When will it be three much?
Will 3-point attempt rate ever stop climbing?
If you consume any NBA media, you’ve probably heard someone bemoan how many 3-pointers are taken. And while it has a tinge of “old man yells at cloud,” they are correct that we’re in unprecedented times. The league average 3-point attempt rate (3PAr) is at a record high of 0.423, and the Boston Celtics are on course to smash the single-season team record for 3-point attempts.
For all the attention the rising volume of threes garners, it isn’t a new trend. Since the introduction of the 3-point line in 1979-80, 3PAr has been steadily increasing and shooting efficiency has come with it. For the past 45 years, the easiest way to improve your offense was to take more threes. The question is will that ever change?
Correlation is not necessarily causation. Long before the 3-point line, shooting efficiency was trending up. In a game determined by accumulating points, it’s no surprise that teams and players have steadily improved their ability to score. However, starting in 2006, the first season where the league average 3PAr eclipsed 20%, the correlation between the league average effective field goal percentage (eFG%) and 3PAr is 0.942. Understanding why 3-point volume and efficiency are so inextricably linked is paramount to discerning what the upper limit of volume is– if it exists at all.
It took a decade, but by 1990, on average, a 3-pointer was a more efficient shot than a two. Now, a trip to the free throw line remains the most efficient way to score, which gives twos back a lot of their power as 3-pointers are far less likely to earn a foul, but for the past 35 years, any random three has been better than any random two. This basic mathematical fact was the early driver of 3PAr, but then something else started to happen.
Excluding the lockout season of 1998-99 (both lockout seasons saw dips in shooting efficiency), the nadir of 2-point efficiency was 2003-04 at 0.92 points per attempt. The basketball of this era was so offensively bleak Ben ‘Career-High 9.7 PPG’ Wallace was a cover athlete. But from these depths, two key developments happened. The league did away with hand checking, opening up more driving lanes, and most importantly, the Phoenix Suns showed basketball the light.
In 2004-05, the Suns broke people’s brains by playing at the breakneck pace of 95.9 possessions and bombing 28.9% of their field goals from three. 20 years later, the Suns look like a Model-T next to a Thunderbird, but make no mistake, the Model-T changed the game and the Thunderbird just has a cool name. Yes, the Suns won the math game by taking more of the more valuable shot, but they also won the math game inside the arc.
Between 2004-05 and 2009-10, the Suns finished in the top-3 in 2-point efficiency every season and led the league for four consecutive years from 2005-06 to 2008-09. Taking more threes provided more value than just the points, they made your 2-pointers more valuable too. After the Suns run we see a slow and steady rise in 3PAr and 2-point efficiency, and in 2016-17, the first season where 3PAr eclipsed 30%, the NBA had its first season where twos were worth over a point per shot.
However, the efficiency landscape of 2024-25 is a lot different than 2004-05. Many of the efficiency gains made over the past 20 years were low-hanging fruit. In 2004-05, 23.6% of all shots were long twos, converted at a 40% rate, and in 2024-25, only 4.8% of shots are long twos, converted at a 40.1% rate. At a certain point, there will be no more long twos to trade in for threes, but once again, the 3-point revolution keeps finding a way.
While trading long twos for threes fueled the early offensive gains of the 2010s, it alone doesn’t explain why teams keep shooting more and more threes. The real secret of the 3-point explosion is what it does in the restricted area.
A random three might be as valuable or more valuable than a random two, but dunks, layups, and shots around the basket reign supreme. Since 2005-06, shots 0-3 feet from the basket have never dipped below a 60% field goal percentage, peaking in 2022-23 at 70.1%. Simply put, the best shots in basketball remain the ones right at the basket. The key to a great offense is generating as many high-quality looks around the basket as possible, and bombing from distance is the key.
*3PAr and FG% on shots 0-3 ft
Since 2004-05, as 3PAr has grown so has efficiency on shots 0-3 feet from the basket. Now, over this same period, the rate of shots 0-3 feet has declined as an overall percentage of shots, but it has grown as a percentage of 2-point attempts. So while there are fewer total shots at the basket, when teams aren’t shooting threes, they’re finding it much easier to generate the highest-value twos.
Since 2005-06 (the first season the league eclipsed a 20% 3PAr), the linear correlation between 3PAr and 2-point efficiency is 0.947. For every 0.5% increase in 3PAr, there’s roughly a 0.15% increase in efficiency on twos. Over the past two decades, the only real downside in taking more threes is 3PAr has an inverse correlation of -0.867 with FTAr, and the league has hovered around 36% shooting on threes, putting a theoretical cap on its per-shot value.
What does all this data mean for the future of the NBA? Well, it means more and more threes in the short term, but the real question is at what point will more threes lead to worse offense? Using a linear forecast model I was able to guesstimate what the league average efficiency on twos, FTAr, and free throw efficiency would be as 3PAr rises. The only assumption I had to make is that 3-point efficiency would stay at 36% as volume rises. Here are the results.
Based on the current trends, the point of diminishing returns is at approximately a 3PAr of 0.647. However, this whole forecast is predicated on 3-point efficiency remaining stable. If it dips, the 3PAr threshold goes down, but if it increases, then it would blow well past that figure. Needless to say, there are going to be more threes. By 2030, it wouldn’t surprise me if over 50% of shots were threes, and there’s no reason to think it won’t get there earlier.
I don’t personally believe the current 3-point volume is a problem, but the league will need to address it at some point. Players and teams are ruthless in their pursuit of maximum efficiency, and there’s overwhelming evidence that we’re not even close to the upper limit for 3-point volume. If you believe the NBA has become too 3-point dependent, you’re in for a rough next decade. This is only the beginning. That’s the problem when three is more than two.
For any inquiries about work, discussion, and the like you can email me at nevin.l.brown@gmail.com.
Humans should watch some of the awful post ups from the 80s and 90s, and then come back to the present and see if they still hate three point shots so much.
The question with an analysis like this becomes whether the league will allow 65% of all shots to be three point field goals, or whether they'll change some rules to throw their body in front of that possibility. They've done this all the time for different phenomena, but mostly when the league had become too physical, and therefore too defensive. This is an entirely different kind of homogenisation, and it depends whether the rule makers see this one as a threat to the game, in the same way they saw the others.
Also, do you think there'd be any justification for thinking of the effect of three point volume on two point efficiency as quadratic? You did speak of low hanging fruit earlier. I know that hasn't been the case at all so far, but we're pretty much out of mid range jump shots to get rid of now. Perhaps this pattern will change soon. I have no evidence whatsoever to support anything I've just said. It's strictly a thought.
This was a very good piece my friend. In short, it will never be three much. Three point rate will only continue rising until they are by far the majority of all shots. I suppose it's on us to decide if that's good or bad for the game.