The following chart describes half-court college basketball in 2008. Burn it into your head for the last few weeks of the season. You'll be a smarter fan for it.
Now for some explanation. The data behind the graph is nearly 4,000 games worth of charted shots over the past five seasons. This includes a total of 340,000 shots. There's a lot of interesting stuff in the data, and the above graph represents a basic overview of it. The red line represents the average number of shots taken at each distance. Actually, that's not precisely correct. There are some warts in the data, and it seems reasonable that the uncertainty of assigning shot location is, on average, greater than a foot. So the red line is the average number of shots within a foot of the charted value. The blue line is the percentage of shots that are made at each distance.
This is pretty fascinating. A lot of analysts lament the death of the mid-range game, and you can see that in this chart. Fewer than half as many shots are taken between 10 and 15 feet as are taken between 20 and 25 feet. What's striking is that accuracy in the mid-range is less than it is for the closest three-point shots. One might ascribe the lack of mid-range shots to players being stupid, lazy or some other negative stereotype which gets associated with the modern game, but the conclusion could also be drawn that there's more mid-range shooting going on than there needs to be. If a player can be as accurate from 20 feet, with a little practice, as he is from 15, then why practice the 15-footers if you're just going to cheat yourself out of a point? Further evidence against the stupidity theory is the lack of shooting going on between 15 and 20 feet. Players and coaches clearly understand basic math.
Of course, there are incidental benefits to attempting two-pointers which aren't quantified in this chart. Increased free throw opportunities are the main thing, but there's also evidence of a marginal increase in offensive rebounding percentage on shorter shots. Additionally, three-point opportunities that are set up by penetration wouldn't be as plentiful if there was no threat of the penetrator making a shot in the 5-15 foot range. Even knowing this, it still makes me think that Mike Kryzyzewski, Ben Howland, Randy Bennett, Trent Johnson and Todd Bozeman have it right when they design defenses that rarely allow an open look from beyond the arc. That quintet constructs its defense to play the shot-selection game by encouraging opponents to drive to that dead zone on the floor where most players are uncomfortable hoisting a shot.
Keeping in mind that this chart reflects an average over thousands of shots, I hope it can give you an appreciation for shot selection the rest of the season. Teams taking mid-range shots--especially early in the shot clock--are just making life easy for the defense more often than not. Even someone like Michael Beasley, who by my figures is an above average mid-range shooter, is someone who should be given a 15-footer whenever he wants. If you're the opposition, that's the smart thing to do considering the other spots on the court where he can be more potent offensively.
The missing link on this chart is shots inside of five feet. There really isn't a way to separate shots off the fast break from shots in the half court, so it made the most sense to leave the shortest attempts off the graph. Just so you don't think the college game is all about three-pointers, about 62% of shots inside of five feet are converted, and they account for about 30% of all shots taken. That accuracy isn't as high for shots taken in half-court confrontations, but you'd also have to assume that the free throw benefit maxes out on shots around the rim. The bottom line here is that a team still gets more bang for its shooting buck inside of five feet than beyond 20.
The analytical world has embraced Dean Oliver's "four factors" concept, and for good reason. They explain offensive and defensive efficiency in a remarkably simple way. However, it's always bothered me how shooting typically dominates the other three factors (turnovers, rebounds and free throws) and I've dreamt of the day when it can be separated into two more manageable components: shot selection and accuracy. We're still far from having the technology to do that, but with data like this, we're getting closer.
In the meantime, I hope you now have a better idea of what quality shots really are. Most good teams are spending their allotted 35 seconds with the ball trying to get short threes or shots around the rim, and opposing defenses are trying to force shots in other locations. What I've learned is that the mid-range jumper is almost dead, but not totally. It has been kept on life support by quality defenses that force their opponents to take low-value shots. Teams that choose to take a lot of mid-range shots better be good offensive rebounders, because they are destined to rack up a low shooting percentage. The most successful teams are in the nation are maximizing the value on their shot attempts and minimizing that of their opponents.
Ken Pomeroy is an author of Basketball Prospectus. You can contact Ken by clicking here or click here to see Ken's other articles.