The NBA has essentially reached the midpoint. As of last night, only five teams – Boston, Charlotte, Chicago, Golden State, and Milwaukee — have yet to play 41 games. As each team reaches the 41 game mark I have measured the productivity – via Wins Produced – of each player these teams employ. After this weekend I will analyze the remaining five teams, and then I will offer a series of posts analyzing what the NBA looks like at the midpoint of the 2009-10 season.
As we wait, though, let me offer a brief comment on the MVP race. And here it is:
For those who think Carmelo Anthony is the most productive players in the NBA… well, I ain’t seeing it. Carmelo leads the NBA in scoring, but after 41 games Melo has only produced 4.5 wins and posted a 0.160 WP48. Yes, Melo is above average. But I don’t think he’s an MVP.
What about Kevin Durant? He has clearly improved from what we saw his rookie season (and yes, I still think he improved during his second season). Currently he is third in the NBA in scoring and after 41 games he has posted 8.2 Wins Produced. And although his WP48 stands at 0.240, he’s also not the most productive player in the game.
And then there is Kobe. He’s fourth in the NBA in scoring but has only produced 6.0 wins after 41 games. And although producing about twelve wins in a season is very good for a shooting guard, it isn’t going to lead the NBA in 2009-10.
So who is the most productive player thus far in 2009-10? Well, I will discuss that and other topics… starting on Sunday.
- 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.
Random question: Why don’t you count charges the same as steals? It seems weird to me that players don’t get more credit for a charge. Kyle Lowry has about 30 this year. If those each counted as steals — and they seem to be the functional equivalent — his great numbers would look even better. Same thing with Battier and Hayes.
Good idea. Give me data on charges drawn for every player I will include these.
That would be the reason. I don’t have the data. But you are right. Drawing a charge is just like a steal (and I don’t think the NBA counts it that way).
From where has the idea come about that a deserving choice for the NBA’s MVP Award should be the “statitistically most productive” player over the course regular season?
khandor,
That is the criteria I use to determine Most Productive Player (MPP). And I tend to think that since players who produce the most wins are the most valuable, the MPP should probably be MVP. Of course, on your website you can argue whatever you like.
Beyond the issue of not having the data: a charge would have the added benefit of giving the opposing team a foul, but the disadvantage of not allowing for a fast break (which I’m guessing typically lead to better scoring opportunities?). So I’d bet that charges would need to be put in the model as a check, like assists were, and would come out as being pretty much the same as a steal plus the absolute (positive) value of a foul.
re: “That would be the reason. I don’t have the data. But you are right. Drawing a charge is just like a steal (and I don’t think the NBA counts it that way).”
Technically, from a pure basketball perspective, drawing a charge is not exactly the same thing as generating a steal … although both do qualify as different means of creating a Turnover.
i.e. A “steal” properly understood is worth slightly more than a drawn charge, as the ball is already in your team’s possession without needing to be inbounded, first and foremost. In specific situations there are fundamental advantages in a Steal-Turnover which do not exist in a Drawn Charge-Turnover. :-)
Hoopdata.com tracks charges.
David,
A follow-up clarifying question for you:
On the old Boston Cetlics teams of Auerbach, Russell, Cousy, Jones, etc., which one player would have been seen as the “statistically most productive player”? … according to the metric used by the Wages of Wins … and, Is this player the same one that usually won the League MVP Award?
Thanks, in advance.
Khandor,
I am not trying to predict who will win the MVP. I am trying to discuss — briefly — who is not the current MPP. I would argue that MPP and MVP would be the same. But having studied the voting for MVP, I know they tend not to be.
Jason Kidd is having a tremendous season (even ignoring his age).
My guess would be MPP would be a fight between Josh Smith, LeBron James, Pau Gasol, and Tim Duncan. Perhaps Gerald Wallace making an appearance, after Charlotte’s recent streak? I don’t have your stats, but those seem to be the players who come to mind who would do well in this metric with their performances this season. Looking forward to the analysis.
Dberri,
A few entries ago I mentioned the “drawing offensive fouls” statistic because it was apparent to me that Jared Jeffries of the Knicks adds value that way. The reason I brought it up was because I saw a note somewhere that it is tracked and will probably eventually become part of the official stats. I just can’t remember where I saw it.
