Talking Stars and Teamwork with Henry Abbott

Posted on January 23, 2009 by


Last night I was watching the Orlando Magic play the Boston Celtics on TNT.  At halftime a heated argument (at least it looked heated) took place between Kenny Smith and Chris Webber.  Essentially, Smith argued that how your team performs should impact whether or not you are placed on the All-Star team.  C-Webb vehemently disagreed (with support from Gary Payton).  The exchange led me to send the following e-mail to Henry Abbott (of TrueHoop):

Not sure you are watching TNT’s basketball coverage tonight, but Chris Webber just said something I think is very interesting. They were discussing who should be an All-Star and Kenny Smith argued that the record of the team should be a factor. C-Webb then disagreed, noting that he played for good teams and bad teams. On both teams his work-out routine was the same and his effort was the same. Unfortunately, outcomes were different because other people (GM, Owner, etc…) put together teams that were either good or bad. Since C-Webb didn’t get to pick his teammates he didn’t think he should be penalized for this outcome.

What is interesting about this statement is that C-Webb is arguing that his performance was essentially independent of his teammates. Gary Payton then chimed in and essentially agreed.

This is interesting because it contradicts a piece of conventional wisdom in the NBA. People believe that teammates play a large role in the performance of the individual. C-Webb and Payton are arguing that this is not the case. And I think the empirical evidence supports their position.

Today, at TrueHoop, Henry responded to my observation.  Before I repost what he had to say, let me note that this resulted in a series of e-mails between Henry and I.  After I post what Henry originally said I will also share our further thoughts (with Henry’s permission):

Henry Responds (see Leaders, Wins, Teammates, and All-Stars at TrueHoop)

Whoa, holy cow is this a complicated issue.

To me, to think about it, it helps to remove the conversation from basketball for a moment. (This is one of those “bear with me” moments. Hang in there, I’ll make it make sense, I promise.)

You go to work. There is a leak in the roof. Water is dripping onto your desk. You can not do your job.

Now … what happens next?


Probably the first thing you’re thinking is: Fixing the roof is not my job. My job has a definition, and this ain’t it. Somebody needs to get on this, and pronto, because I need to do less mopping up my desk, and more of the stuff that’s in my job description.

So then you go and talk to … who? Your landlord? Your boss? Your maintenance guy?

Somebody whose job this is.

The person you talk to about it, they aren’t loving this either. This might be something they can handle, but it is still unexpected and unpleasant. Fixing that leak was on nobody’s to do list for this day.

If you have good leadership where you work, though, I put it to you that the people you talk to about it will embrace the challenge. and take responsibility for getting it done, even if they are not able to fix it themselves. They might ask you to take a role. (“Could you move your desk for a couple of days?” kind of thing.) But somebody needs to make clear that getting this fixed is their deal.

Nothing like being self-employed, or running a start-up, to teach you this. Whatever walks through the door is your job in that setting. (If there is no boss to turn to, then guess what: You’re the boss. Not as fun as you’d thought it be, huh?)

(So, now, here we go, trying to connect this to Chris Webber, and reality.)

The reality is that, as nice and normal as it is to want a finite set of obligations … that’s not how the world is.  Your little list of duties could never be long enough to address everything that might come up. Lightning strikes, life, the weather, happenstance, coincidence, the economy, the Spanish Inquisition, … crazy things happen all the time. 

And if you are insulated from all that, if you have a nice, tidy list of responsibilities, then that means you are living under the protective umbrella of somebody else. Some leader or protector is taking those lumps for you. Which is as it should be. Not everyone is equipped to deal with everything.

So, I don’t know if Chris Webber is right or wrong about how you can be just the same player on a good team or a bad team. But I do know this, any good team — in business, in basketball, in anything — has at least one if not several people who are equipped to take responsibility for whatever comes up. On many teams this may be a GM, coach, trainer, etc.

But it is often a player. In an environment where top players have more power, money, and longevity than coaches, a star like LeBron James or Kobe Bryant is in the best position of anyone in the franchise to lead in almost everything the franchise does, from charity work to boosting morale.

Some players, some of the very best players, obsess about how to make their team winners. Other players, also some of the very best players, obsess about how to be the best players they can be. 

Those are slightly different things. And for the first kind of player, there really is no tidy definition of what their job might be. I sincerely respect the people who embrace this challenge.

