An Intro to Shot Selection

By:
Published: December 20, 2017

Introduction

This is a cursory post on shot selection. Many of you may already be familiar with concepts explained below, so bear with me. For those who aren’t, this post aims to contextualize the value of 3 point shots. A lot of people wonder why our star big men have moved further away from the paint and even players like Dante Cunningham are seeing most of their shots behind the arc. The goal of this is post is explain why this is happening and why the league as a whole is trending this way. I know it’s easy to pin these actions on a coaching staff, however I want to explain the math behind this methodology. There are actual mathematical reasons behind why the league continues to fire away behind the arc. First a look at two hypothetical teams, then a look at how this relates to the Pelicans.

Mid-range Shots and Shot Selection

I think the best way to examine this topic is to imagine two perfectly identical teams, team A and team B. Let’s pretend that both teams shoot the exact same percentage from each spot on the field. Both teams shoot 50% at the rim, 40% from mid range, and 35% from three. When team A and team B play each other, both take 100 shots. The only difference between the two teams is how many shots they shoot from each spot. For simplicity let’s pretend the breakdown is like this:

This would lead to the following point totals

Team A ends up scoring 95 points. Team B ends up scoring 100 points.  Both shot the exact same percent from all spots on the field, but Team B wins the game. Why? They took more shots that had a higher points per shot value. 35% from three is worth 1.05 points for every shot you take. 40% from mid-range is worth 0.80 points for every shot you take. So it makes sense to want more threes in general.

Here is a fun example. Same situation as above, but what if Team B only took 1 more three than team A?

Team B still beats Team A 95.25 to 95! I know what you are thinking, you can’t score 95.25 points in a real game and that’s true. But over the course of a season, those percentages add up.

How does this apply to the Pelicans?

Last year the Pelicans shot 1,618 mid-range shots, that’s about 19.7 per game. They hit them at about 39% which ranked 18th in the league. The best team (Warriors) shot 44.4%. They only took 17.6. Let’s ignore them for a second and focus on the Pelicans. They shot 35% overall from three last year. At a glance there was only a 4% difference between us taking mid-range shots and us taking threes. But the per shot difference comes out to 0.78 points per shot vs 1.05 points per shot. Per 100 shots that’s 78 vs 105. No contest on what gives you the better return. Let’s go back to the Warriors for a second. A team with the best shooters in the world only hit 44.4% of their mid range shots. That’s still only 88.8 points per 100 vs 105.

Here is where it gets really crazy. The Pels took 235 threes last year that were defined as having “Tight” defense. They made 28.5% of those threes. Per shot that comes out to 0.855. That means even their contested threes last year were worth more than their average mid-range shot! And it was almost worth the same as a Warriors mid-range shot! Crazy.

This year the Pelicans are taking 12.8 mid-range shots per game, roughly 7 less per game. On the surface this is a big change. But if you think about as AD taking two less from about 8 per game to about 6 per game, and Jrue, Dante, Solomon, DeMarcus, E’twuan all only taking one less mid-range shot per game, it’s easy to get to that 7 less total number. And I don’t think taking 7 less mid-range shots means you have to replace those shots with threes. You could totally replace those shots with shots at the rim. AD  and Jrue are taking almost a career high percentage shots at the rim this year and are having their most efficient years to date. Over the course of the year 7 shots per game is 574 shots, which is a lot of shots. To take the extreme example, if you replaced those 574 mid-range shots with contested threes which you made at 28.5%, you would score almost 43 more points! Obviously you’d want to replace those with much better shots than contested threes, but I think this conveys the point.

On a related note, the Pelicans offense is currently ranked 7th in the league and scoring about 4 more points/100 possessions from last year.

My Personal Thoughts

I don’t think less twos means more threes. I think less twos means more shots in the paint or rim and more threes, whichever is easiest and more suitable for your team. For our team I think it means more shots in the paint and rim, but if the situation occurs where one of our poor shooters has to shoot, it’s better they shoot a three. It’s all about maximizing expected points per shot. Now I don’t fully buy into the Houston style “eliminate” all mid-range twos offense totally. I think having the ability to do that is necessary because good teams will try to take away good shots. You need to make them pay for trying to limit your efficiency. It keeps the defense honest and adds needed variation to offense. This is important in the playoffs especially I think. And lastly, when the game is on the line, expected outcomes go out the window in my opinion. If the game is tied and you need a bucket, you take the shot that has a higher percent chance of going in. 40% > 34%. But for the other 99 possessions of the game and the thousands over the season, you need the maximize what you are getting out of your offense. Easiest way to do that is to cut out the item that gives you least amount of return.

 


 

 

 

Leave a Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.