A few days ago John W. Davis – of Pistonscast – posted a link to the following article (by Rick Bonnell of the Charlotte Observer):
Bobcats’ guard going beyond the score
The article detailed the development of Shannon Brown, a back-up point guard with the Charlotte Bobcats (and graduate of Michigan State). Here is the bulk of the story:
Charlotte Bobcats guard Shannon Brown had a simple, yet stark, question for one of his coaches a couple of weeks ago:
Why don’t you like me?
Brown now says he was joking, but assistant coach Dave Hanners had a serious reply. He liked Brown just fine. It was Brown’s game Hanners didn’t like.
“I said Coach (Larry Brown) is asking you to do certain things, and you’re not doing them,” Hanners recalled. “Most guys are misguided about how to do well. They think, ‘If I score 15 or 18 points, Coach has to play me!’
“We have Jason Richardson and Adam Morrison and a whole lot of guys who can make shots. We need Shannon to do something else to help.”
Apparently that registered, explaining how Brown slipped into the rotation Saturday. He played 13 minutes of solid defense, mostly on Miami Heat star Dwyane Wade (5-of-15 from the field), in the Bobcats’ first victory this season.
So it figures that Brown will get more chances tonight at home against the Detroit Pistons.
There was a time in training camp when Shannon Brown looked in danger of being cut, despite a guaranteed contract. The best thing in his favor was the boss’ faith; managing partner Michael Jordan likes Brown’s potential.
Yet he was blowing this chance, playing out of control with quick shots and risky passes.
“When you see something you want that bad,” Brown said, “you can push too hard and sometimes I’ve just got to slow down.”
For Shannon Brown to play, he needs to demonstrate enough playmaking skills and judgment to fill in as the Bobcats’ third point guard. That isn’t the most natural thing for a guy who grew up an explosive athlete and scorer.
So Hanners saw a breakthrough in a decision Brown made against the Heat. His teammates were running, looking for a fast-break advantage, but Brown recognized they didn’t have a numbers advantage in transition. So he backed off and called a set play.
“Two weeks ago,” Hanners said, “he would have done something to get himself a shot.”
The Shannon Brown story is reminiscent of a story I posted last month on the Jerry Sloan and the Utah Jazz. In both cases a young player (and in Utah’s case, more than one young player) was under the impression that the key to more playing time was taking more shots. And in both cases, the head coach disagreed.
These stories highlight a problem that coaches face throughout the Association. Players have an incentive to shoot. The more a player scores, the more he will get paid and the more acclaim he will receive. You can see this when you look at the determinants of free agent salaries. And you can see this when you look at coaches’ voting for the All-Rookie team.
Given the player’s incentives, it’s not surprising that Shannon Brown would look for his own shot. What’s encouraging is that it appears the Bobcats are getting through to him. After seven games, Brown is shooting 50% from the floor and his overall production of wins – which was in the negative range during his first two seasons – is now well above average.
Unfortunately, although Brown appears to be following the dictates of his coaches, he is not getting rewarded. After six games (not including Tuesday night’s game against Denver), he has missed two games (DNP-CD each time), only averaged eleven minutes per game, and has only taken ten shots from the field. So although his WP48 [Wins Produced per 48 minutes] stands at 0.421 – a mark that leads the team – he is sitting on the bench while the Bobcats give minutes (and shots) to Jason Richardson and Adam Morrison.
Let me go back to what the above article notes: “We have Jason Richardson and Adam Morrison and a whole lot of guys who can make shots.”
Here is the problem with this statement. Both Richardson and Morrison had WP48 marks in the negative range after six games. Against Denver, Richardson’s numbers were quite respectable. Morrison, though, once again struggled. In sum, Morrison is not hitting shots. And it should be noted – contrary the statement from his coach –Morrison has yet to hit shots in the NBA.
Although Morrison is not hitting his shots, he still gets to shoot. Meanwhile, Brown has hit the few shots he has gotten to take. But he isn’t getting minutes (or shots). Against Denver on Tuesday night Brown only got to play 3:45 minutes. In sum, it looks like Brown – in the few minutes he gets — is playing “the right way” (a favorite phrase of his head coach, Larry Brown). But he’s not being rewarded with much playing time.
