Is pau statistically more important than kobe?

Moderator: Doctor MJ

User avatar
dockingsched
Retired Mod
Retired Mod
Posts: 51,734
And1: 10,717
Joined: Aug 02, 2005
     

Is pau statistically more important than kobe? 

Post#1 » by dockingsched » Thu Nov 5, 2009 6:18 am

seeing the lakers struggle to win games without pau even though bynum is putting up 20/11 and kobe is putting up 30pt+ games had be reflecting on pau's role on the lakers...

why do these two stats show that pau gasol was more critical to the lakers winning games than kobe? what can be taken from these stats?


http://www.wagesofwins.com/2009MVP.html

wins produced:
gasol 16.8 vs. bryant 13.9


http://www.basketball-reference.com/pla ... nko01.html
http://www.basketball-reference.com/pla ... lpa01.html

wins shares:

gasol 13.9 vs. bryant 12.7
"We must try not to sink beneath our anguish, Harry, but battle on." - Dumbledore
TrueLAfan
Senior Mod - Clippers
Senior Mod - Clippers
Posts: 7,508
And1: 433
Joined: Apr 11, 2001

Re: Is pau statistically more important than kobe? 

Post#2 » by TrueLAfan » Thu Nov 5, 2009 1:11 pm

In a nutshell, this is what is all wrong about statistical analysis with basketball. We're asking the wrong question. If we have a statistical analysis system that determines that Pau Gasol is more important to the Lakers than Kobe Bryant, people (specifically, the inventors of the system) want us to ask the question "What does this tell you about the players?" We should be asking, "What does it tell you about the system?" We don't ask the question, because the answer would be "Well, it must have some problems."

Not that I don't like David Berri and company...they do some good neat stuff. But have a look at that list. Calderon over Bosh? Kidd over Dirk? McDyess over, well, a lot of people? (McDyess played less than 1900 minutes, for God's sake.) Miller over Jefferson? Mike Conley over...well, almost all the other starters? I see a bunch of pretty obvious howlers in there.
Image
drza
Analyst
Posts: 3,423
And1: 1,639
Joined: May 22, 2001

Re: Is pau statistically more important than kobe? 

Post#3 » by drza » Thu Nov 5, 2009 9:10 pm

TrueLAfan wrote:In a nutshell, this is what is all wrong about statistical analysis with basketball. We're asking the wrong question. If we have a statistical analysis system that determines that Pau Gasol is more important to the Lakers than Kobe Bryant, people (specifically, the inventors of the system) want us to ask the question "What does this tell you about the players?" We should be asking, "What does it tell you about the system?" We don't ask the question, because the answer would be "Well, it must have some problems."

Not that I don't like David Berri and company...they do some good neat stuff. But have a look at that list. Calderon over Bosh? Kidd over Dirk? McDyess over, well, a lot of people? (McDyess played less than 1900 minutes, for God's sake.) Miller over Jefferson? Mike Conley over...well, almost all the other starters? I see a bunch of pretty obvious howlers in there.


I think that, in a nutshell, it's what is all wrong about trying to use any one stat as a holy grail "basketball goodness" stat, and that the down-side is that a lot of knowledgeable basketball fans therefore don't give enough credence to statistical analysis. In reality, there is a lot of outstanding statistical info available that really could help shape how we watch and understand basketball...if it's only used correctly.

Rule number one to any statistical analysis: a stat can only answer the question that it is asked. A stat is like a binary computer...it is black and white, not interpretation. The interpretation is left to the person that uses the stat.

So the actual Wins Produced stat, for example, does NOT say that Pau is a better basketball player than Kobe. It does not even say directly that he is a more important player than Kobe. What Pau's higher Wins Produced number REALLY says is that Pau measures out as a more EFFICIENT basketball player according to the method used here. That's it. Now, Dr. Berri has attempted to correlate his stat to previous NBA success rates, but once you enter that arena you are getting into Dr. Berri's INTERPRETATION of the stat. NOT the stat itself.

