↓ Login/Logout ↓
Schedule/Results
↓ Roster ↓
Salaries
↓ Archives ↓
↓ About ↓

Who should be shooting more?

All the discussion about how many shots Dwight Howard should get had me interested in how well the Pistons handled shot distribution this year.

Here’s a player-by-player breakdown of how the number of shots each attempted correlated with Detroit’s winning percentage:

Alex Acker

Optimal number of shots per game: 1

Average number of shots per game: 1.6

graph (5) 

FGA W-L
0 1-0
1 2-1
2 0-1
3 0-2

Analysis: He played in seven games before being traded to the Clippers. Nothing to see here. Move along.

Arron Afflalo

Optimal number of shots per game: 6

Average number of shots per game: 4.1

graph (7)

FGA W-L
0 4-3
1 3-7
2 5-7
3 5-5
4 3-4
5 4-3
6 5-2
7 2-1
8 2-1
9 2-0
10 1-2
12 0-1
13 0-2

Analysis: There’s some weak evidence the Pistons would be better off with Afflalo taking as many as five shots more per game than he did this year. In all likelihood, that’s not actually a good idea.

Chauncey Billups

Optimal number of shots per game: 11

Average number of shots per game: 10.5

graph (8)

FGA W-L
8 1-0
13 1-0

Analysis: The numbers say Detroit was better when Billups took shots.

Kwame Brown

Optimal number of shots per game: 6

Average number of shots per game: 2.9

graph (9)

FGA W-L
0 6-5
1 5-4
2 4-6
3 5-4
4 3-4
5 1-1
6 4-1
7 1-2
11 0-1
13 0-1

Analysis: There’s a slight chance Brown is the Pistons’ starting center at the beginning of next season. On the bright (less dim?) side, Detroit appears to do slightly better when Brown takes more shots than he averaged this year.

Will Bynum

Optimal number of shots per game: 8

Average number of shots per game: 6.0

graph (11)

FGA W-L
0 4-1
1 3-3
2 1-3
3 3-1
4 3-5
5 2-3
6 1-3
7 2-2
8 1-1
9 2-3
10 0-1
12 1-1
14 0-2
15 1-0
16 1-1
18 0-1
19 1-0

Analysis: The numbers here are so up and down, there’s not much to take from them. Bynum’s role changed so much — from out of the rotation to go-to guy in the fourth quarter.

Richard Hamilton

Optimal number of shots per game: 15

Average number of shots per game: 15.6

graph (12) 

FGA W-L
8 1-1
9 2-1
10 1-1
11 3-3
12 4-4
13 5-4
14 1-1
15 3-1
16 1-3
17 0-2
18 5-2
19 0-6
20 1-2
21 1-2
23 0-1
24 0-1
25 2-0
26 1-0
29 1-0

Analysis: Once Pistons coach Michael Curry benched Hamilton, the guard decided to take control of the offense. He held the ball longer, which results in more scoring — but more turnovers, too.

The Pistons are better when he’s an efficient scorer (9-to-15 shots per game) instead of a volume shooter. Here’s hoping he returns to that mentality next year (if he’s still in Detroit).

Walter Herrmann

Optimal number of shots per game: 6

Average number of shots per game: 3.5

graph (13)

FGA W-L
0 5-5
1 5-5
2 3-4
3 4-6
4 1-3
5 2-1
6 4-2
7 2-1
8 1-0
9 1-1
11 0-1
13 0-1
15 0-1

Analysis: The Pistons had a jump in production once Herrmann took five shots in a game. This made Curry’s decisions to bring Herrmann of the bench for the first time in the fourth quarter all the more troubling. A shooter like Herrmann needs the time on the court to develop a rhythm.

On the flip side, if you’re relying on Walter Herrmann to take double-digit shots, you’re probably not going to win.

Allen Iverson

Optimal number of shots per game: 15

Average number of shots per game: 14.7

graph (15)

FGA W-L
4 0-1
7 1-0
8 0-2
9 2-2
10 2-2
11 0-3
12 2-2
13 1-4
14 2-3
15 1-0
16 3-1
17 4-1
18 2-2
19 1-3
20 0-1
22 1-1
23 1-0
24 1-1
28 0-1

Analysis: Warning to everyone considering Allen Iverson for next year: His teams still do better when he takes a lot of shots. Besides no signs he’d even take a lesser role, these numbers don’t show he could make that work.

