→ Sports
Precision Shooting
| Sun Nov. 8, 2009 11:29 AM PST
And now for something completely different, here is Matt Yglesias on three-point shooting in the NBA:
In general, there’s not enough three point shooting happening in the NBA. In the 2008-2009 NBA season the average possession resulted in 1.083 points. The league average on three point shooting,
meanwhile, was .367 meaning that the expected value of a three point attempt was 1.101 points. Better than average. Indeed, last year only four teams scored at a more efficient rate than 1.101 points per possession.
My takeway from this is a little different: it's astonishing how close these two averages are. If you assume that players generally attempt the best possible threes, then additional attempts are going to have a poorer success rate. Probably much poorer, since I imagine that effectiveness falls off exponentially with distance and coverage. In other words, if NBA players attempted even 1% more threes, the expected value of triples would probably fall below the average possession.
That's no big surprise, I suppose. Professional basketball players, it turns out, have an extremely precise sense of how effective various plays are. Still, the fact that they take threes at something like 99+% of the ideal rate is pretty remarkable.
Advertisement
Advertisement
Comments
Speaking of global warming
Submitted by Anonymous (not verified) on Sun Nov. 8, 2009 11:55 AM PST.
Speaking of global warming and green policies coming from the White House, http://www.azcentral.com/news/articles/2009/11/08/20091108Xmastree1108.html
Using similar logic, NBA
Submitted by Anonymous (not verified) on Sun Nov. 8, 2009 12:38 PM PST.
Using similar logic, NBA players don't dunk enough. They make 99% of their dunks, or 1.98 points per possession. They should do it every time! Oh, wait, they can't. Because the other team plays defense and keeps your from dunking at will. Same with 3-pointers. You generally only take them when you manage to beat the defense and have an undefended shot.
Don't Forget Fouls!
Submitted by Detroit Dan on Sun Nov. 8, 2009 1:10 PM PST.
You rarely draw a foul on a 3 point attempt. There's a slight benefit from scoring points on free throws as opposed to 3 point attempts, since opposing players will be eliminated if they commit too many fouls...
3 Point Shooting
Submitted by Smitty (not verified) on Sun Nov. 8, 2009 1:13 PM PST.
Yglesias provides us with the average points per 3 point shot, not of 3-point possession. Probably 1/3rd of missed 3 point shots will be rebounded by the offensive team.
If one assumes that the expected points obtained from an offensive rebound is roughly one point, then the expected scoring from a made 3-pointer plus the offensive possession that frequently follows should total at least 1.4.
What's more, the assumption that an offensive team would be expected to score as a rate of 1.0 is quite conservative since offensive rebounds are extremely likely to lead to scores.
Don't try three % more threes ?
Submitted by RobertWaldmann on Sun Nov. 8, 2009 1:40 PM PST.
I'm with Yglesias on this one. You note that the marignal three is a lower percentage shot than the average three. However, also note how teams end up trying threes. An almost always necessary condition is no fast break (I know some guards pull up and take 3 but very few and rarely).
A three point attempt is very often the result of failing to get to the hoop with either a fast break or with the half court passing game. Three point efficiency should be compared with 2 point efficiency with the same number of seconds left on the shot clock. I guess it wouldn't be close.
There's a reason Yglesias
Submitted by B on Sun Nov. 8, 2009 3:08 PM PST.
There's a reason Yglesias hasn't been hired as a color commentator by NBA basketball. He's almost never interesting and even less likely to be right. Talk about an inept nerdfest.
Three pointers
Submitted by etaoin shrdlu (not verified) on Sun Nov. 8, 2009 5:08 PM PST.
Want three pointers? Women's college basketball is the place to see them. You'll also see some different basketball, frequently based on team work and strategy, instead of the dribble-dribble-dribble-dunk of the men's game. I recall that some smart sportswriter once observed that the pro game could be made more efficient if each team were initially given 100 points and then played for five minutes.
I bet
Submitted by Ralph Kramden (not verified) on Sun Nov. 8, 2009 6:02 PM PST.
if you did a similar analysis on stolen base attempts in baseball, you'd find about the same thing. (The break-even point for stolen bases in terms of runs expected is about 2/3.)
Wrong stats
Submitted by Ken D., proud to be (not verified) on Sun Nov. 8, 2009 7:03 PM PST.
I think Matt Yglesias weakened his argument by choosing the wrong comparison. Based on his source, it looks to me like points per shot for 2-pointers is .97, which is a more direct comparison. I agree, however, that his conclusion is simplistic and unpersuasive for many of the reasons advance by Kevin and previous commenters. Coaches these days have plenty of stats, and I don't think they are missing anything this basic. I think there is a strong analogy with run v. pass in football. Most teams average more per play passing, but that does not justify jumping to "more passing is a no-brainer."
Yes Kevin, it's shocking
Submitted by James2 (not verified) on Sun Nov. 8, 2009 10:52 PM PST.
Yes Kevin, it's shocking that people seem to figure out the best possible option all on their very own, with no league intervention or 3 point attempt mandates!
Also, the utter lack of understanding regarding basic statistics on this thread is disturbing.
Three Point Shots
Submitted by Anonymous (not verified) on Mon Nov. 9, 2009 11:32 AM PST.