So, on topic, I say that Pau having more Wins Produced and more Win Shares doesn't of itself say that he is more statistically important than Kobe. Like Wins Produced, Win Shares is also a measure of EFFICIENCY though it has some production and impact components as well. PER is also a measure that incorporates efficiency and production. The +/- family of stats all measure in some way how a player's presence correlated with their team's success. All of these stats are different (some slightly, some completely) and it is by understanding all of them and considering what all of them are saying...in conjunction with your own observations and knowledge, that IMO the most educated conclusions can be reached.

But the last sentence is the key: YOU have to be the one to supply the final analysis. Because yeah, looking only at any one stat or even looking at multiple stats incorrectly can lead to some funky results. But if the same player measures out as the most efficient, the most productive, AND with the biggest impact on the team's output then perhaps there is some validity to what the stats are telling you. Because despite what we all think of our own opinions, they are extremely fallible as well. So sometimes it is worth keeping an open mind that perhaps your conclusions might not match up entirely with successful basketball.
Creator of the Hoops Lab: tinyurl.com/mpo2brj
Contributor to NylonCalculusDOTcom
Contributor to TYTSports: https://www.youtube.com/playlist?list=PLTbFEVCpx9shKEsZl7FcRHzpGO1dPoimk
Follow on Twitter: @ProfessorDrz
wreck
Banned User
Posts: 512
And1: 0
Joined: Oct 29, 2008

Re: Is pau statistically more important than kobe? 

Post#4 » by wreck » Thu Nov 5, 2009 10:49 pm

Because the general public has been so brainwashed by the media's love affair for Bryant, they can't possibly believe what the statistics say, and that is Gasol is the team's most valuable player.

Just a common case of perception being greater than reality.
User avatar
Wile E. Coyote
Assistant Coach
Posts: 3,767
And1: 94
Joined: Apr 29, 2008
Contact:

Re: Is pau statistically more important than kobe? 

Post#5 » by Wile E. Coyote » Fri Nov 6, 2009 4:32 am

wreck wrote:Because the general public has been so brainwashed by the media's love affair for Bryant, they can't possibly believe what the statistics say, and that is Gasol is the team's most valuable player.

Just a common case of perception being greater than reality.


So the Lakers would be better if Kobe were injured right now and Gasol wasn't?
Jimmy76
RealGM
Posts: 14,547
And1: 7
Joined: May 01, 2009

Re: Is pau statistically more important than kobe? 

Post#6 » by Jimmy76 » Fri Nov 6, 2009 11:22 am

Wile E. Coyote wrote:
wreck wrote:Because the general public has been so brainwashed by the media's love affair for Bryant, they can't possibly believe what the statistics say, and that is Gasol is the team's most valuable player.

Just a common case of perception being greater than reality.


So the Lakers would be better if Kobe were injured right now and Gasol wasn't?


I think because Vujacic is Kobe's backup and Bynum is Pau's thats probably not true

Then again they can move Artest to SG and use him as the wing creator so maybe not

Maybe Kobe will get hurt and we'll get a chance to see but it seems like there is an argument there since Pau is so ridiculously efficient and a great passer so running the offense through a post player like that instead of a wing scorer might help the team offense more
TrueLAfan
Senior Mod - Clippers
Senior Mod - Clippers
Posts: 7,508
And1: 433
Joined: Apr 11, 2001

Re: Is pau statistically more important than kobe? 

Post#7 » by TrueLAfan » Fri Nov 6, 2009 7:36 pm

No. Just…no. Berri (and let’s not give credit for inapplicable titles…my doctorate and three graduate degrees don’t necessarily relate to basketball any more than Berri’s Ph.D. in economics does) is not talking about relative player efficiency. Wins produced are absolutes…totals for all the players on a team should add up to the actual team wins. That’s why to get the amount, actual minutes played are used…the formula is WP48 / 48 * Minutes Played) It’s true that offensive and defensive efficiency are used to determine WP48…but that’s neither why Berri created Wins Produced nor how it is applied. It is means to be assessment tool (to break down individual player impact on total team wins) and comparative tool (between the total WP for individual players).

