The Historical Evolution of Three Point Shooting
I observed the other day that the average points per NBA three point attempt in 2008 exceeded the average points per possession. Then the next day I read in Bill Simmons’ The Book of Basketball about how low three point percentage was in the first few years of NBA three pointers. So I thought it might be useful to take a historical look at this question. Thus the following charts points per possession and points per three point attempt ever since the three pointer was introduced in the 1980 season:
In the early years, three point percentages were terrible and a possession ending in a three point attempt tended to be a badly below-average possession. But pretty quickly either the existing players got better at shooting from long range, or else better long range shooters got recruited into the association, and three point attempts became about average. Ever since that happened, the two indicators tend to be pretty close, though with the points per three point attempt slightly above the average points per possession.
It’s be interesting to compare the comparable figures from NCAA and different European leagues. Have all leagues converged to this same equilibrium, or are there some places where the average points per three pointer is somewhat lower than the average points per possession.
22 Responses to “The Historical Evolution of Three Point Shooting”
Register
RSSLogin
Topic Cloud
ArchivesBlog RollAbout Matt YglesiasContact Me
Donate
November 10th, 2009 at 3:24 pm
The comparison is slightly misleading, since possessions that lead to turnovers are counted in points per possession, but not in points per three point attempt, even though presumable a significant number of such possessions would have led to a three point attempt if not for the turnover. For a truly accurate comparison, you would need to compare points per two point attempt(counting attempts leading to free throws) with points per three point attempt (counting attempts leading to free throws). I suspect two point attempts still generate slightly more expected value than three pointers.
November 10th, 2009 at 3:26 pm
the best value of the three pointer is when it is actually a two
as in pickup to 15 by ones and twos
November 10th, 2009 at 3:30 pm
I don’t mean to threadjack, but I think this is an interesting story and was wondering what the rest think of it:
http://www.huffingtonpost.com/huff-wires/20091110/us-priest-attacked/
Some Marine reservist beat the crap out of a Greek Orthodox priest he mistook for an Arab, then claimed the priest was a terrorist and was shouting Allahu Akbar. The priest only speaks Greek and was visiting the US.
Why is it that whenever a Muslim does something horrible (or is accused of doing so), it’s always the Sikhs, Bahais, Mediterraneans, etc. that get beat up?
November 10th, 2009 at 3:31 pm
Wouldn’t another possibility be that teams got better at deciding when best to go for a three point attempt and when not to?
November 10th, 2009 at 3:39 pm
Part of the low percentage of the pre-1989 three pointers is that so few of them were taken. Many of those attempts were last second shots or desperate attempts to catch up. And players rarely practiced it, since the shot wasn’t really integrated into the offense.
November 10th, 2009 at 3:40 pm
There is one other consideration, and it is a big one!! The distance of three point line (or arc if you will) has been changed several times. In the beginning it was further out than it is now. Today’s players are no better at outside shooting than the NBA players of 20 years ago or 30 years ago. Bill Bradley (the former Senator) could out shoot most of today’s players, even the three point shooters (Ray Allen et all) that are hyped up today.
November 10th, 2009 at 3:55 pm
In order to make the analysis complete, it’s important to go beyond points per possession in isolation. What I’d like to see is the impact of missed three point attempts on the resulting possession, does it lead to more easy transition baskets as many coaches believe?
So I’d like to see those two factors, the points per possession for a team that takes a trey, and the points against per possession on the next possession (for the other team), especially after misses (but not limited to that).
A third factor I’d throw in is the offensive rebounding advantage that could conceivably result from missed treys, since rebounds often come out longer and boxing out is reduced in effectiveness (so, how many more possessions might you earn?).
Otherwise, interesting analysis Matt, I didn’t realize how close to the equilibrium this is.
November 10th, 2009 at 4:07 pm
Here’s a pretty good article on this subject.
November 10th, 2009 at 4:08 pm
Interesting that the two lines seem to cross right around 94-95, when the Rockets won two championships with an offense consisting of Hakeem Olajuwon and a bunch of guys who could (and would, often) shoot the 3.
November 10th, 2009 at 4:16 pm
Tom,
94-95 was when they shortened the 3-point line to 22 feet in all directions. Not all that surprising that there was a big jump.
November 10th, 2009 at 4:24 pm
Matt,
You were too young to remember, but the 3-point line used to be out at a more manly distance.
November 10th, 2009 at 4:27 pm
One thing that Kevin Drum pointed out with respect to this is that it is very impressive how closely correlated the two lines are now. NBA players have a very good sense of what 3-point shots are good shots – so much so that they take almost the exact optimal number of 3-pointers.
November 10th, 2009 at 5:33 pm
You need one other control besides the one identified by the first commenter regarding turnovers. I assume shots followed by an offensive rebound and eventual score are counted as a single “possession.” If that’s the case, efficient requires that the point yield from a three-point shot be lower than for a “possession”–after all, three-point shots also might be followed by an offensive rebound and eventaul score.
November 10th, 2009 at 5:57 pm
You were too young to remember, but the 3-point line used to be out at a more manly distance.
No he’s not, because I’m the same age as him and I remember when the 3-point line was moved inwards. I’m guessing he doesn’t remember because he didn’t watch basketball growing up as a kid–which is no knock on Matt, I’m just sayin’.
November 10th, 2009 at 6:29 pm
Bill Bradley (the former Senator) could out shoot most of today’s players, even the three point shooters (Ray Allen et all) that are hyped up today.
I assume you’re talking about skeet, otherwise you’re insane.
November 10th, 2009 at 7:52 pm
Or perhaps neither. A three pointer was considered more a desperation shot than a regular offensive option. The strategic game was still to work the ball in for the closest possible shot. It took them a while to figure out that attempting to isolate a good post player down low while the other four guys stand around just behind the line could be a workable strategy, though unbelievably boring to watch.
And yes, I am a basketball luddite who watches very little NBA anymore because the three-pointer has ruined the game (amongst other reasons).
November 10th, 2009 at 8:29 pm
The NBA 3-point line was moved in to a uniform 22 feet in 1994, but it was moved back to its original 23′9″ distance (22′ in the corners) for the 1997-8 season and has been there ever since. So no, the improved efficiency is not related to the line being moved in (though that may have played a role in the Rockets’ championships).
November 10th, 2009 at 9:35 pm
In the early years of the NBA trey, shooting wasn’t just terrible; it was practically non-existent. The early Magic Johnson-era Lakers won two titles in the first three years under the rule, without almost ever putting up a three-point shot. In fact, the famous game in which Magic tipped off at center against the Sixers for the 1980 NBA Title (when Kareem was hurt), featured almost no attempts from distance, either by the backcourt-dominant winners or by Philly, even though the Sixers trailed most of the game by a significant margin and needed the points for the comeback.
November 11th, 2009 at 9:43 am
[...] them back at their best against Jazz Scores Report Who needs Stephen Jackson? Yglesias The Historical Evolution of Three Point Shooting NESN Time For Rajon Rondo, Deron Williams to Step Out of the Shadows Six Celtics Land on [...]
November 11th, 2009 at 11:31 am
Yeah, Bradley wasn’t a great shot. If he had said Sam Jones (OR was it KC Jones?), or Rick Barry, he’d have a point.
November 11th, 2009 at 1:52 pm
You were too young to remember, but the 3-point line used to be out at a more manly distance.
As usual, Steve Sailer is wrong. I know it’s been said already, but the current 3-pt line is the same distance it’s always been, except for a brief experiment in the mid-90s when they shortened it. In other words, what dbeach said at #17.
November 12th, 2009 at 5:05 pm
The nba voted to move the line out before they voted to move the line in.