The Carmelo Effect and the GlenGarry leads

This is an unfair post. 

However, using my methodology, I wanted to see how the Knickerbockers have been performing — in a relative sense — after two games with Carmelo Anthony. It turns out that they are not performing that well so far, at least on defense. 

I calculated the Road Team and Opponent Win Score for the Milwaukee Bucks (35.50-42.90) and the Home Team and Opponent Win Score for the Cleveland Cavaliers (41.90-50.70) and then averaged them and compared the average to the two game Team and Opponent Win Score the Knicks have achieved in the Carmelo Anthony Era (46.50-49.25).

The Knicks offense has been okay — just about average (+0.30)– but their defense has been atrocious — well below average (-10.25). 

Prior to the trade, the Knicks Team and Opponent Win Scores were (47.94-46.84).  The 2010-11 NBA average is (43.01).  Thus, their offense was well above average without Carmelo (+4.93) and their defense was below average (-3.83), but not as far below average as they have been with Carmelo.

Again, its early and this analysis is therefore unfair.  So take it for what it is worth. 

The Carmelo Effect

I want to comment on the so-called “Carmelo Effect” that was discovered by some author a bit ago.  Essentially, the author pointed out that playing with Carmelo Anthony tended to raise a player’s shooting percentage.  The author suggested the “Carmelo Effect” was the result of Anthony’s scoring prowess — drawing double teams and the like.

This supposed effect has to be bunk.  At least, I am not convinced it exists, chiefly because I cannot see any connection between Carmelo’s ballhogging and his teammates shooting effectiveness. 

First of all, Carmelo rarely passes the basketball.  If he is double teamed, he will often force shots up anyway.  How does that help his teammates get better looks?  If anything, I would think it would tend to frustrate them and lead them to force up shots indiscriminately simply to “get theirs”.  (This is what coaches like George Karl refer to when they talk about “trusting the pass”.  Players want to get “x” amount of shots up per game, and will force their attempts rather than passing if they do not believe the ball will come back to them)  

I will buy a certain “improvement effect” where the player in question is a playmaker.  Carmelo is not a playmaker.  But “usage” fans always argue that high usage players make their teammates appear more efficient because the high usage players take the tough, low percentage shots that would otherwise lower their teammates shooting averages if they were forced to take those attempts.

The Usage Debate

Here’s my take on the Usage Debate.  Let me lay out the “high usage” proponents basic argument as best I can, and to do that I will use the movie Glengarry Glen Ross as an analogy (if anyone thinks I’m describing a straw man version of the argument, please let me know). 

In that movie you had a group of real estate salesmen who worked off “leads” — names of people who had expressed some degree of interest in purchasing their product.  The leads, however, varied in quality.  Some of the leads identified potential customers who had either expressed a low interest or who had backed out of past transactions.  These were people who would likely be difficult for the salesmen to “close”.  These were the “weak” leads.

Then there were the “Glengarry” leads.  These leads were described by Alec Baldwin (in one of the greatest scenes in movie history) as “gold”, meaning they contained the names of people who were very likely to consummate a transaction.

Now, lets forget for a moment that the firm of Mitch & Murray apparently reserved these Glengarry leads for their best salesmen.  Lets assume the firm of Mitch & Murray acted rationally and distributed the Glengarry leads to their weakest salesmen while tasking their best salesmen with closing the weaker leads.

Now at the end of the month Mitch & Murray came out with raw sales statistics.  These stats seemed to show that the weaker salesmen were more “efficient” than the better salesmen.  In this situation, the better salesmen would argue that this interpretation is grossly unfair.   They would argue that their sales numbers ought to be augmented because they were required to chase the weak leads whereas the poor salesmen’s numbers were buoyed by the Glengarry leads. 

In essence, I’ve just described my understanding of the usage argument.   Usage proponents argue that high usage players ought to be given credit for mopping up many of the more difficult shot possessions while low usage players ought to be debited for cherry picking the easier shot possessions. 

The problem with that argument in my mind is that it presumes that in any given game there will only be “x” amount of high percentage shots (the Glengarry leads).  In other words, a team cannot avoid taking a certain amount of “tough” shots.  I don’t accept that central premise except in one situation.      

