Don’t Be Deceived by Carmelo Anthony’s Scoring Totals

(Photo: Keith Allison)

Here is how the Associated Press led the story describing the Miami Heat’s elimination of the New York Knicks in the 2012 NBA Playoffs:

The final horn sounded, and LeBron James wrapped his arms around Carmelo Anthony in a warm embrace.

Their head-to-head scoring matchup in this series was even, 139 points apiece.

Just about everything else tipped Miami’s way — so the Heat are moving on and the New York Knicks are going home. 

Such a lead gives the impression that Carmelo Anthony and LeBron James played about the same in this series.  If we delve a bit deeper, though, we see that the scoring totals are quite deceptive.  Here is each player’s level of shooting efficiency in the series: 

Carmelo Anthony: 0.435 Effective Field Goal Percentage, 0.489 True Shooting Percentage

LeBron James: 0.517 Effective Field Goal Percentage, 0.604 True Shooting Percentage

Because Melo was a far less efficient scorer, he had to attempt 34 more shots from the field than LeBron in the series.  And because shot attempts are a finite resource, this means that other players on the Knicks had to attempt fewer shots (with the exception of J.R. Smith, everyone else on the Knicks who attempted at least 10 shots from the field was more efficient than Melo in this series). 

In contrast, LeBron was able to achieve his scoring total with fewer shots, giving him more opportunities to set up his teammates.  One can see this clearly when we look at assists.  LeBron finished the series with 28 assists while Melo only had 11. 

Melo did finish with more rebounds and fewer turnovers.  But when we consider the vast differences in shooting efficiency and assists we see that King James – as has been the case throughout each player’s respective careers – had a far bigger impact on his team’s fortunes. In fact, Andres Alvarezat the Wages of Wins Journal – notes that LeBron led the Heat in Wins Produced during this series (he did the same for the regular season).  In contrast, Melo’s production of wins – because he was a very inefficient scorer – was in the negative range.

Such a story highlights an important lesson: scoring totals in the NBA can be quite deceptive.  A player can boost his scoring totals by simply taking more and more shots. But if this shooting is inefficient, teams actually suffer from this choice.  Of course, as the AP story illustrates, the inefficient scorer’s reputation often does not decline from this choice.  Hence we see this story repeat itself over and over again in the NBA. 


Frank

#1 - nice self-referral to the site devoted to what you think about basketball
#2 - Andres's analysis lost me when he said "a very good Landry Fields", who when on the floor allowed the opposing D to play 5 on 4. Any statistical system that suggests that Landry Fields was "very good" this year and that Troy Murphy was EVER in the top 20 players in the league has some serious problems.

Jack

Dave,

Carmelo was doubled the entire series. No one on the Knicks takes the pressure off of him. LeBron was not doubled because the Knicks had to control Wade as well. Obviously, LeBron is a better overrall player, but Carmelo is just as lethal as a scorer in a half-court offenses as James. If not better.

Jonathan Cantor

Great point, which is the irony here that these "metrics" capture these things. The actual organizations probably has these, while ESPN probably is too lazy to calculate them because they are not as sexy of a story

Jonathan Cantor

You know what the problem is with your argument? The fact it does not take into account the fact that Melo was being defended by two of the best wing defenders in the league (Shane Battier and Lebron James). While Lebron was guarded by Melo (an average defender at best). But of course the reason actual researchers don't engage in these subjects (only pop statisticians) is because there are too many endogenous variables to not be considered. Thats the inherent flaw in the work done by guys like John Hollinger (who I still read avidly).

Mike

How does it not take defense into effect? If Melo is facing a good defense the entire time, perhaps that's why he was such a poor shooter. All the more reason to dish the ball off to a teammate.

Jonathan Cantor

I agree with you but thats the problem no? The fact we can't really say anything based on the actual subject since there are so many factors going on at once. You might be able to say some broad descriptives but nothing definitive.

Jon K

I don't understand why Wage of Wins has any place on this blog. To put it mildly, it is pseudo-science at best. This is what happens when you let someone run a regression model without understanding the inputs. I've posted on Mr. Berri's blog several times, but he seems unable/unwilling to discuss the severe limitations of his model. It is no surprise that he has created one tool that accounts for only the most basic of statistics (box score stats), and flat out refuses to consider metrics that use "results when player is on court" vs "results when play is off court". Doing so would cause him to realize basketball is much, much more complicated that his model appears to be.

Specifically in the case of Carmelo Anthony, Nate Silver, statistician extraordinare who forecasts elections and writes for the NYT thoroughly thrashed Mr. Berri's analysis. Anyone interested should search for such and see exactly why Mr. Berri's simple model is ineffective. The basic result of Mr. Silver's analysis showed that while Carmelo's shooting percentage is average, his teammates shooting average is significantly higher when he is on the court. He creates positive externalities for his teammates which Mr. Berri's model does not account for. As an avid basketball fan and an economist, it saddens me to see the level of subpar analysis offered by wage of wins continually referred to in major publications.