In this case, this is abundantly clear. Berri clearly making a reference to the more/most valuable players on individual teams and in the league--hence the title “The 2009 Most Valuable Players” and notation for “Non Top Players that Received MVP Votes. Trying to say that an analysis titled “The 2009 Most Valuable Players” that lists players by “Wins Produced” is actually talking about efficiency is wrong. In the book The Wages of Wins—which is a great read, as I said, Berri and co. have extensive discussion involving total player wins produced—which is the ranking criteria on the linked page—and correlate it to overall player value and importance. So, yeah, “Top Player's Wins Produced” is all about who is a better/more important/more valuable basketball player.

So if it’s true that most use statistical analysis “in conjunction with your own observations and knowledge,” and through that “that … the most educated conclusions can be reached,” what does it say about anyone who uses a statistical analysis that says that Jason Kidd was worth 13 more wins—a year—over 250% more—than Dirk Nowitzki in 2009. Kidd is a savvy vet who still dishes the ball well, has developed a good three point shot, and rebounds as well as any PG in the league. He doesn’t score much, has lost a step on D, still shoots under 42% from the field, and doesn’t get fouled because nobody respects his midrange game. I’m not going to go into the nuts and bolts of faulty WP analysis—Dan Rosenbaum, King Kaufman, and a bunch of people at APBR and other sites can handle that if they like. I’m contesting the idea that Berri is doing something in conjunction with observation and knowledge, or why we should look at a clearly flawed conclusion...because, let’s be real…there’s no way on earth that Jason Kidd was worth 13 more wins, or 250% more in terms of wins, to the Mavs in 2009 than Dirk Nowitzki. And that's exactly what total WP shows in this case. The question with regard to Kidd and Dirk—or Pau and Kobe—shouldn’t be “What does this tell you about the players?” It should be “What does this tell you about the tool being used?” And the answer is, “It’s got some serious flaws.” And, frankly, Berri is falling into the same trap as Hollinger and the ilk that design a system and spend their time defending the system and its results rather than looking at them to see if they are accurate…because a lot of time, they aren’t.
Image
drza
Analyst
Posts: 3,423
And1: 1,639
Joined: May 22, 2001

Re: Is pau statistically more important than kobe? 

Post#8 » by drza » Sat Nov 7, 2009 5:18 am

TrueLAfan wrote:No. Just…no. Berri (and let’s not give credit for inapplicable titles…my doctorate and three graduate degrees don’t necessarily relate to basketball any more than Berri’s Ph.D. in economics does) is not talking about relative player efficiency. Wins produced are absolutes…totals for all the players on a team should add up to the actual team wins. That’s why to get the amount, actual minutes played are used…the formula is WP48 / 48 * Minutes Played) It’s true that offensive and defensive efficiency are used to determine WP48…but that’s neither why Berri created Wins Produced nor how it is applied. It is means to be assessment tool (to break down individual player impact on total team wins) and comparative tool (between the total WP for individual players).

In this case, this is abundantly clear. Berri clearly making a reference to the more/most valuable players on individual teams and in the league--hence the title “The 2009 Most Valuable Players” and notation for “Non Top Players that Received MVP Votes. Trying to say that an analysis titled “The 2009 Most Valuable Players” that lists players by “Wins Produced” is actually talking about efficiency is wrong. In the book The Wages of Wins—which is a great read, as I said, Berri and co. have extensive discussion involving total player wins produced—which is the ranking criteria on the linked page—and correlate it to overall player value and importance. So, yeah, “Top Player's Wins Produced” is all about who is a better/more important/more valuable basketball player.