If a player is forced to take an inordinate amount of shots when the shot clock is running down, then okay, I will credit him for that.  He’s basically taking a suck for the team.  But according to 82games that really doesn’t happen very often.  It seems that most every player takes about 13% of his attempts within 3 seconds of a shot clock violation — even Kobe Bryant.  So its not as if the Lakers are saying “if you can’t find a shot, and the clock is running down, find Kobe!”  Its simply doesn’t happen that way (in fact, Pau Gasol takes a higher percentage of “last 3 seconds” shots than Kobe does).

If you eliminate those kind of “involuntary” desperation shots, then you are left with only voluntary attempts.   In other words, low percentage shots that the “high usage” player affirmatively choose to take rather than passing the ball.  When high usage players take such shots, they preclude the possibility that the team will produce a higher percentage shot. 

Why should the player be credited for such a decision?  How can he know at the time that the team would not be able to produce a lay-up or open jump shot out of the particular possession?  How does he know at the time that the best shot the team can produce in the particular possession is a spinning, fade away jumper?  Unless the player has special clairvoyant skills, I don’t believe he can. 

And that is why I just don’t buy into the idea that players who are willing to use a high percentage of a team’s possessions have any intrinsic value.  I don’t think they do.  Anyone can miss a shot or fumble away the basketball. 

Thus, my position remains that high usage players have great value if and only if they combine that high usage with high point conversion. 

One of these days I’m going to calculate each player’s “points over attempts” average per game.  This will identify the best scorers.

6 Responses to “The Carmelo Effect and the GlenGarry leads”

  1. arturogalletti Says:

    This, a thousand times this. Great post Ty!

  2. Max Says:

    I mostly agree with this post, but I have one other question. What about the defense? Is it possible that Carmelo’s presence on the floor means that the defense have to play more honestly when Carmelo doesnt have the ball, whereas with a less talented offensive player, they could sag off of him and put more pressure on the ballhandler meaning that other teammates have easier shots even when they didnt get a pass from Melo?
    (Note: I’m not sure if this applies to Melo as much as it does say Bron and D-Wade, but those players also pick up a ton of assists so it would be harder to disentangle the different effects)

    • tywill33 Says:

      Here’s how I view that, Max. NBA teams rarely overtly double team any player who is not set up in the post. Meaning, its rare to see a player — any player — get double teamed on the perimeter (its simply too easy to beat such a defense).

      Therefore the effect you’re describing would have to be subtle — almost a leaning of the defense. That might be happening, but I would argue that it happens all the time. Players are taught to “hedge” to the ball no matter who is possessing it. Thus I don’t see how a Carmelo Anthony’s presence would cause any different arrangement.

      The player who really provides better shots for his teammate is the penetrate-and-pass player. Those type players cause what Van Clausewitz called “friction” within a defense, meaning penetrators cause a defense to react to the penetration and this “reaction” almost always results in a certain amount of chaos. That chaos can often be exploited, especially when the defense is not disciplined.

      That is where I would expect an effect. Carmelo would not produce such an effect because he doesn’t like to pass. So where his effect would come from, I don’t know.

  3. jbrett Says:


    This post is dead on, in my opinion. A question: The favorite WoW argument last season was that Boston lost because LA didn’t have to defend Rondo outside the lane, and Kobe (primarily) could hedge toward the shooters. The implication is that a Derrick Rose, or Jameer Nelson–really, any PG with more range–would have been more valuable because he would have ‘kept the defense honest.’ Any thoughts on refuting this line of unreasoning?

  4. Carlos_XL Says:

    How about the fact that Melo simply always commands the best defender on the floor? Occam’s Razor.

    I also think the “hedge” effect is real – the big men especially have one eye on star players all game. Here’s a testable hypothesis: if the hedge effect is true, players that take a high percentage of their shots at the rim should result in the highest TS% improvement amongst their teammates (let’s call it the Silver Effect, after Nate), since help defense occurs more often at the rim than at the perimeter.

    Of course, that doesn’t describe Melo, who loves to take long jumpers over tight defense.

    • tywill33 Says:

      He might draw the best G/F defender, but not necessarily the best defender. The Celtics best defender is Rajon Rondo. You will never see Rondo on Anthony in any setting.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: