passes that are thrown off-target or are mishandled due to the extra force required to make the pass.