All of this must make the relationship between the player and his coaches quite confusing. On the one hand Shannon Brown is being told to take good shots and focus on the non-scoring aspects of the game. On the other hand, Morrison and Richardson – who have not consistently hit shots this year – are not being punished with less playing time (or fewer shots).
Of course it’s still very early in Charlotte’s season. Certainly it’s more than possible that the Shannon Brown we saw during his first two seasons will re-emerge. In fact, given that his effort to play the right way is not being fully rewarded, it seems likely that Shannon Brown might want to go back to taking shot after shot. After all, the data strongly suggests (although his coaches voice disagreement) player evaluations and decisions in the NBA are still driven by scoring (which happens when you take lots of shots).
Let me close by thanking John for finding this story. And if anyone else sees stories like this, please send them along.
– DJ
The WoW Journal Comments Policy
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.
Tball
November 12, 2008
“We have Jason Richardson and Adam Morrison and a whole lot of guys who can make shots.” If you really break down that statement, it reads we have these two guys (stop) then we have a whole lot of guys who can make shots. Jason and Adam are technically excluded from the shot-making-guys grouping. And there are ways to rephrase the sentence to include Jason and Adam in that grouping, but those words were not adopted.
I don’t think Hanners was taking a shot at Jason and Adam (he may not have been consciously excluding them), but linguistically analyzing the statement, he did imply they are not included in the group of guys that make shots.
And DJ, what is the best way to send similar articles along?
dberri
November 12, 2008
Tball,
Not sure I buy your interpretation of Hanners statement, but I guess it is possible to interpret it that way.
John Davis just posted a link in the comment section here. So that is as good a way as any.
Erich
November 12, 2008
I’m sure Dave will see this, but MarginalRevolution has posted a related story just hours after this was posted. See the link on my name
dberri
November 12, 2008
What follows is the post Erich is referring to. When I get back from teaching I might offer a comment.
When one team wins, another loses. If the Celtics win the championship, the Lakers cannot. Sports at the team level, within the context of a single season, is more or less a zero-sum game. But ranking the quality and fame of players is more multi-dimensional and thus it is more positive-sum. Maybe the advent of LeBron James diminishes the luster of Tim Duncan (or maybe it doesn’t), but the total amount of fame produced still goes up because of LeBron and his efforts.
Players who maximize team wins are investing more resources into the zero-sum game. (In fact team players in small markets with few fans are especially destructive of human welfare and it is those players who should be most encouraged to become ball hogs.) Players who pursue individual glory — even if at the expense of the team — are investing more resources into the positive-sum game and thus they are doing more to benefit society.
So why is it again that we glorify the team players?
William
November 12, 2008
Dave,
This is off-topic, and I apologize.
I’m interesting in tracking a college team’s WP48 over the course of the year. Could you (and/or your commenters) review my numbers to see if they look okay. I’m using Win Score in lieu of calculating WP48, understanding that there is a slight difference between the two.
I’ll use a SG who played last night.
WS=6.5
WS/Min = 6.5/40 = 0.1625
PAWSmin= WS/min – Average at Position/Min = 0.1625 – 8.43/40 = -0.04825
WP48 = (PAWSmin*1.614) + 0.100 = 0.022125
And a PG (this could give away the team I’m analyzing!)
WS = 21.5
WS/Min = 0.5375
PAWSmin = 0.355
WP48 = 0.67297
Important note – To find PAWSmin, I took the average position win scores you recently provided and divided them by 40. For the shooting guard, it was 8.43/40. Is that the correct way to determine this value?
Final question – When calculating WP48 from PAWSMin, is the correct formula for COLLEGE players (PAWSmin*1.614)+0.100?
Many thanks, and again, I apologize for straying off-topic like this.