David,
Then, perhaps the title of this article should read as, “Who is not the MPP in 2010?”, instead of, “Who is not MVP in 2010?” for the purpose of increased accuracy.
When you speak of MPP, then, I can understand exactly what you mean, and not get it confused with the perception of which player should [or will] get the NBA’s MVP Award Winner this season.
David,
re: I would argue that MPP and MVP would be the same.
I would be very interested to hear your thoughts and ideas on this exact subject.
If need be, I have no problem giving you my ideas first … as whichever way works best for you, is fine with me.
I don’t know if you checked the value of offensive vs. defensive rebounds, but if you eventually make a change for charges, I would also consider taking a closer look at that.
It seems to me that offensive rebounds lead to a higher level of Points Per Possession because many result in easy tip ins, put backs etc… When they don’t, they are similar to other possessions anyway.
Duncan is having a great year, but I’d guess Lebron is still #1. Could be close though.
Yes, in hoopdata.com we have at least one site that tracks charges. 82games.com also tracked charges at one point.
Here’s a direct link to the page: http://hoopdata.com/defrebstats.aspx
“I don’t know if you checked the value of offensive vs. defensive rebounds, but if you eventually make a change for charges, I would also consider taking a closer look at that.”
It’s been brought up. No evidence that offensive rebounds contribute more to wins than defensive (that I recall). Someone created an alternate wp48 formula that weighted ORBDs heavier than DRBDs.
Charges aren’t exactly the same as steals. Steals will result in more easy fast break points, so they should correlate well to efficient scoring.
Of course, drawing charges brings the opposing team closer to the penalty, and increases the chance of the player’s team getting to the foul line, so there should be a correlation between drawing charges and efficient scoring as well, though I’d expect it to be less strong.
Yep. A DRBD prevents your opponent from getting a ORBD, thus precluding the easy tip-in, layup and so on–so it seems like it should be equally valuable.
My best guess is that the WoW metric will reveal that Lebron James has been the MPP at the season’s mid-point.
i think it could be dwight howard. he always does really well in wp. probably lebron though.
With Chris Paul’s injury earlier in the season I will assume Lebron has been the most productive so far.
It will be interesting to see if Josh Smith is still up there after his incredible start.
In reference to the charging/steals issue; if on average each possession is worth one point then – regardless of a need to inbound the ball or the efficiency of a fast break etc – both steals and charges have essentially the same value, since they gain your team a possession without conceding a basket.
What happened to the second highest scorer in the league Dave?!!!??!….
Dr. Berri, so regarding shooting guards….Is it that much harder for SG to get above ~.200WP 48 compared to other positions? Italian Stallion mentioned that perhaps SG’s tend to be ‘undervalued’ when it comes to production. I would consider a SG who’s 6’1-6’5 to be very very good with a WP48 of say .150
You know who I think might sneak in as MPP for the first half of this season?
Marcus Camby
Duncan, Howard, James, and Bosh should be in the conversation too.
A few players in the next tier who deserve honorable mentions: David Lee, Joakim Noah, Al Hortford, Troy Murphy, and Samuel Dalembert. All those guys are underrated.
docrostoy,
I’ll second Gerald Wallace having a shot. WP has rated him pretty highly in past, and he’s doing basically what he did last year (when his WP was over .250), plus pulling down 3.5 more boards a game. That’s a huge net possession increase.
khandor,
Though the box score from Russel’s Celtics days isn’t as comprehensive as today, I seem to remember that he getting about 20 rebounds a game, led the league in FG%, and people estimated that he blocked easily half a dozen shots a game. Hard to see him not leading his team in WP.
bagsflyfree,
What do you mean, hard? Jordan had a WP of .500 one year, and I believe he was playing SG. Subjectively, I can see Durant, Roy or Kevin Martin all having a chance to exceed .300.
The std dev for SGs does seem quite low, especially compared to other positions. So yes, I think there is some merit to the concept that it is more difficult for SGs to exceed relative to their peers.
Then again, maybe players become SGs because they’re ball dominating scorers. I know that’s the stereotype. And players such as this are rarely productive – at least, from a WP standpoint. They’ll make bank, though.
Tindall,
I recently read an article that provided some evidence that OREBs are more valuable than DREBs. Some of the math was outside my range, but if I can find it I’ll provide a link.
