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

Sorting the Detroit Pistons by how much they each play with Ben Wallace could be the key to building their best lineup

TrueHoop’s Henry Abbott often talks about the value of advanced statistics. His opinion, and one I share, is they’re valuable. But they shouldn’t always stand alone. They should be combined with qualitative observations. Here’s my attempt at doing that.

In his recent Q&A with Keith Langlois, Joe Dumars talked about what he’s learned this season. The general manager said, with all of Detroit’s injuries, you really can’t tell what kind of team you have. But you can evaluate individual players.

I’m just not sure we’ve been doing that properly.

If you’ve watched the Pistons this season, you know Ben Wallace has been their best player. You don’t need advance statistics to tell you that, but here are a few anyway:

  • His plus-minus per 48 minutes (-1.3) leads the team.
  • He leads Detroit in win shares (3.4).
  • He leads the Pistons in offensive rating (117) and defensive rating (104), too.

By almost any measure, quantitative or qualitative, Wallace is Detroit’s best player.

He gets most of his credit for defense. He blocks a lot of shots and alters even more. He’s a very good defensive rebounder. He also collects many of steals.

Wallace provides plenty on offense, too. He leads the league in offensive rebounding percentage. He sets excellent screens. He’s a good passer for his size. His true shooting percentage is even its highest since his second season in Detroit.

On both sides of the court, Wallace knows his assignment. When Wallace is playing, Pistons coach John Kuester can focus on the other four players – just another way Wallace makes Detroit more effective.

After Wallace, ranking the Pistons gets pretty murky.

Think about your ideal Detroit lineup. Wallace is obviously in it. But who are the other four players? Besides DaJaun Summers, you could make a semi-reasonable case for anyone else on the roster.

(And before you say it, playing Chucky Atkins at point guard might be the way to make Rodney Stuckey most effective, if that’s what you’re going for.)

Still, you have an idea of which players besides Wallace are best. But I think most evaluations of the other Pistons miss a critical factor.

My theory: Someone who plays a high percentage of his minutes with Wallace is more likely to be overrated. Someone who plays a low percentage of his minutes with Wallace is more likely to be underrated.

That’s not to say this is a rule. But if someone has more opportunities to play with Detroit’s top player, especially one who does so many little things well, I think that’s a tremendous advantage. That player could easily end up looking better than he is.

Let’s look at a visual.

Chart

image

Some keys to reading this graph:

  • The farther right the player, the more minutes he’s played with Wallace.
  • The higher on the graph a player, the better his plus-minus per 48 minutes.
  • The red line represents the expected value of a a player’s plus-minus per 48 minutes based on the percentage minutes he’s played with Ben Wallace. The farther a player is above the line, the better he’s done that expected based on the percentage of his minutes he’s played with Wallace. The farther a player is below the line, the worse he’s done that expected based on the percentage of his minutes he’s played with Wallace.

Most likely to be overrated

Obviously, there are a lot of ways to assess whether a player is overrated. But let’s just stick with this measure. Here are the percentage of each player’s minutes he’s played with Wallace:

  1. Tayshaun Prince (73.7 percent)
  2. Jonas Jerebko (73.7 percent)
  3. Richard Hamilton (73.6 percent)
  4. Rodney Stuckey (71.2 percent)
  5. Ben Gordon (63.7 percent)
  6. Charlie Villanueva (60.4 percent)
  7. Chucky Atkins (56.6 percent)
  8. Jason Maxiell (48.1 percent)
  9. Will Bynum (42.5 percent)
  10. Austin Daye (36.1 percent)
  11. Chris Wilcox (29.3 percent)
  12. DaJuan Summers (23.2 percent)
  13. Kwame Brown (7.3 percent)

So, if the goal is to evaluate which players are best, the players near the bottom of that list should play more minutes with Wallace instead of the players near the top.

Let’s look at a couple specific switches the Pistons should make to their rotation:

Kwame Brown for Chris Wilcox

Brown’s plus-minus per 48 minutes is farther above his expected value than any other Piston. But getting on the court with Wallace could be difficult.

Wallace and Brown are the only two Pistons who can legitimately play center in the NBA (and judging by Brown’s minutes, Wallace is really the only one). That explains why Brown has, by far, played the fewest minutes with Wallace.

Wilcox hasn’t played a lot with Wallace, either. But it’s been more than three times as much as Brown.

And despite more playing time with Wallace, Wilcox has been less effective than Brown in terms of plus-minus per 48 minutes.

Before the season, there was talk of Wallace and Brown starting together. I think it’s time to give that duo another chance. Maybe Brown wouldn’t look as terrible as he has.

