“Threes And Layups”: The Evolution of NBA Offense In One Stat

On this here site, D’Antoni Index has become a catch-all stat for measuring “modern” NBA offense. If the goal is to shoot your outside shots from 3-point land, your inside shots inside the restricted area, and to draw lots of fouls, adding up the rate at which teams accrue such things must by definition measure how well teams are achieving that goal.

What’s more, what constitutes a “high” 3PAR, for example, has radically changed over the year.

For example, the Indiana Pacers, through games of January 17, 2021, stand 21st in 3PAR at .373; the majority of NBA teams take at least 40 percent of their shots from beyond the arc, with Chicago standing 16th at .403.

But as recently as 2016, the 2021 Pacers would’ve led the league in 3PAR; the Mike D’Antoni-coached Houston Rockets took 37.0 percent of their shots from beyond the arc and ended up tops in the Association.

But one reason the Pacers shoot so few of their shots—by 2021 standards, anyway—from beyond the arc is that they’re the best team in the league at attacking the basket. 36.1 percent of the Pacers’ attempts come from 3 feet and in; layups, dunks, putbacks, all the high-percentage looks you want from an offense.

Indeed, the Pacers shoot 65.1 percent from that range, the kind of eFG% you’ll only get if you shoot 43.4 percent from three.

Add in a .223 FTR (fifth-worst in the league; the Pacers, due to referee bias or inability to draw contact, simply do not get to the line) and what you’ve got is a raw D’Antoni Index of .957.

League average is .904. That’s .396 3PAR, .249 FTR, and .259 0-3 foot attempt rate. Indiana has an adjusted DAI of plus-.053.

But let’s go back to 2016, when Houston “dazzled the league” with all those 3-pointers that would rank in the bottom third of the league just five years later.

That year, 3PAR was .285, FTR was .276, and 0-3 foot attempt rate was .293, for a league-average DAI of just .854. This is the reason why that last step of normalizing to league average is so important for determining whether a “modern offense” continues to be so.

The 2016 Rockets had a raw score of 1.043 and a league-adjusted DAI of plus-.189. They may have shot “only” 37 percent of their shots from long range, but they led the league in FTR and stood fifth in shots within three feet of the rim. They were over 22 percent higher than the rest of the league and would still rank way up there even today.

So let’s look at D’Antoni Index since the Dark Ages in 2002, 20 seasons’ worth of data counting the current one, and see if we can learn anything about not just how “threes and layups” have changed but if there are any patterns in efficient basketball that are changing in ways we might not expect in 2021.

By year, and in order of DAI, 3PAR, FTR, and 0-3FTAR:

2002: .761 (.181, .293, .287)
2003: .770 (.182, .302, .286)
2004: .787 (.187, .303, .297)
2005: .824 (.196, .324, .304)
2006: .844 (.202, .333, .309)
2007: .854 (.213, .327, .314)
2008: .836 (.222, .306, .308)
2009: .844 (.224, .306, .314)
2010: .842 (.222, .300, .320)
2011: .798 (.222, .300, .276)
2012: .789 (.226, .276, .287)
2013: .806 (.243, .270, .293)
2014: .829 (.259, .284, .286)
2015: .829 (.268, .273, .288)
2016: .854 (.285, .276, .293)
2017: .877 (.316, .271, .290)
2018: .870 (.337, .252, .281)
2019: .910 (.359, .259, .292)
2020: .926 (.384, .260, .282)
2021: .904 (.396, .249, .259)

So What Have We Learned?

Well, for one thing, you’ll notice that 3PAR has been on a sharp upward trend since 2012, and even going back to 2002, teams have been steadily taking more and more 3-pointers. We saw a lull between 2008 and 2012 as 3PAR leveled off league-wide, but in general, every year, teams shoot more long balls.

For another, FTR has since 2014 been in steady decline and is way down since the mid-aughts, where a parade of players to the free-throw line was consistent with the bullying low-post play that was common at the time as teams strove for the highest possible raw FG% they could squeeze out of a possession. This was back when eFG% was still largely confined to esoteric intellectual discussions rather than talked about as it is today, when guards are measured by that stat and people are told not to focus on that raw FG% number.

And at the last, it’s hard to believe that only 25.9 percent of shots for a whole season will be taken within three feet of the basket as they have been so far in 2021. It seems reasonable based on a decade of trends to expect that number to rebound to between 28 and 30 percent of overall shots.

For the most part, you can see a couple of philosophical shifts in basketball over the past 20 years.

The first one, in 2012, was teams finally starting to understand on a conceptual level “defend without fouling”, sharply reducing free throw rate.

And, partially as a reaction to that and partially due to Mike D’Antoni changing the way the league looked at the 3-pointer once he got to Houston, in as much time as it took free throws to decline, we saw an explosion in long-range shots.

But overall, the league had this weird sine wave of efficiency over a 14-year stretch before the world went mad in 2016 and basketball saw this explosion in modern play that continues to the present day and leaves open questions about the limit of just how deep this rabbit hole can go.

First, a rise in 3-pointers to complement rough inside play led to an efficiency boom at the end of the Dark Ages.

That was followed by efficiency getting dampened as defenses adjusted to the new reality and took away a weapon from the offense in the low post.

And finally, efficiency rebounded as offenses reacted to losing the inside game by launching a ton of 3-pointers.

Will defenses be so effective at taking away the 3-pointer and defending on the perimeter that we see a recovery in low-post efficiency to the point where the big man again becomes ascendant, where superstars like Giannis Antetokounmpo and even low-post threats like Domantas Sabonis become much more valuable? What about Clint Capela and Andre Drummond, guys who are as old-school as they get as far as low post play that wouldn’t have been out of place in 2004 or even 1974?

We’re seeing the efficiency curve at an inflection point, and history has taught us that when that happens, there’s often a radical change in philosophy just waiting for a team to win a title with it and change the way the game is played.