The NBA introduced the 3 point line during the 1979-80 season. That year, only 3.0% of field goal attempts were 3-pointers, and only two players (Rick Barry and Brian Taylor) averaged over 3 3-point attempts per game. Last year in the 2014-15 season, 27.1% of field goal attempts were 3s, and 153 players averaged at least 3 a game (minimum 20 games played).
This disparity is exaggerated by the fact that, up until the 90s, most players in the league spent some portion of their college/pro career playing without a three point line (the NCAA did not adopt the line for all games until 1986), yet the trend towards more three pointers has remained very strong throughout this millenium. Where have all these new 3s come from?
Let's explore 3 possible sources for the increase in NBA 3 pointers: 1) Players tend to take more 3s as they get older and more experienced. 2) 3 shooting has been increasingly valued in the draft, allowing more high-volume 3 shooters into the league. 3) Team strategy has evolved to emphasize 3 pointers.
The Age Effect
During the first 3 years of his career, just 37/2804 (1.3%) of Chris Bosh's field goal attempts were 3 pointers. Over the past three seasons (including this 2015-16 so far), 608/2465 (24.7%) of his shots have been from 3 point range. Despite only having played 53 games this year, his 222 3-point attempts would have been second most in the 1979-80 NBA (only 17 behind Brian Taylor's 239 in 78 games).
Is Bosh typical of other NBA players? It's unsurprising that players take more 3s as they age and gain NBA experience. Coaches believe (with plenty of supporting evidence) that shooting, particularly long distance shooting, is coachable and can be improved with practice. Moreover, developing a 3 point game helps players stay productive on offense as they age, offsetting declines in the necessary quickness and strength to drive to the basket.
The data bears this out, as players tend to take more 3s as their careers evolve. Moreover, the average NBA experience level has gone up a bit over the last 35 years (though it hasn't change much in the last 15).
But these effects alone don't come close to explaining the overall surge we've seen in 3 point shooting during the past 35 years (including the last 15 in particular). An extra year of experience is worth about 0.5 extra 3PA per 100 FGA, and in the last 35 years the average NBA experience level has only gone up a little more than 1 year (and not at all the past 20 years). Moreover, there's little evidence that the aging effect is stronger now than before; that is, the rate at which NBA players increasingly shoot 3s as they age has been pretty constant. Looking at the data, we see that the increase in 3 point attempts has been picked up pretty uniformly across player experience levels, rather than primarily by aging players:
Thus, while Bosh is a striking example highlighting the global evolution of NBA offense, he's not a good poster child for how the NBA has changed. We do see slightly more 3s now than we used to thanks to longer careers and aging effects---yet Bosh is an anomolous exaggeration of this trend rather than its prototype.
Over time, NBA rookies have taken more and more 3s. As we see below, the historical rise in proportion of 3 point shots for rookies parallels that of the NBA as a whole. This rookie effect is a pretty significant ancestral path to today's NBA shooting patterns. As we mentioned, an aging effect alone does not explain the overall rise in 3 point shooting, since the average age and experience level of NBA players has barely changed over the last 35 years (and not all the last 20). If anything, the graphic below shows a stronger aging effect during the 80s and earlier 90s than in recent years, though this is confounded with (a) the shorter 3 point line from 1994-1996, (b) inertia from the era before 3 point shooting, and (c) lack of measureable aging effect for recent rookies.
A vivid example of a uniquely modern NBA rookie is Pablo Prigioni. Besides his age (35), another remarkable feature of Prigioni's rookie season is that 61% of his shots were from 3 point range. While 61% is an extraordinary number, recent years have seen an explosion of rookies taking a large portion of their shots from downtown. In the last 3 full seasons (2012-14), 38 rookies took at least 40% of their shots from downtown---that's more than all NBA rookies between 1979 and 2003. As the graphs below show, there's been steady increase in high-3P-volume rookies, who have dominated the overall increase in rookie 3 point attempts.
We don't think an aging effect alone would explain the dramatic rise in 3 point shooting we've been witnessing; however, an aging effect intensifies the overall trend when rookies enter the league with increasingly higher 3 point shooting rates.
It's no secret that "Moreyball" (prioritizing 3s, dunks, layups, and free throws) has made its mark on the NBA and influenced shooting patterns on offense. It stands to reason that part of the rise in 3 point shooting (particularly in the 10 years, when teams have started doing more math) is due to changes in team strategy, alongside league demographics and individual behavior. These factors are tough to disentangle, as a team's strategy and personnel decisions exist symbiotically.
One thing we can look at is the extent to which offenses feature proficient 3 point shooters. Specifically, we'd like to see whether guys who take a large portion of their shots from 3 point range (and make them at a high rate) get more playing time in recent years, and take more shots during the time they're on the floor. We calculated correlations between usage statistics (minutes per game, field goal attempts) and three shooting (3PA per 100 FGA, 3P make percentage) to see if teams have utilized 3 point shooters differently over time.
These plots don't reveal much. While more sophisticated analyses might reveal more interesting patterns, we're convinced that changes in team strategy, conditional on teams' personnel, don't play a major role in the rise of NBA 3 point shooting.
The phenomena we've looked at in this post could be rehashed in a number of interesting ways, and by no means paint a complete picture of how NBA offenses have come to take more 3 pointers. Moreover, they don't tell us what's "right" for the game, or how shooting patterns will continue to evolve in the future. Caveats aside, we think there are a few important take-aways from this analysis.
Players take more 3s as they age. This has been a trend as long as 3 point shooting has existed.
The sports analytics movement probably hasn't really affected individual players' styles. The age effect has plenty of reasonable explanations from a skill and physique standpoint; moreover, the strength of the aging effect isn't stronger now than it used to be.
There's been a big shift in the types of players entering the league, particularly increasing numbers 3 point specialists. This might reflect the growing global popularity of basketball, as good 3 point shooting is a route for shorter players to stand out offensively.