Three Pointers and Skill Displacement

Feb 3, 2016; Washington, DC, USA; Golden State Warriors guard Stephen Curry (30) dribbles the ball as Washington Wizards center Marcin Gortat (13) and Wizards forward Jared Dudley (1) defend in the third quarter at Verizon Center. The Warriors won 134-121. Mandatory Credit: Geoff Burke-USA TODAY Sports
Feb 3, 2016; Washington, DC, USA; Golden State Warriors guard Stephen Curry (30) dribbles the ball as Washington Wizards center Marcin Gortat (13) and Wizards forward Jared Dudley (1) defend in the third quarter at Verizon Center. The Warriors won 134-121. Mandatory Credit: Geoff Burke-USA TODAY Sports /
facebooktwitterreddit
Stephen Curry, Golden State Warriors
Geoff Burke-USA TODAY Sports /

It’s not Steph’s fault. Well, it shouldn’t be at least. But soon-to-be-repeat-MVP Curry and his bombs-away style seems to have sparked a discussion about what the era of the three-pointer means.

There are several layers to this debate. The question as to whether the three-pointer itself, or the present distance and dimensions of the line are “good for the game” is complicated. As I have noted before, the notion that some item, ethic or tactic is “hurting the game” is a cudgel quite often employed in the service of a sort of mindless conservatism which, given that this is an analytics site, doesn’t sit well with me. Not all arguments are based on a fear of change coupled with an overly rosy view of the past though. Basketball is an entertainment product, so questions of aesthetics have to be taken seriously, whether the “hack-a” rules, the way the block/charge call is administered[1. Team #BanTheCharge or at least change the rule to be much more pro-offense to discourage collisions between a jumping player and one on the ground, but that’s a whole other kettle of fish…] or the three-pointer.

If NBA offenses became homogenized to the point where threes, threes and little else but threes were the only way to compete, it would undoubtedly be bad for the entertainment value. Most objective and/or statistical evidence suggests this isn’t actually the case at present — the top five offenses in the league by ORTG sit 2nd 5th, 14th, 20th and 27th in the percentage of their shots taken from beyond the arc[2. Per Basketball-Reference.] — but if the perception of “one true way” became widely accepted, it might as well be true anyway. In fact, the aesthetically-based argument is largely immune to statistical analysis. No one with credibility on the issue still thinks “three-pointers just don’t work” anymore, rather there is a wistfulness for a time when a more subjectively pleasing variety of basketball was equally or perhaps even more encouraged by the competitive environment than the current trend towards paced-up and spaced-out. If that’s what floats your boat, no amount of proof of the efficiency of MoreyBall is going to move you because the conversation is happening in completely different languages.

However, (and you had to know there was a “however” coming) I do think some of those arguments are pining for a time which didn’t really exist. The laments for the “art of the midrange” being lost are so common to be beyond cliche at this point. And it’s true, leaguewide, the choice to take midrange shots has been made less and less over the years.[3. I like to date the truly “modern” era of the NBA to the revamped hand-checking rules imposed prior to the 2004-05 season, so for the purpose of this post, I’m going to be drawing comparisons between that season and today’s game.]:

Dashboard 3 (33)
Dashboard 3 (33) /

Since 2004, the proportion of shots in the restricted area, as well as the rest of the key[5. An important distinction to draw.] have remained relatively constant. However, three pointers have skyrocketed essentially in tandem with the fall of the midranger[5. Defined here as all two pointers from outside the paint.]. But what has really been lost? Is creativity really being stifled? I don’t think so.

I would argue instead that the less creative players are on teams, those reliant on their teammates to set them up for shots rather than able to generate their own looks consistently are simply standing in different places. The following illustrates the percentage of made shots which came via assist in each basic shot zone in 2004-05 and then in 2014-15:

assistedby
assistedby /

Obviously, assists aren’t a perfect measure what with the persistent though inconsistent bias in their award across arenas. But as a general trend, the above is at least suggestive that the guys who are creating space for those doing the heavy offensive lifting[3. And just rewatch some games from the 80s or 90s if you think there is a lot of standing around and watching now — there used to be an illegal offense rule in place because teams were so interested in creating space on the strong side of the floor for one or two dominant players. In fact modern defensive complexity almost requires intelligent use of space away from the ball. All else being equal in terms of accuracy from various zones on the floor, 2004/05 shot selection against 2014/15 defense would result in an average Effective Field Goal Percentage two full points lower than was actually observed, 48.1% to 50.1%] are standing in different spots. More specifically they are spacing out to the three point arc much more than spotting up for a 19-footer.

From an effectiveness standpoint, why shouldn’t teams seek to maximize scoring on plays where the defense has been broken down, as is the case on catch and shoot jumpers? Further, the additional skill required to shoot effectively from distance[1. Don’t get me started here. Developing shooting range or not is a make or break point in the careers of many young NBA players, and if there were no special skill involved, anyone could do it. That marginal players who “stick” tend to have become competent deep shooters is a reflection of the skill being selected for rather than being easy to acquire.]  has its own benefits. Namely increasing the area of the floor an opponent is obliged to defend, which opens up driving lanes and passing angles for more efficient/creative/pleasing on the eye play for others.

Meanwhile, the in-between game is still regularly practiced, and is even aided by that additional room to operate. But it has become the province of the star player, talented enough to make an otherwise  “bad shot” — that is to say a low percentage, inefficient look, into a good or at least acceptable one[7. Which, by the way is precisely what Curry himself does with the pull up three pointer.]. Either that or the guy everyone recognizes as dragging an offense down with his poor shot selection. So I guess my question is this, why is replacing Jordan Crawfords with Jared Dudley a bad thing?

As a final aside, for all the pining for the midrange game there is the equally frequently stated desire for more team ball. The very thing seen as eliminating the former all but requires the latter. For players aside from the few like Curry or Damian Lillard who can effectively create their own looks from three off then bounce, the longer shot is usually the result of just the sort of cooperative play called for. Around 85 percent of all made threes coming via assist as compared to the rate of just under 60 percent on two-point shots, depending on the season. Asking for both at once is at minimum inconsistent, and indicates a lot of the fuzzy thinking which tends to pervade this and related topics.


Thanks as always to Darryl Blackport for assistance with some of the underlying data for this post.