Eli Witus – of Count the Basket – and I are both fans of the Pistons. After my post on Carl Landry, he suggested a similar post on Amir Johnson.Â
Like Landry, Johnson is a big man who has been quite productive in limited minutes. Unlike Landry, this production was somewhat expected.
The Unexpected Pistons
Before I get to that story, though, let me talk about the unexpected performance of the Pistons. Last year this team won 53 games with a 4.6 efficiency differential (offensive efficiency minus defensive efficiency). In the off-season the Pistons lost Carlos Delfino and Chris Webber. And they added Jarvis Hayes and two rookie guards. Furthermore, the team promised to give Webber’s minutes to Jason Maxiell. Since Delfino and Webber were above average performers, and Hayes and Maxiell were below average (and the rookies were expected to be below average), I expected the Pistons to decline in 2007-08.
But that hasn’t happened. The Pistons have actually improved tremendously, and currently boast the second best record in the NBA. So how was this possible?
Table One: Projecting the Pistons after 57 games
Table One offers two projections of Detroit’s Wins Produced. The first assumes that what we saw from these players last season (except the rookies) will be seen this year. The second projection assumes that what we have seen so far this season will be seen the rest of the 2007-08 campaign.
If the Pistons had maintained their performance from last season the team would currently be on pace to win 47 games. Again, the moves they made should have left them worse off. But when we look at what the players have done this season we see a projection of 62 wins.Â
When we look at the individual players we see that this leap can be linked almost entirely to the play of four players: Chauncey Billups, Rasheed Wallace, Richard Hamilton, and Maxiell.Â
To see what these players are doing better, we turn to Table Two.
Table Two: The Improving Pistons
Table Two reports each of these player’s statistics from 2006-07 and 2007-08. Looking over these numbers we see one consistent difference. All four players have improved with respect to shooting efficiency and assists.Â
So how did this happen? Flip Saunders is considered an offensive-minded coach, so it is possible he has done something to change how these players perform. Yes, coaching can impact performance and this might just be an example of this phenomena (then again, maybe it isn’t). Regardless, we can see that it is these four players – and their improved shooting efficiency – that has played a major role in Detroit’s rise.
Focusing just on Maxiell, though, we see more than just a change in efficiency. Maxiell has also improved on the boards. These changes have increased his WP48 from a mark of -0.002 last season to 0.105 in 2007-08. In other words, Maxiell has gone from well below average to average. Clearly increasing his minutes was a good idea.
The Amir Johnson Story
Well, maybe not. There’s another young big man that has also posted outstanding numbers for the Pistons.
In 2005 the Detroit Pistons selected Amir Johnson with the 56th pick of the draft. The following year the NBA instituted an age limit. This implementation of this rule meant that Johnson was the last player selected by an NBA team out of high school.
In 2005-06 Johnson played only 39 minutes. Last year his minutes more than tripled. And in these 124 minutes he posted a 0.247 WP48. After two years and 163 minutes, the Pistons signed Johnson to an $11 million contract.   This means – according to HoopsHype — that Johnson is currently the 6th highest paid player on the Pistons.
This season he has once again seen his minutes more than triple.  And his WP48 in these 409 minutes currently stands at 0.263. For his career he has now played 572 minutes, produced 3.0 wins, and posted a 0.248 WP8. Such a per-minute mark – both this season and for his career – bests anything posted by any big man on the Pistons roster.
When we compare Johnson to these big men we can see why he is so good.
Table Three: Comparing Amir Johnson to the Other Big Men in Detroit
Relative to Maxiel, Wallace, and Antonio McDyess, Johnson is a more efficient scorer, a better rebounder, and he blocks more shots. Plus, only Wallace gets more assists.
And it is important to note that much of what Johnson is doing this season we saw in limited minutes in 2006-07.
Playing Time for Johnson
So why isn’t Johnson playing more minutes this season?
I am not the only one asking such a question. Here is Terry Foster of the Detroit News (hat tip to Detroit Bad Boys and TrueHoop)
I love the idea of Antonio McDyess coming off the bench for the Pistons. He plays better off the bench, is more comfortable coming off the bench and can be a major spark against opponent’s second units. That is why the Pistons should entertain the idea of inserting Amir Johnson in the starting line up.
He is ready. The last nine games he has averaged 6.3 points, 4.8 rebounds, blocked 16 shots and shot 64 percent from the floor. He is a guy who makes things happen whenever he is on the floor.
Back in November, Detroit Bad Boys devoted an entire post to the wonder that is Amir Johnson. In other words, people have been noticing Johnson for awhile now.
Yet on Wednesday night Johnson never left the bench in a loss to the Utah Jazz. How could Flip Saunders keep such a talent out of the game?
One could point to Johnson’s youth. Although he is in his third NBA season he’s currently the youngest player on the roster (McDyess and Wallace are both 13 years older than Johnson).   Maxiell is only four years older, but Maxiell is a former number one draft choice.
Certainly youth (and maybe draft order) play a role in keeping Johnson out of the game. But there might be another reason. Although many of the numbers in Table Three suggest keeping Johnson glued to the bench is a mistake, there is one number that tells a different story.
Per 48 minutes Johnson averages 9.27 personal fouls. In other words, he commits six fouls every 31 minutes.  No player who started half his team’s games in 2006-07 committed fouls at this pace. And I think it is this issue that might be contributing to the decision Flip Saunders is making with respect to his rotation.