Amir Johnson

Optimal number of shots per game: 6

Average number of shots per game: 2.6

graph (16)

FGA W-L
0 5-6
1 5-9
2 3-6
3 2-6
4 4-4
5 2-2
6 3-0
7 1-2
8 2-0

Analysis: These numbers probably don’t reflect much because Johnson doesn’t typically look for his own shot. He can finish on the fastbreak and gets putbacks. So, if he’s taking a lot of shots, the Pistons are probably doing other things right that lead to wins.

Jason Maxiell

Optimal number of shots per game: 4

Average number of shots per game: 4.1

graph (17)

FGA W-L
0 1-1
1 4-6
2 4-4
3 8-6
4 10-8
5 4-6
6 3-1
7 1-2
8 1-1
9 1-2
10 1-1
11 1-1

Analysis: His chart is pretty horizontal, which makes sense. Maxiell doesn’t do much besides score inside. His ability to do that doesn’t seem to change much, regardless of how many looks he gets. So, Detroit performs about the same — no matter how many shots Maxiell takes.

Antonio McDyess

Optimal number of shots per game: 8

Average number of shots per game: 8.5

graph (18)

FGA W-L
3 2-1
4 1-2
5 3-4
6 4-6
7 3-3
8 6-2
9 2-3
10 2-3
11 2-2
12 1-1
13 0-1
14 1-2
15 0-1
16 0-2
18 1-0
20 0-1

Analysis: The Pistons are a little better when McDyess takes a lower-to-average number of shots. He’s too talented to disappear offensively. But at his age, he can’t be relied upon to carry too much of the load.

Tayshaun Prince

Optimal number of shots per game: 14

Average number of shots per game: 12.4

graph (20)

FGA W-L
2 0-1
4 0-1
5 1-0
6 1-2
7 1-1
8 1-3
9 1-4
10 4-6
11 5-3
12 5-3
13 2-4
14 4-2
15 7-3
16 3-4
17 1-2
18 3-0
19 0-1
21 0-2
26 0-1

Analysis: If anyone should get more shots, it’s Prince. But it’s his passivity that’s to blame more than coaching strategy.

Walter Sharpe

Optimal number of shots per game: 2

Average number of shots per game: 1.4

graph (21)

FGA W-L
1 3-2
2 3-0

Analysis: Give Sharpe the ball! Look at that steep upward slope above.

Rodney Stuckey

Optimal number of shots per game: 10

Average number of shots per game: 11.6 

graph (22)

FGA W-L
2 0-2
3 0-1
5 0-1
6 2-2
7 1-2
8 6-5
9 5-5
10 3-2
11 1-3
12 6-6
13 3-1
14 2-4
15 3-3
16 0-1
18 0-1
19 1-1
20 0-2
24 2-0
25 0-1
29 1-0

Analysis: He definitely needed a bigger role than what he had at the beginning of this season. But he’s not ready to carry the scoring load. Detroit’s performance was way too up and down when his shot totals reached the 20s.

Rasheed Wallace

Optimal number of shots per game: 15

Average number of shots per game: 10.9

graph (23)

FGA W-L
1 0-1
2 1-0
4 1-0
5 0-1
6 2-1
7 1-3
8 2-2
9 3-3
10 3-4
11 5-7
12 6-3
13 0-1
14 1-2
15 4-1
16 2-1
17 1-2
18 0-1
19 0-1

Analysis: When he shoots too little, he’s disinterested in the game. When he shoots too much, a lot of those are probably 3-pointers. Neither is good for his team. His shot total should be in the middle range.

12 Comments

  • May 15, 200911:32 am
    by gulkerbr

    Reply

    Is this metric actually used in sports analysis? I don’t see how you can make any conclusions based on this data. There is only a noticeable trend for a few of the players (Kwame, Aflallo, maybe a couple more). For the most part, there is far too much scatter in the data to even see a trend.

    It seems misleading to say, for example, the Pistons were %100 when Iverson shot 7 times. How many times did that happen? I’m guessing there were more instances of him shooting much more than that.

    Also, what decides an optimum shot number when several numbers are tied? Why is Iversion better at 15 than 7 or 23?