There are two externalities that Matt Yglesias' analysis overlooks. First, a three point shot has virtually no chance of being fouled (at least compared to a driving shot). Second, a missed three point shot is more likely to lead to fast break points going the other way.
Welcome to basketball: a sport that is almost completely immune to statistical analysis.
pros and collegians
Submitted by keith on Mon Nov. 9, 2009 4:56 PM PST.
anonymous (the most recent one) is mostly correct. james2 is, unsurprisingly, mostly wrong.
yes, driving to the hoop leads to more fouls, which is why Dirk Nowitski became so much more effective when he added drives to his standard repertoire of long-distance jumpers. and yes, long rebounds can lead to fast-break points for the other team, although the break is largely a lost art in the NBA. and although the 3-point shot does also lead to more offensive rebounds, they aren't likely to be taken near the rim, and are therefore less likely than you might think to lead to easy baskets.
coaches can be surprisingly averse to changing the patterns they've grown used to using, or that they learned from their mentors. that's why coaches who embraced the 3-point shot soon after it was introduced to college basketball were often so quickly successful. think of the rick pitino kentucky teams that took what might have seemed an absurdly high percentage of 3-point shots. if you charted their shots, you'd see very few from close inside the 3-point line (lower frequency of success, and lower reward), and relatively few from 12-15 feet. like yglesias, pitino figured out that the 3-point shot had a higher return than all but the closest 2-point shots, with the added advantage that you could take a decent 3 earlier in a possession (before the defense had a chance to establish good rebounding position), and that you could get more shots per game and more running, which gave an advantage to a team with a deep bench. this would also allow a relatively short team to negate the advantage of a team with a large, skilled center.
unfortunately for pitino, the advantages didn't translate as easily into the NBA, which has faster, larger, more skilled defenders and a more distant 3-point line. the rewards for 3-point shots were less, as were the associated advantages. in fact, the NCAA basically wised up to this, and pushed the collegiate 3-point line back, which restored some balance to the college game. like the distance from the mound to the plate in baseball (and between the bases), there seems to be some generally optimal distance for the 3-point line (and also for the width of the lane, but that's another question), and setting that distance requires . . . the "league intervention" that james2 so fears.
Since you don't seem to have
Submitted by James2 (not verified) on Mon Nov. 9, 2009 6:05 PM PST.
Since you don't seem to have a sense of humor, I'll spell it out for you: that was a joke. And what part of what I said was wrong? As Kevin himself states, players take 99%+ of the ideal rate of threes. And, if there's more "league intervention" and they move the line back, guess what? Players will still take the optimal number of threes.
Also, which NBA are you watching? Defense is better than in college? Maybe, in some series, in the playoffs, but certainly not in regular league play. Defense is atrocious to non-existant in all but heated rivalries and playoff preview games.
Pitino's offensive strategy worked so well because he was able to make 12 guys run a full court press all game long. Try getting even 8 guys who are getting $2-25 million a year to bust their ass up and down the court for 48 minutes. Never gonna happen.
optimal?
Submitted by Anonymous (not verified) on Tue Nov. 24, 2009 8:05 PM PST.
Dude, "I was only joking" stopped working around the fifth grade. You were wrong. Man up and admit it.
It kind of depends on what you mean by optimal. Obviously, players and coaches have figured out a way to maximize their returns. Nonetheless, from a fan's point of view, a league where there was a pull-up three on every possessioni would be less-than-optimal. The league "intervened" by creating a three-point line, and the NCAA did the same thing. The NCAA then decided that their line was too close, resulting in less-than-optimal games, and decided to move it back out.
As for defense, if you ever watched a CBA game of a development league game, you'd see that there are a lot of skilled offensive players who don't make NBA rosters. In fact, if you watch pre-game shoot-arounds, you'll see that virtually everyone in the league can hit uncontested shots with astonishing regularity. The reason players don't shoot 50%-60% is defense -- individuals and teams play good enough defense to keep good shooters under wraps, and only extraordinarily skilled offensive players can overcome that. Such players can make it in the league without a great deal of defensive effort, but most can't, and therefore have to put a lot of effort in on defense.
As for Pitino, lots of college coaches and even high school coaches press. Nolan Richardson and John Thompson won NCAA titles doing it, and the John Wooden UCLA teams were famous for their three-quarter court press. Several college coaches have also tried the shoot-quickly strategy, most notably Paul Westhead at Loyola Marymount. The difference was that Pitino had his players race downcourt and take a (usually) open three-point shot almost every possession unless there was an open lay-up or dunk. It worked at Providence, and it worked at Kentucky, but it didn't work in Boston. That isn't because NBA players wouldn't "bust their butts" (marginal players certainly would, and the possibility of making the playoffs or winning the NBA championship motivated hard-to-handle cases from Pete Maravich to Dennis Rodman).
Rather, it's because:
1) the NBA 3-point line is far enough out to make that a less successful strategy,
2) NBA players get back on defense extremely well,
3) defensive rotations keep wide-open threes to a minimum, unless offensive players are willing to shoot from well behind the line.
But there's no need to let reality interfere with your opinions and your "jokes." Keep watching NBA games without seeing what's going on, and keep loudly asserting your opinions. Young children may think you're knowledgeable because you're loud.
Post new comment
MoJo Comments: Send Us Your Feedback
We changed our spam software to better filter comments. Should you encounter any issues, please let us know.