So if it’s true that most use statistical analysis “in conjunction with your own observations and knowledge,” and through that “that … the most educated conclusions can be reached,” what does it say about anyone who uses a statistical analysis that says that Jason Kidd was worth 13 more wins—a year—over 250% more—than Dirk Nowitzki in 2009. Kidd is a savvy vet who still dishes the ball well, has developed a good three point shot, and rebounds as well as any PG in the league. He doesn’t score much, has lost a step on D, still shoots under 42% from the field, and doesn’t get fouled because nobody respects his midrange game. I’m not going to go into the nuts and bolts of faulty WP analysis—Dan Rosenbaum, King Kaufman, and a bunch of people at APBR and other sites can handle that if they like. I’m contesting the idea that Berri is doing something in conjunction with observation and knowledge, or why we should look at a clearly flawed conclusion...because, let’s be real…there’s no way on earth that Jason Kidd was worth 13 more wins, or 250% more in terms of wins, to the Mavs in 2009 than Dirk Nowitzki. And that's exactly what total WP shows in this case. The question with regard to Kidd and Dirk—or Pau and Kobe—shouldn’t be “What does this tell you about the players?” It should be “What does this tell you about the tool being used?” And the answer is, “It’s got some serious flaws.” And, frankly, Berri is falling into the same trap as Hollinger and the ilk that design a system and spend their time defending the system and its results rather than looking at them to see if they are accurate…because a lot of time, they aren’t.


Just out of curiosity, what are your degrees in? Always nice to see brains on the board.

On-topic, your post in a lot of ways proved my point. You just gave a long, well thought out rebuttal to the way that Berri interprets his efficiency stat. The actual stat...the core question that it can answer, is "what is the player's offensive and defensive efficiency"? Everything else is Berri's interpretation...

Berri noted a correlation between team efficiency (offensive and defensive) and winning

Berri decided to you could multiply the original efficiency ratio by some constant factor to make it match up with # of wins

Berri decided to make reference to "most valuable players" based on Wins Produced

Berri decided that "Top Players Win Produced" is all about who is a better/more important/more valuable basketball player.

My point is, it doesn't matter HOW or WHY Berri decided to interpret his efficiency stat...the point is that it IS an efficiency stat that has a large emphasis on ending possessions defensively and shooting efficiency, and everything else is Berri's window dressing. So your whole Jason Kidd/Dirk Nowitzki debate isn't a rebuttal to my original point, as I read from the start that Wins Produced is telling me that Kidd measured out as more efficient than Dirk in this measure. That's it.

I can then go and check PER and get a slightly different angle. I don't care why Hollinger developed his stat or that he claims it's the best way to tell how valuable a player is. I care that it's a measure of efficiency and production that has more of an emphasis on scoring.

I can then make the rounds to the other stats that I like, that I know the strengths and weaknesses to. I could care less what the author claims about what the stat can do as they try to sell their books or earn their ESPN paycheck or push their blogs or whatever. If the way the stat is calculated is in the public domain, if the strengths and weaknesses of the actual stat are public domain, then how the author interprets it isn't IMO a good reason to disregard the information that is there.
Creator of the Hoops Lab: tinyurl.com/mpo2brj
Contributor to NylonCalculusDOTcom
Contributor to TYTSports: https://www.youtube.com/playlist?list=PLTbFEVCpx9shKEsZl7FcRHzpGO1dPoimk
Follow on Twitter: @ProfessorDrz
NetsForce
Banned User
Posts: 20,711
And1: 24
Joined: Dec 27, 2006

Re: Is pau statistically more important than kobe? 

Post#9 » by NetsForce » Sat Nov 7, 2009 4:28 pm

All I know is that the Lakers wouldn't be 5-1 if Pau was playing and Kobe wasn't.
Doctor MJ
Senior Mod
Senior Mod
Posts: 41,143
And1: 9,150
Joined: Mar 10, 2005
Location: Treys are for 'Kicks!...and Steph Curry
   

Re: Is pau statistically more important than kobe? 

Post#10 » by Doctor MJ » Sun Nov 8, 2009 2:51 am

Piggybacking a bit on what TrueLA's said, but also moving onto a tangent: To me what's wrong with statistical analysis is the attitude of guys like Berry, Winston, and Hollinger to slightly lesser extent. No matter what the conclusions are from their chosen stat, they never doubt. It's simply ridiculous.

Because we simply cannot do the experiments that would prove any one theory right, the common wisdom of the "yokels" cannot be dismissed, and anything that varies dramatically from that wisdom has to justify itself, not the other way around. Most stats guys get that, but it's so frustrating that some of the names most well known don't.
In another world...
Your Longshoremen of Long Beach...