Austin Daye for Tayshaun Prince

Next to Brown, Daye has performed the second best based on how much he’s played with Wallace. Prince is third worst behind Wilcox and DaJuan Summers.

Prince and Daye both play small forward, so the switch makes a lot of sense. A bonus: Daye is younger and a key part of the Pistons future, and Prince isn’t.

Daye has played a lot in garbage time, and Prince has been injured. So, there’s a good chance Daye would flop if he took Prince’s minutes.

But I’d be willing to take the chance to find out.

A new lineup

To restate, I’d like to see Austin Daye and Kwame Brown play more minutes with Ben Wallace. Those three haven’t played together much this season, just 7:20.

I know the sample size is small, but in that limited time, they’ve been awesome. They’re plus-11 with an offensive rating of 135.7 and a defensive rating of 57.1.

By conventional wisdom, Wallace, Daye and Brown would best be paired with a backcourt of Stuckey and Hamilton – maybe Stuckey and Gordon. Neither of those combinations has played together this season.

The Pistons should take a serious look at these changes. It would help them evaluate their players better. And it might even help them win some games.

Summary

  • Ben Wallace is the Pistons’ best player.
  • Players who play most of their minutes with Wallace might look better than they really are.
  • Kwame Brown, in lieu of Chris Wilcox, should play more minutes with Wallace.
  • Austin Daye, in lieu of Tayshaun Prince, should play more minutes with Wallace.

12 Comments

  • Jan 27, 201012:58 pm
    by TDP

    Reply

    I had to read the entry’s title three times.  Interesting stats, nonetheless.  I’d love to see the lineup of Stuckey, Rip, Daye, Big Ben and Cakes.

  • Jan 27, 20103:05 pm
    by Pratik Narula

    Reply

    I agree that the lineup should be changed….but I think we need to trade tay or rip before the trade deadline and get a center or PF who can really make our team more efficient on the offense and defensive side.
    I’d like to see Stuckey, Gordon, Rip, Ben, and Jason/Kwame. We should try and get Amare before the season ends, trading Tay and possibly summers could help phoenix as well. Stuckey, Gordon, Rip, Ben Wallace, and Amare would help us  get to the playoffs and be a really good team next year. With Bynum, Charlie, and Max coming off the bench along with Daye and Jerebko getting good minutes as well

    • Jan 29, 20105:10 am
      by Dan Feldman

      Reply

      Pratik, I’m actually looking into a trade for Amar’e Stoudemire for a future Friday Trade Idea. I think it would cost Jerebko and/or Daye, though.

  • Jan 27, 20104:54 pm
    by Stat_Police

    Reply

    This is a classic case of correlation vs. causation confusion – I feel obliged to make a point or two:

    Demonstrating that players who play a higher % of minutes with Ben Wallace tend to have higher per 48 +/- stats is neat, but does nothing to establish causation.  In fact, here’s a simple alternate theory: as an aging veteran who is, as you say, unquestionably his team’s best player, there is only one category of game time we expect Ben Wallace to miss consistently: garbage time.  Of course, the players who play in garbage time are not selected randomly – they are generally weaker players.

    This also completely disregards any possibility of players being complementary: Ben Wallace may enable some players more than others.  Hint: the big men can be disregarded or, at minimum, must be analyzed separately from the guards.  Obviously the impact of one big on another is very different.  In the specific case of Ben Wallace, one might expect him to hurt the numbers of other bigs due to his strong rebounding and post presence.  Since his outside shot is limited, it makes perfect sense that he doesn’t share much court time with Wilcox OR Kwame.  Shooting “bigs” (Villanueva, Maxiell to a much lesser degree) likely thrive more and are clearly playing with him more…  Does this mean they are “over-rated”?

    Brown is outperforming the “Ben Wallace line” significantly, but do you truly believe he’d move up the more he shares time with BW?  Seems unlikely…  It’s debatable whether he would even maintain production.

    Finally, if you take out two points (Summers and Wilcox), the entire correlation looks much weaker.

    Please, if you would like to leverage numbers to enhance your understanding of the game, do so in a comprehensive way.  Viewing player performance along a single variable (in this case % of time with Ben Wallace on court) is so statistically flawed that any conclusions drawn are probably less likely than random hypotheses to be correct (example being Kwame Brown’s insane overperformance relative to everyone).