Read more...

jaeicsl@gmail.com

another reason for Melo to pass rather than just take stupid double teamed shots then isn't it? I mean your argument is that when Melo is on the floor his teammates shoot better, so if Melo is double teamed, then someone who shoots 'better' than Melo at that point is open, yet Melo does not pass, he just shoots....

Justin

This is nothing people who care about basketball already know. And Wins Produced? You picked the worst and least respected individual player rater.

HB

For those who say that it's not a fair comparison because Carmelo was being doubled by Battier and James the whole series - the same results are seen over their careers:

Carmelo's career EFG% and TS%: 47.8, 54.4
Lebron's career EFG% and TS%: 51.6, 56.9

http://www.basketball-reference.com/players/a/anthoca01.html
http://www.basketball-reference.com/players/j/jamesle01.html

I'm a lifetime NYK fan and this just proves statistically what I know intuitively. Melo is not a transcendant player like Lebron, Wade, Kobe, Durant, etc. He's overrated. He is good enough to have some amazing offensive games but not good enough to be the leader of a championship team.

Melo sucks motivation and points out of his teammates (remember Amare and Fields last year?). He insists on being the focal point of the offense regardless of the talent or system around him, but delivers less than half of the time. The media continues to place him alongside the aforementioned superstars because he was touted as Lebron's equal out of college and can still muster a Sportscenter highlight or two a week.

I preferred the high energy, quick ball movement and team play under Lin and D'Antoni. The team was gelling, learning to work together, and winning. The only guy that was unhappy was Melo, we rearranged our entire offensive for him, and now we have another first round exit.

Read more...

Jacob

What Jon K said.

Bill Keating

Mello is known agent of the "volume shooting class". Westbrook of OKC is often put in this class although in the last series he was much more efficient. Also known as "black holes" or offense stoppers these players can put up showy scoring stats while severely disadvantaging their teams. Colin Cowherd had some great stats on how Mello's scoring has little correlation to a Knick win. It would be interesting to see if a volume shooter/scorer can exist on a winning team, especially playoff basketball in which possessions are much more scarce.

keith

I was going to ask why this hadn't been moneyballed out of the game, as perhaps having star players is worth more to a franchise than actually winning, but the other comments suggest it's just a bad use of stats. That's a shame because it looked interesting.

Mike

Or perhaps players like Melo are still good enough to deserve a maxed out contract. Since there's an artificial cap on how much a single player can be played, two guys might both deserve maxed out contracts even though one is better than the other.

Calvin

One thing to consider about this, though, is that Melo was forced to take more shots than his teammates. The offense called for him to shoot low-efficiency isos at the end of quarters and the game that he probably wouldn't have chosen for himself. In addition, when playoff defense ramps up and nobody else can create their own shot, it falls on the playmakers to do so to a ridiculous degree, which also lowers efficiency in a way that is more a roster-construction issue than anything

David Quinones

Well-known secret to hoops fans and reporters. But impossible to apply in a Moneyball methodology because it is too situational and the polynomial equation of a basketball team's DNA is far different than baseball, where it really boils down to one pitcher and one hitter.

For example, consider the 2007 NBA Finals, where the better team, clearly, was the San Antonio Spurs. Tony Parker and Tim Duncan were the most efficient players on the court, and Parker won Finals MVP. The least efficient, as it usually is, was the clearly inferior team's best player -- in this case, the Cavalier's LeBron James, who outscored both Duncan and Parker. But James' inefficient scoring was worth exactly zero finals wins in that series. Now, with a clearly superior team, his game is afforded space to be more efficient. So, it's all situational.

Zack

Basketball advanced statistics are very tricky. In baseball, the game has a finite number of states it can take based on the number of outs, the inning, and the configuration of runners on base. Basketball is fluid and continuous. You really need a certain amount of video analysis to get a real sense of what was going on in an NBA game. It's possible Carmelo Anthony, who was surrounded by a worse team than LeBron, was forced into less than optimal shots because he was being double-teamed. You can't double team LeBron when Dwyane Wade and Chris Bosh are also on the court.

Caleb b

hey Melo-defenders: open your eyes! Melo chucks up the rock every chance he gets and misses all the time. Westbrook was really bad about this last year and he only got better about passing bc he has Durant. People were hounding him day and night to pass and the peer pressure got to him (well that and maybe the long term deal he signed in the offseason).

Melo is diet version of Allen Iverson. Tons of skill, but too much of a ball-hog to make his team better.