This may come as a surprise, but after intensive research I have learned that players who foul out of games don’t help a team win games. Yes, when you don’t play you don’t help (well some players help by not playing, but that’s another story). Now Johnson hasn’t fouled out of a game yet this season. But his propensity to commit fouls might suggest that Johnson is not quite ready to be a prime-time player.
If Johnson could stop committing fouls – or maybe just get a bit older – he might see his playing time increase. And if that happened the Pistons could take another leap forward. The numbers suggest that giving Johnson the minutes going to Maxiell would vault the Pistons past the Celtics. In other words, it’s possible that come playoff time, the Pistons could be favored to take down Boston.  At least favored by people looking at these numbers.
– DJ
Our research on the NBA was summarized HERE.
The Technical Notes at wagesofwins.com provides substantially more information on the published research behind Wins Produced and Win Score
Wins Produced, Win Score, and PAWSmin are also discussed in the following posts:
Simple Models of Player Performance
What Wins Produced Says and What It Does Not Say
Introducing PAWSmin — and a Defense of Box Score Statistics
Finally, A Guide to Evaluating Models contains useful hints on how to interpret and evaluate statistical models.
Wages of Wins Journal 
Nate
February 28, 2008
As a fellow Pistons fan, I was wondering when you’d get around to an Amir Johnson post. :)
I definitely agree that Amir should be playing more minutes, since he seems to be developing into a PF version of Tyson Chandler. That said, there’s a factor besides the fouls that isn’t helping his cause. He’s still filling out physically, and he’s vulnerable to being pushed around by big men with a lot of upper-body strength. That’s also a big part of why he fouls so much–he gets put in positions where it’s foul or give up an easy hoop. But once he gets stronger and learns to control the fouls, he’ll be a monster.
Pete23
February 28, 2008
cool network diagrams of nba players, http://arbitrarian.wordpress.com/2008/02/22/nba-similarity-networks/
Westy
February 28, 2008
Just a thought as I consider the strength of a team like the Pistons.
So far this year, they’ve played 36 of their 58 games against their fellow Eastern Conference members. As the undoubtedly weaker conference, would this serve to somewhat inflate their Wins Produced as compared to a Western Conference team who plays on average much tougher competition?
I pointed the same thing out when we were looking at Eastern Conference versus Western Conference All Stars, but I wonder if it would be worth checking if there is a noticeable difference in player performance versus the West or the East.
dberri
February 28, 2008
Westy,
Yes, strength of opponent is another factor that will impact Wins Produced. So in addition to diminishing returns, we have schedule strength, coaching, minutes, injuries, roster stablity, etc…
What I think is interesting… despite all the factors that shift productivity, basketball players are still more consistent across time than baseball or football players.
Jefferson Smith
February 28, 2008
Does your methodology adequately gauge the cost of fouls? My sense is that a player who fouls a lot should lose a lot of “efficiency” credit — it gives free points to the other team, adds to the team foul count (which allows for more free points), not to mention what it does to the flow of the game.
Ben
February 28, 2008
David,
Have you done any research on how much fouls are a function of role? That is, there are big men who come into the game who know they’ll only be in for limited minutes. This affects their playing style. They’re much more likely to foul given that they’re going all out for the short time they’re in the game. But that doesn’t mean if their role changes they will continue their rate of fouling.
I’d be interested to know if this is true with regards to Amir Johnson, or other, similarly foul-prone players like Anderson Varejao or DeSagana Diop.
grover
February 28, 2008
Great point, Ben.
‘Energy’ big men seem to be encouraged to take chances. That kind of scrappy play leads to extra rebounds, blocks, and steals (and most importantly, extra possessions), but also leads to extra fouls.
I expect that most bench big men would see a drastic drop in their foul rate once they moved to the starting lineup.
murph
February 28, 2008
In Joe D we trust.
Animal
February 29, 2008
A haiku about Dave Berri:
Skilled statistician
Unconventional thinker
Nemesis of Dan
Aaron
February 29, 2008
Say what you will about adjusted +/- (e.g. Richard Hamilton’s low rating), but Amir Johnson has the highest rating of any Piston (http://basketballvalue.com/teamplayers.php?year=2007-2008&team=DET). By unadjusted +/- (Overall Rating), he’s the only notable low minutes player with a positive rating. I just thought I’d mention one more datapoint supporting Amir Johnson.
Thanks,
Aaron
www.pistonscast.com
February 29, 2008
Hey Im John W. Davis from Pistonscast.com. I like the numbers you showed to prove that Amir should receive more playing time?
We just did a whole podcast episode on Amir Johnson at http://www.pistonscast.com
~John W. Davis
Leon Douglas
March 2, 2008
David, have you done any work on the value of minor league performance as a predictor of NBA success? Amir might be an interesting subject.
Two years ago, some of us at pistonsforum.com and realgm.com were getting excited because this 18-year-old kid was shooting over 60% in the DLeague. Some will reflexively discount such stats, but in baseball they mean a lot — especially when a player is significantly younger than his competitors. What’s impressive is that Amir has ALWAYS been efficient offensively, no matter what level he plays at.
dberri
March 2, 2008
Leon,
I have a student working on this at the moment. There is a link, but I don’t think it is very strong. If I get a chance, I will work through my student’s data and maybe write a post on this.