Even just little things on the court: Let’s say you’re a big man, and know the point guard is supposed to close out on that open shooter on the perimeter. You’re supposed to stay with the big man on the low block.  But that shooter is wide open, and makes 44% of his shots from there. Are you going to just do your job and stay with your guy (everyone will know whose fault it was) or are you going to take responsibility for the team’s success, and close out the shooter? Or even more importantly, are you going to think of a way to help the point guard handle it better next time?

Just being a good player … That’s huge. But it’s not to be confused with taking responsibility for the team.

And I think that on some level people know that. And I think that on some level, maybe it’s even subconscious, that’s why coaches tend to vote together an All-Star team made mostly of winners. You can’t really know, from afar, which top players have fully taken on responsibility for the team, and are playing a role in seeing to it that the roof doesn’t leak. But you can know that if everything on the team is not working, there’s not much of a chance that the leadership is stellar. And if the team is humming along, well then the star of that team tends to get the benefit of the doubt. 

If you’re a great player on a losing team? Consider yourself warned that becoming an All-Star is not just about working out and being talented, as Chris Webber suggests. The riddle you have to solve is also about figuring out that whole team thing, which is kind of messy. But it’s a team game.

Too many of the things that go into being a team leader are hard to quantify, and undercelebrated. You don’t get to be in the evening’s highlights for boosting the morale of a teammate, hitting the open man, or rotating crisply on defense. But you do get to win some games for that, I believe, and you do get the inside track on being an All-Star because of that. It’s a system with some bias in favor of winners, and I’m OK with that.

My Response

Here is an e-mail I sent to Henry after I read his post:

My sense is that for teams to work they have to buy into the idea that they are all in it together.  So the Cavaliers have to see themselves as a “team”.  But clearly, LeBron is the key to the team.  Take him away, and the team has real problems.  Take away Pavlovic and the Cavs still keep winning.

I think this is the point Webber and Payton were making.  At one point Payton argued that if he played for the Clippers he would still be a great player.  His team wouldn’t win much, but he would still be great.  I think that is true.  At least, I think moving a player like LeBron to the Sacramento Kings isn’t going to transform the Kings into a championship contender.

I think what this ultimately boils down to is a conflict between a necessary myth (everyone on a team is it in it together) and the fact that there are substantial differences in the productivity of individuals.  And these differences do  not depend on the teammates.  Webber and Payton were great players, regardless of who they played with.  It should be the case that we can still say that, even when their teammates were not that good.

And Henry Responds Again

Yes. Big range in talents and productivity among players.

But I think that maybe 90% of players perform well short of their best-case-scenario potential.

Title teams tend to have very productive role players, compared to other role players.

I think stars on the leading edge do things to inspire those role players, and to get more productivity out of them over the long haul.

And then I respond again

It is true that title teams have very good role players.  I think the evidence suggests, though, that these players were also good role players before they got to the title team.

I would add, the one issue you see when you look at the performance of players on a team is diminishing returns.  The more productive a player, the less productive his teammates.  This is not a large effect, but it is there in the data.  It also suggests the opposite of what people argue.  Productive star players diminish the productivity of their teammates.  After all, there is only one ball.  If one player takes a large number of shots or gets a large number of rebounds, there is less for others to do.  Again, it is not a large effect.  But it is there.


All in all, this was an excellent exchange.  The Kenny Smith view (supported by Henry) argues that teams win and lose together.  The view of Chris Webber and Gary Payton (which I support) is that a star player should not be penalized by the poor play of his teammates.

Let me close by noting that I think the views expressed by the players reflects the productivity each player offered in his career.  Here is each player’s career performance in terms of Wins Produced and WP48:

Kenny Smith: 26.4 Wins Produced, 0.057 WP48

Chris Webber: 109.1 Wins Produced, 0.170 WP48

Gary Payton: 169.5 Wins Produced, 0.173 WP48

Both Webber and Payton were above average performers.  Smith was generally below average.  Given these differences, I think it’s not surprising that Smith emphasizes the “team” aspect of production.  Webber and Payton – two very productive players – naturally emphasize their importance to their respective teams.  In sum, it appears production – in this case – dictates perception.

– DJ

The WoW Journal Comments Policy

Our research on the NBA was summarized HERE.

The Technical Notes at 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

Wins Produced vs. Win Score

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.