DonD
November 12, 2008
The coach may be making a perfectly rational choice with “Ammo” in another way: drafting him is a sunk cost. If he is going to realize any benefit, it will be in scoring, since that is the only chance for a decent ROI. A little defense or rebounding would help, certainly, but if Adam is to have any decent trade or perceived value at all, he must score. Perhaps Adam could become an above-average rebounder, passer, and/or defender, but the odds are even more against that and it would be an unwise bet. Trying to get him to be average at all is a safer bet, but doesn’t have the risk/reward value as if he were to become a good scorer.
Brett
November 12, 2008
2 things:
1st -Adam scored 16 points on 11 fga against denver, so bad example for this post. For the year, his WS/min is 0.08 – below average, but just barely, and a large improvement from his rookie year.
2nd – Shannon Brown may have a nice WP48 number – the Bobcats are getting torched when he is on the court – 122.2 defensive efficiency, versus their overall mark of 107.8. Might be a factor, no?
dberri
November 12, 2008
Brett,
Adam Morrison’ Win Score last night was 2. This is a per 48 mark of about 3.4. This is not only below average for an NBA player, but also below what he has done for the season.
After seven games, Morrison has a WP48 that is essentially zero. So he is really not helping.
dberri
November 12, 2008
William,
You can’t use the NBA averages for college. I did post the average at each position for players drafted out of college. That post went up recently so it should be easy to find.
dberri
November 12, 2008
About the Marginal Revolution post…
I think I am going to write a post on this. There is a problem in the argument being made that I want to note.
William
November 12, 2008
Dave,
Just to be clear, I did use the college averages from your recent post in my calculations.
dberri
November 12, 2008
William,
Oops, didn’t see that. Now you have another problem. I have never calculated the link between College PAWSmin and College WP48. So if you want to know “good” or “bad”, college PAWS will work. If you want to know “wins”, there is quite a bit more work to do.
slackerjoe
November 12, 2008
There are some 76er fans who believe Andre Miller’s horrible start is due to his increased shot attempts in this his Free Agent walk year. A higher PPG average could be what Andre thinks is the ticket to his next (and likely last) big Free Agent contract.
Brett
November 12, 2008
Dave (can I call you Dave?), I was looking at win score per minute, where Adam was at 0.07 against Den, and 0.084 for the season. I had read somewhere that WinScore/min was a decent approximation of WP48 – was I misled?
dberri
November 12, 2008
Brett,
You can use Win Score to approximate WP48. But there are steps you have to follow. First you need to calculate PAWSmin. Then you have to use the formula — in the comments below — to estimate WP48. I think you are looking at WSmin and thinking that is the same as WP48.
mrparker
November 12, 2008
Brett,
New Orleans has a similar drop off in defensive productivity when CP3 is on the floor. They make up for it with a bump in offensive productivity. I’m not sure this is the case with Brown. My point is their could be some other dynamic getting thrown into Brown’s floor numbers because New Orleans is ceratainly not a worse defensive team when Paul is on the floor.
Erich
November 12, 2008
William,
Draftexpress.com maintains an NCAA player database that will report WS/40 (See Stats/Usage stats). Personally, in the past I have used sportsline’s stats to calculate NCAA WS. Contact me via email (see linked site) about your goals and I’ll see how I can help.
Owen
November 12, 2008
This is a great great post.
It would be very interesting to see what would happen to the NBA if GM’s started to compensate player’s according to Wins Produced…
dustin
November 13, 2008
Owen,
Wouldn’t baseball be an example of this?
Random question for everyone;
Does anyone know of a web service that provides fairly up-to-date nba box scores?
Jason E
November 13, 2008
What format do you need the box scores in?
dustin
November 13, 2008
I would prefer .wcf , but anything would do.
John W. Davis
November 13, 2008
When I saw this story, I immediately thought of you.
You would think that Tom Izzo taught Shannon Brown this lesson in college, apparently not.
When you start off on a team with Lebron James wouldn’t you just do the little things to help the team win. I just don’t get it.
I can care less about someone averaging 27.7 points a game for their career average if it takes them 40+ minutes and 20+ shots to do it.
COUGH (Iverson) COUGH
I want efficient shooting, way above average assists per minute played, and very few turnovers.
I’m sorry I just had to vent.