The Charlotte Bobcats won 18 games in its first season in 2004-05. The next season the Bobcats won eight additional games to finish with 26 victories. And then last year the team set another franchise record with 33 wins.
This year, though, the drive to the top has stalled. If the Bobcats manage to win their final three contests – and that would be a bit if – Charlotte will only tie the mark the team achieved last year.
Forecasting Charlotte
When we look at the roster the Bobcats employed this year, and what these players did in 2006-07, we see that Charlotte should have expected another substantial leap in the standings.
Table One: The Charlotte Bobcats after 79 games in 2007-08
As Table One reveals, after 79 games the Bobcats should have already won 43 contests. In other words, if each player continued to play as well as he did in 2006-07 the Bobcats could have expected to have already qualified for the playoffs.
Before we get to why this forecast wasn’t realized, let me first note why this expectation should have existed in the first place.
The Bobcats in 2006-07 employed Adam Morrison at small forward, who managed to be the least productive player in the league. When the season was over Morrison had produced -6.7 wins.
Two events happened in the off-season. On draft night the Bobcats acquired Jason Richardson. And then before the season started, Morrison was lost to injury. Although Richardson is typically a shooting guard, Charlotte appears to have given him substantial time at small forward 2007-08. In essence, it appears that the Bobcats replaced Morrison with Richardson.
Again, Morrison was very bad last year. And Richardson has been slightly above average this season. Consequently, this one move –from a below average (okay way below average) performer to an above average player — is worth about 12 wins. In sum, replacing Morrison with Richardson should have been enough to vault this team into the playoffs. Had that happened the Bobcats would have been one of THE stories of the 2007-08 season. But this didn’t happen.
What happened to Charlotte?
When we look at the individuals employed we see three players who are responsible for Charlotte not making the expected leap: Matt Carroll, Emeka Okafor, and Gerald Wallace. Although most of the players the Bobcats employed this season (including Richardson) are about as good as they were last year, this trio declined to the aggregate tune of 13 wins (after 79 games). In sum, virtually all of the team’s drop-off – from the forecast to what we observed – is connected to changes in the performance of Carroll, Okafor, and Wallace.
And this leads us to ask, what happened to this trio?
Again we turn to the numbers. Table Two reports the per 48 minute production these players offered in 2006-07 and 2007-08.
Table Two: Carroll, Okafor, and Wallace
The Carroll Story
Let’s start with Carroll. With respect to Net Possession (Rebounds +Steals – Turnovers), Carroll is actually doing more this year. And his shooting efficiency is also the same. Where we see a change is in field goal attempts. Last year Carroll took 16.8 shots per 48 minutes. This year his shots from the field are down to 14.0. This decline in shot attempts is the big reason his WP48 [Wins Produced per 48 minutes] has gone from 0.093 (or nearly average) to 0.030.
Now why did his field goal attempts decline? One suspects that the addition of Jason Richardson – who easily leads the 2007-08 team in field goal attempts – took some shots away from Carroll. In other words, Carroll’s performance is impacted by the expected diminishing returns we see with respect to points production. As you add additional scorers, the production of existing scorers should decline.
The Okafor Story
Okafor’s WP48 has declined from 0.290 to 0.202. So Okafor is still very good in 2007-08. Just not quite as good as he was last year.
When we look at the individual numbers, we see virtually no change in scoring. Shooting efficiency, shot attempts, and points production look to be about the same. When we look at Net Possessions we do see a drop-off of one per 48 minutes. Part of this is due to a small decline in rebounds. The other part is due to a small increase in turnovers.
Turning to the remaining numbers we see a drop-off in blocked shots and a smaller decline in assists. And that is all the changes we see.
None of the changes we see in Okafor are that dramatic. All we see are just little changes here and there that combine to result in the observed decline in wins. My sense – and I am speculating – is that changes we see in Okafor’s numbers are just part of the random fluctuations we are going to see in any player’s numbers over time. Players are not robots. Although performance tends to be similar across time, players do not do exactly the same thing year after year.
Although I am sure we can invent a reason for why Okafor declined, I would prefer to file this one under “random fluctuations.” In other words, I think the best explanation for Okafor is no explanation.
The Wallace Story
Like Okafor, Wallace is also offering less this season. But unlike Okafor, I think there is an explanation. Before we get to that, let’s talk about the observed changes.
Relative to last year, Wallace did spend more time at the free throw line and accumulated more assists. But we see substantial declines in shooting efficiency, rebounds, and turnovers. And that led to a decline in WP48 from 0.209 to 0.094. In words… Wallace has transformed from an extremely good player to just average.
For Wallace, the best explanation is injury. Wallace has missed 17 games this season to a variety of ailments and the numbers suggest that he has not been fully healthy when he has played.
Going Forward
One might expect that next year Wallace will be healthy. And the bounces we see in Okafor could go the other way. So the Bobcats might expect to improve next year without making any real changes.
Of course, changes always occur, especially on teams that miss the playoffs. Morrison will probably recover and return to the line-up. If his production does not increase dramatically he will be a drag on Charlotte’s outcome next season.
Barring a trade, Charlotte will also be adding a lottery pick next year. Rookies tend to play badly, so it’s possible the lottery pick – whoever that might be – will also cause wins to be reduced.
Although Morrison and the lottery pick might not help, there are a few changes that could lead to more success in 2008-09. Looking at the roster, the Bobcats only have two productive big men: Okafor and Nazr Mohammed. Jared Dudley – as a rookie – spent some time at power forward and was not a complete disaster. Maybe he will improve and be of more help. If not, another big man (perhaps the often injured Sean May) capable of being at least average would be useful.