Phil,
“What do you mean, hard? Jordan had a WP of .500 one year, and I believe he was playing SG. Subjectively, I can see Durant, Roy or Kevin Martin all having a chance to exceed .300.”
IMO the reason that Jordan had such a high WP (aside from being above average at almost everything) was that he was probably the best player ever at getting to the hoop and finishing from the G position. So he was able to score way more points than average at way above average levels of efficiency relative to a typical SG or even PG .
However, he wasn’t a great outside shooter and IMO people should consider that in the overall evaluation. In fact, in some ways he was a below average outside shooter until later in his career.
So the Bulls still needed the outside shooting a team usually gets from their SG that he couldn’t provide at way above average rates. That’s why they teamed him up with guys like BJ Armstrong and Steve Kerr. The two created a deadly combination because they were both hyper efficient at one of the scoring requirements.
IMO, this is why positional adjustments sometimes fail.
To me, skills are skills. It doesn’t matter which position they come from. Certain skills just “tend” to be clustered at certain positions and lead to higher efficiency ratings.
Re; offensive rebounds. It may have been the prof’s work. I recall that teams with more offensive rebounds tend to lose more often; poor shooting=more offensive rebounds.
I think the intent and point of this post is pretty clear: who are the guys that are currently being discussed as MVP candidates by all the sports writers/bloggers/etc./ that are not as productive as their gaudy scoring totals suggest?
Dr. Berri has done this multiple times and provided rationale for it in the past.
The title of the post is obviously a hook — meant to grab one’s attention. Given the context (i.e., this blog and Dr. Berri’s book), I think it’s pretty obvious that it’s not misleading at all.
Steals are an over-rated part of the game. More often than not, players are out of position and defenses break down because of failed steal attempts. It’s also a high variance play—like presses and traps. If it doesn’t succeed the consequences are dire. They don’t compare to the value of drawing a charge; protecting the basket (admittedly less tangible), gaining possesion of the ball and assigning a foul to the other team.
Ima go with Chris Paul.
todd2,
From a purely basketball perspective, “Failed Steal Attempts” are in a different category altogether from “steals”.
The 1st category demonstrates relatively “poor” judgment, as an elite level player; while, the 2nd category reveals the exact opposite attribute about a specific player, re: his actual Basketball IQ/Judgment.
In terms of equating with actual “points scored” in a basketball game, IMO, “a steal” is worth more than a “drawn charge” … although exactly how much more is certainly open to debate, given the other +’s and -’s associated with each.
brgulker,
re: “The title of the post is obviously a hook — meant to grab one’s attention. Given the context (i.e., this blog and Dr. Berri’s book), I think it’s pretty obvious that it’s not misleading at all.”
… and, a terrific “hook” it is.
I wasn’t trying to suggest that David was somehow being disingenuous with the title of this piece; only that, in the interests of further clarity, it might help others [like me, for instance] … especially, newcomers to the site … to see that it is actually David who is trying to make the case that the “real” League MVP should be determined by something like the metric involved in generating his MPP scores, according to the model used by the WofW.
Very much like David, himself, I am someone who is concerned with bringing increased clarity to a situation which is otherwise opaque.
I’d like to see WP48 after 41 games for Nash, Kidd, Ben Wallace and Duncan, for their age they are playing some inspiring basketball, it’s a joy to watch players like that, not as athletic as they used to be but high understanding of the game still allows them to be an asset to their teams.
khandor,
Gotcha. I just ran a quick search of this blog: http://dberri.wordpress.com/?s=mvp
I think if you have the time to even a handful of these, it will become quite clear that this particular post is just one of many that argue that players who are the most productive are not always perceived to be as productive as they actually are.
A recent example that I enjoyed was the Chris Paul article.
@ bags fly free:
Here’s an article on Ben Wallace through 20 games: http://dberri.wordpress.com/2009/12/07/big-ben-cannot-save-detroit-from-ugh/
I can’t say with certainty, but I think his production has slipped just a little bit since … he tends to struggle (and have his minutes limited) especially in back-to-back games, which is to be expected.
He’s been a joy to watch, though.
Nash, Kidd, Stockton…any other PGs who were that productive at such an advanced age?
Pingback: Disagreeing with Doug Moe « The Wages of Wins Journal
Pingback: MVP Race Preseason « Nerd Numbers the Blog