From the Courts

Quality Still Trumps Quantity When it Comes to 3-Point Shooting

Geoff Burke/USA TODAY Sports

The three-point shot is on a steady rise in the NBA. We’ve known this for a while now. Players and teams are realizing that it’s generally prudent to take a good chunk of deep shots in order to take advantage of those attempts being worth 50 percent more points more than shots taken inside the arc.

Last season taught us that teams who jack up a bunch of threes can be extremely successful. All four conference finalists were in the top seven in attempted threes per game, and six of those top seven won at least 53 regular-season games. It was fitting that the Golden State Warriors, who feature the two best long-distance shooters in the business, won the title.

Like most things in life, though, it’s best to be smart with the resources at hand rather than exploit them haphazardly. That’s a lesson the league is showing us this season with its usage of the three-point line.

There’s been a significant increase in three-point attempts between last year and this year. Those type of shots made up 26.8 percent of all field goal attempts in 2014-15, but the number has ballooned to 29.1 percent in 2015-16.

Great! Teams are being smart, taking lots of threes and becoming more efficient offensively.

Not so fast, though — the league-wide points per 100 possessions figure has held steady at 103.0 both season, despite all the extra threes. So are the offensive attacks heavy on three-point attempts actually succeeding in 2015-16, as they seemed to do last year?

Let’s take a look at the data for all 30 teams from both seasons. The x-axis represents what percentage of a team’s field goal attempts are taken outside of the arc, and the y-axis is the team’s points per 100 possessions:

There’s a decent correlation there. Outside of the Houston Rockets and Philadelphia 76ers both attempting way more threes than you’d expect out of their respective offenses, there’s a pretty clear grouping on a slight upward slope.

And here are the same numbers for 2015-16:

Sure enough, the points are a lot more scattered this season and that 0.2329 coefficient is a lot closer to zero, which indicates more randomness in the numbers. The San Antonio Spurs and Oklahoma Thunder are huge outliers in a good way, accumulating great offensive ratings in spite of low three-point attempts, while the Los Angeles Lakers and Sixers fall on the other end of the spectrum. Several other smaller outliers ruin the correlation here, as well.

Sure, the graph generally has a slight upward slope, but the data from this season is totally inconclusive. There are plenty of good offensive teams attempting relatively few threes, and vice versa.

What could be the reason behind a lot less correlation between three-point attempts and offensive efficiency this season?

With all the hoopla surrounding the rise of three-point shot, it could be that teams are defending the arc better and contesting more threes in fear of behind-the-arc barrages. Some squads are still willing to take an inordinate amount of those tightly-contested shots anyway, but they’re more difficult and end up being very low-percentage for almost the entire league (*cough* Steph *cough). That seems like a likely reason, for now.

But what about three-point percentage, though? If a team makes the most of its attempts from downtown, is that more of an indicator of offensive strength than pure volume? Let’s see.

Once again, we’ll look at data from both the 2014-15 and 2015-16 seasons. The x-axis here is three-point percentage, and the y-axis is, once again, points per 100 possessions:

Now that’s more like it. The correlation of 0.7275 is pretty close to 1, which would indicate a totally predictable slope. There are no massive outliers here.

For 2015-16, the data is pretty similar:

The correlation has gotten even stronger this season, despite a few more minor outliers like the Suns (high three-point percentage, low offensive efficiency) and 76ers (mediocre three-point percentage, dreadful offensive efficiency).

The Golden State Warriors look like they’re an outlier, but they’re not. They’re in la-la land in both statistics, but their correlation is strong with most of the rest of the league.

It’s pretty simple what this tells us: it’s more effective for teams to be smart with their threes than it is for them to toss long balls at the hoop because three is greater than two.

If a team’s personnel allows it to take a lot of threes and still convert a good percentage, it should go right ahead and do so. That’s just being smart.

If not, though, it’s better for the squad to focus on hitting a high percentage of them when the shots are open, then use the threat of an accurate deep-ball attack to open up the rest of the floor.

Don’t be surprised if the three-point shot’s growth slows down in the next few years as analytics show its poor correlation with offensive efficiency and teams continue to improve at defending the arc.

For related reading, be sure to check out Seth Partnow’s piece for Vice Sports on correlations between teams’ shot locations versus their shot-making ability.

Note: ESPN.com was used as the source for all possession-related statistics. 

Click to comment
To Top