If the Bobcats added an average big man, it would allow the team to resist the urge to play Wallace at power forward. And with Wallace spending more time at small forward, Richardson could move back to shooting guard. Those two shifts would also result in more wins next season.
And then we have the point guard spot. Jeff McInnis and Earl Boykins each offered production that was in the negative range. And although the starter – Raymond Felton – was in the positive range, he was still below average.
Felton’s performance is especially interesting. His WP48 his rookie season was 0.092, which is quite close to average. Last year his WP48 declined to 0.072. This year it is 0.050. In sum, Felton is gradually offering less and less.
Going forward the Bobcats need to figure out why Felton is getting worse. And if they can’t reverse that trend, then the team needs a new point guard.
In sum, Charlotte needs to find another big man and they need more production at the point. If those things happen – and Morrison and the lottery pick don’t subtract too much – Charlotte can actually make the leap we expected. And if that happens, look for the Bobcats in the playoffs in 2009.
– 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 
BobcatsPanthersfan
April 13, 2008
I don’t see how Felton has “gotten worst” if anything he’s improved the most since taking over the PG spot our idiot “coaches” is the reason Felton hasn’t broke out yet give him a full 82 games and he would be averaging double doubles in dimes and points just look at his numbers since we waived Touce. But yea pretty good read I think with some draft night luck, Larry Brown, and a healthy Crash, Morrison and May we could make some noise in the East.
GWIII
April 13, 2008
Couple of things….
First, if you DO look at Felton’s numbers since they waived McInnis, he’s pretty much the same player he was before McInnis was waived. Queen City Hoops took a pretty good look at that (http://www.queencityhoops.com/EverybodyLovesRaymond.php). And one thing hasn’t changed; Raymond still can’t finish inside. Combine that with him shooting a career worst from 3, and you can see where his production is going down.
As far as Wallace is concerned, even when he was healthy in the first part of the season before the concussion, he wasn’t the player he was last season. I look at it as a part of the change of the offense, where he had been taking 3 times as many three pointers as he has over the course of his career. He’s only making 32% from deep, and this pulls down both his regular and true shooting percentages. His rebounding was down, and his blocks were reduced by half from last season. All of that negates the face that he has more assists and is going to the line more frequently and making more. Add an increase in turnovers to almost 3 a game, and there’s your sinking back to average right there.
Tim
April 14, 2008
Hi Professor Berri,
What’s your take on adjusted plus minus? I ask because I saw this startling analysis from “professor Stephen Ilardi, who has consulted with Roy Williams’ staff at Kansas, and is an expert in adusted +/-” on the TrueHoop blog:
“As you know, the best measure of a player’s overall impact on the game’s bottom line is his adjusted plus-minus rating, which adjusts statistically for the effect of all teammates and opponents, and also factors in both offensive and defensive contributions. Paul’s adjusted plus-minus rating right now (through games of April 10) is +0.76 points per 100 possessions, indicative of a modest overall positive contribution. Put simply: his superb offensive contributions have been largely offset by his apparent defensive liabilities. Based on this season’s adjusted plus-minus numbers, a much stronger MVP case can be made for either Kobe (+12.04) or LeBron (+11.35), and perhaps the best case of all for Dwight Howard (+16.57).”
The full post is in this link:
http://myespn.go.com/blogs/truehoop/0-32-45/The-World-s-Most-Unlikely-MVP-Showdown.html
Tim
April 14, 2008
And it looks like Henry Abbott bought Ilardi’s argument, rating Paul fourth in his MVP vote because of it:
http://sports.espn.go.com/nba/news/story?page=MVP08-Abbott
dberri
April 14, 2008
Tim,
Wayne Winston has a book coming out (sometime this year) that will discuss adjusted plus-minus in some detail. Having communicated with Wayne a number of times, I find his work to be interesting. But I am not so sure his work is easy to replicate.
Tim
April 15, 2008
Hmm, that measured response is hardly the refutation I expected. I’ll have to look for Wayne Winston’s book.
Personall I find adjusted plus-minus frustrating because I find it hard to keep track of it. At least with the Wages of Wins I can keep an eye on rebounds and shooting percentage and turnovers and other underappreciated stats and get a rough idea of who is underappreciated without making the calculations.
The big question in my mind is whether Wages of Wins does enough to credit great defensive players and to penalize bad defensive players. Ilardi apparently argues that adjusted plus/minus reveals that Chris Paul is a horrible defensive liability, so bad that it cancels out his historically-high (for a short man) offensive stats. Wages of Wins reveals no such thing. Are you saying that the jury is still out?
dberri
April 15, 2008
Tim,
Let me put it this way….
I think Wayne’s work is very interesting and I am looking forward to his book. I have spoken to Wayne, and his take on Paul is different than Ilardi.
People are working on incorporating defense differently in a Wages of Wins framework. Defense is there, but it is only based on the team (not the individual). As I have said before, I think keeping it at the team level makes sense. Of course, looking at it from the individual perspective has value also, and that step will be taken at some point. As I have also said before, I do not think that step (which I have tried in the past) will dramatically change the final results.
Tim
April 15, 2008
Thanks, that clarifies matters. I do hope that Wayne Winston specifically addresses Ilardi’s argument about Paul in his book. Or would a guest column be possible?