    • Jan 29, 20105:37 am
      by Dan Feldman

      Reply

      Stat police, thanks for the response. A lot of great stuff in there. A few rebuttals:

      • Proving causation is nearly impossible in these arguments, yes. But my intent wasn’t to provide a scientifically verifiable conclusion. I wanted to raise the question of which Pistons should have a chance at playing more with Ben Wallace (and at whose expense).
      • Other teams’ worst players are playing during garbage time, too. I think that who “wins” garbage time tends to even out. In the long run, the Pistons’ bench warmers’ plus-minuses aren’t hampered by playing in garbage time.
      • My intent wasn’t to analyze how playing with Ben Wallace would help the the numbers of players. I’m looking at how playing with Ben Wallace would improve a player’s ability to help the Pistons. If Wallace and Kwame Brown go up for a rebound, and Ben Wallace gets it, that hurts Brown’s stats. But it doesn’t affect the Pistons.
      • I definitely see the logic of pairing Wallace with a shooting big like Villanueva or Maxiell. The Pistons have played decently when one of those two joins Wallace. But they’ve played decently with Wallace and Brown, too. Maybe they should give that combo more of a chance  – at least just to see how it works. The numbers suggest it might work better than what Detroit has been doing.
      • I’m not sure whether Brown would maintain his production. But if the goal right now is to evaluate individual players, I’d give him more minutes with Wallace to see how he responds. If you don’t, you’re comparing apples to oranges.
      • If you were to remove any point it would be Brown, who has played, by far, the fewest minutes with Wallace. If you did that, the correlation would be much stronger.
      • I chose to use this single variable, because I think it’s very significant. Watching games, it’s clear Ben Wallace is the Pistons’ best player. After him, the order is muddled. It’s hard to compare players whose playing time with Wallace vary so much. The attempt was to find which players might deserve more of it.
  • Jan 27, 20105:58 pm
    by joe

    Reply

    Isn’t this what the adjusted plus/minus stats are supposed to sort out?
    http://basketballvalue.com/teamplayers.php?year=2009-2010&team=DET
    The one year column is a better representation of the overall value of each player. Wilcox, Summers & Prince haven’t played enough minutes to have accurate calcs.
     
     
     

    • Jan 29, 20105:41 am
      by Dan Feldman

      Reply

      Joe, it’s similar. But with all the injuries, I think the picture of  how valuable other players besides Wallace are is muddled. By isolating players’ time with Wallace, there’s not the noise of other players — which affects adjusted plus-minus.

      At least, I think. I admit, I don’t totally understand how adjusted plus-minus is calculated. I get the point of it, but the formula is a little tougher to get a grasp of.

  • Jan 27, 20106:55 pm
    by koz

    Reply

    why not start wallace, villanueva, prince, stucky and hamilton, your best offensive punch and start competing from the outset of games instead of getting down early all of the time. Then sub in gordon or bynum or atkins along with daye and jerebko wilcox as needed. pick a solid core, stick with it and develop some leadership as a coach and an organization that both fans and players can respect and rely on.

    • Jan 29, 20105:43 am
      by Dan Feldman

      Reply

      Koz, there have been too many injuries to stick with a core. And if the Pistons are rebuilding, wouldn’t it make more sense to tinker with different lineups and figure out which combinations work best?

  • Jan 28, 201012:11 am
    by Gabe

    Reply

    Awesome post! Very interesting idea, and I generally agree with premise and definitely agree with the conclusions.
    It really makes no sense to be giving Wilcox minutes, he’s been unbelievably terrible. I’ve been happy with Jerebko’s performance so far this season, but I wouldn’t be opposed to Daye getting more minutes even if some of them keep at Jerebko’s expense. Daye has played well considering his youth, inexperience and slight frame, and getting him more time on the court with Ben Wallace could be good for his development.

    • Jan 29, 20105:44 am
      by Dan Feldman

      Reply

      Thanks, Gabe. Daye has looked pretty good playing with some unfavorable lineups. I’d like to see what he can do with the big guns. At this point, if he fails, so what? At least find out what you have.

  • Jan 29, 20108:45 am
    by Chris

    Reply

    Very interesting conclusion.  I think with such a melting pot affect with this team and inability to stay healthy it is hard to gauge this teams true potential.  I have to agree with some of the posts here that at some point there has to be more playing time for the “youth” of this team.  It is kinda funny how at draft time the pro’s were saying that summers was the most ‘NBA’ ready player of the teams picks.  What happened?  Being a die hard Pistons fan it is tough to see that they will be lottery bound unless they get their act together.  And if you look at most of Prince’s stat lines he has been pretty useless this season in his few games that he has played.  I say trade him even though his value against salary is low.  Too many players at shooting guard and small forward, not enough true points or bigs.  I am not envious of Kuesters job this year.

  • Leave a Reply

    Your Ad Here