    • May 16, 20095:02 pm
      by Dan Feldman

      Reply

      You’re right. For many, players there isn’t a strong trend. But there was no way of telling which players have noticeable trends without charting everyone.

      And the optimal number of shots is a bit subjective. I also included how the team did when a player shot just one or two off from what could be an optimal number.

  • May 15, 20091:31 pm
    by Mike

    Reply

    Your comments on Johnson are incomplete.

    Last year Johnson shot 45% from outside the restricted area. That was 2nd best among all Pistons players. McDyess shot 47% from outside the restricted area.

    On the other hand Jason Maxiell shot just 31% from outside the resticted area. Hamilton shot just 40% from outside the restricted area.

    So the problem is not that Johnson can’t shoot and can only make put backs and run outs but that the Pistons froze him out of the offense last year. I saw most of their games and there were many games in which Johnson received fewer 2 or less passes per 10 minutes played.

    Johnson needs to play for a team that will pass him the ball on offense and utiltize his vast array of offensive skills.

    Trade Johnson this summer.

    • May 16, 20096:31 pm
      by Dan Feldman

      Reply

      Johnson shot outside far less than almost everyone else on the team. Twenty percent of his shots were jumpers. Twenty-one percent of Kwame Brown’s shots were jumpers. Jason Maxiell was next lowest at 39 percent.

      So, to say Johnson made a high percentage of his outside shots is a bit misleading. He just didn’t take enough of them to make those numbers reliable.

  • May 15, 20096:14 pm
    by snafu

    Reply

    Although these charts are interesting, you might be better off batching shot groups together to reduce the effect of small sample sizes that make the graphs go haywire. Maybe for each player have only 5 data points, so in Sheed’s case that would mean grouping 1-4, 5-8, 9-12, 13-16, and 17-19 shot attempts. Obviously it wouldn’t change much but it would make the graphs slightly more effective imo.

    • May 16, 20096:33 pm
      by Dan Feldman

      Reply

      The problem with that is the cutoff points would be arbitrary. Shots that shouldn’t be grouped together could be, and vice versa.

  • May 16, 20099:28 pm
    by jason

    Reply

    Basically what this tells us that when bench players shoot more, the Pistons have a higher winning%. Of course bench players don’t get green-lit unless Pistons have a lead, so that correlation makes sense.

    It also says that when the top players (except Sheed)take shots that approach their average, the team tends to win more games. That Sheed is the anomaly also makes sense as he rarely plays consistently (sometimes passive, sometimes active), meaning you don’t get reliable results. Which is a result in itself.

    What does this ultimately say about the Pistons? That none of their players are good enough to carry a team. When any individual dominates the ball, the team loses. When they share the ball and each player gets near to their average, the team wins. And if they win by alot, then Afflalo gets to shoot 6 times in the last couple minutes.

  • May 17, 20094:19 am
    by jankx

    Reply

    These numbers are misleading, some guys shots (Kwame Browns’) will be higher becuase the team was already had a big lead. Kwame shooting more will not help your team win.
    There is so much more to basketball than statistics and no coaching decisions should be made on this data alone.

  • May 17, 20095:39 am
    by Ignarus

    Reply

    neat to look at the numbers, but it’s hard to say much of anything without knowing if the player is shooting efficiently or just chucking it. with guys playing small roles especially, they can be getting shots because it’s garbage time and they won or because it’s garbage time and they lost.

    interesting, but very, very complicated issue.

  • May 17, 20098:26 pm
    by Dave

    Reply

    Yeah, but you could have a scaled bin size. The problem with your current presentation is that there are just too few data points for your bins, especially charting against the Binary Variable W/L.

    Also, there is no idication of Makes, let’s face it that’s what we want not FGA but FGM. Charting FGA vs FG% might be more meaningful, but again you don’t have a big sample.

  • May 18, 20099:51 am
    by Josh

    Reply

    This analysis indicates that just about every player needs more shots. There just aren’t that many shots to go around so the whole thing is kind of moot. Unless they play seven seconds or less type basketball.

  • May 20, 20097:59 am
    by John W. Davis

    Reply

    I think the team would need like 10-15 more shots by your calculations.

    How can we do that?

    I like what you were trying to do but we have to take away shots to get those extra shots.

  • Leave a Reply

    Your Ad Here