And now introducing...
The Song of Westeros...
LeBron James
John Stockton
Bobby Jones
Rudy Gobert
Khris Middleton
Connie Hawkins
Otto Porter
Ryan Anderson
TrueLAfan
Senior Mod - Clippers
Senior Mod - Clippers
Posts: 7,508
And1: 433
Joined: Apr 11, 2001

Re: Is pau statistically more important than kobe? 

Post#11 » by TrueLAfan » Sun Nov 8, 2009 3:29 am

Agree with DoctorMJ. The designers of systems have an interest in making sure what they say/show is "right," which causes pretty sezrious tunnel vision. The general response is to jusitfy a conclusion rather than determine why/if it's actually correct.

And I respect what you’re saying drza, but it’s kind of disingenuous.  The moment WP is used to look at more than one player, it becomes a comparative tool—and having read plenty of Berri and co.’s work, that’s almost always how it is used.  I cannot see any possible way to reasonably say that this is not what it was constructed for.  It’s kind of like saying, “We look at On Base Percentage to see how a player is doing…but not to see how well he does compared to league averages or the rest of the team to see if this person should be a lead off hitter, or for contract purposes if he’s going into arbitration.”  The system is designed to be used as a comparative and correlative tool with respect to other player and league averages for the purpose of deriving conclusions of player effectiveness in general and comparatively. 
 
Why is why I agree with your statement about use of statistical analysis “in conjunction with your own observations and knowledge.”  But that statement should be applied to and by designers of statistical models as well.  People who have developed a system that tells them that Jason Kidd was worth 260% more wins to the 2009 Dallas Mavericks than Dirk Nowitzki should not be asking, “What does that tell you about the players?”  They should be looking at the system and asking, “That can’t be right.  We need to fix this.”
 
But that doesn’t happen often—or, at least, often enough.  People become wedded to their modelling and analytic tools. 
Image
drza
Analyst
Posts: 3,423
And1: 1,639
Joined: May 22, 2001

Re: Is pau statistically more important than kobe? 

Post#12 » by drza » Sun Nov 8, 2009 10:02 am

TrueLAfan wrote:Agree with DoctorMJ. And I respect what you’re saying drza, but it’s kind of disingenuous.  The moment WP is used to look at more than one player, it becomes a comparative tool—and having read plenty of Berri and co.’s work, that’s almost always how it is used.  I cannot see any possible way to reasonably say that this is not what it was constructed for.  It’s kind of like saying, “We look at On Base Percentage to see how a player is doing…but not to see how well he does compared to league averages or the rest of the team to see if this person should be a lead off hitter, or for contract purposes if he’s going into arbitration.”  The system is designed to be used as a comparative and correlative tool with respect to other player and league averages for the purpose of deriving conclusions of player effectiveness in general and comparatively. 
 
Why is why I agree with your statement about use of statistical analysis “in conjunction with your own observations and knowledge.”  But that statement should be applied to and by designers of statistical models as well.  People who have developed a system that tells them that Jason Kidd was worth 260% more wins to the 2009 Dallas Mavericks than Dirk Nowitzki should not be asking, “What does that tell you about the players?”  They should be looking at the system and asking, “That can’t be right.  We need to fix this.”
 
But that doesn’t happen often—or, at least, often enough.  People become wedded to their modelling and analytic tools. 


I don't think we're disagreeing as much as you might think, just in the take home message.
You seem to be arguing that you don't like how Berri or Hollinger and their ilk defend their models come hell or high water...and that's fine. That's cool. Whether I agree or disagree with some of their conclusions doesn't even matter, as I agree with you that their systems aren't perfect and shouldn't be presented as such. But where you and I diverge is that I believe that just because the creator may be trying to force us to use their stat in questionable ways doesn't mean the stat itself isn't useful. I agree with you that Wins Produced is a comparative tool, just like On Base Percentage (your example). But I think you are misrepresenting my position in your analogy.

Using your analogy, what I'm saying is that On Base Percentage is one good way to gauge how a player is doing, but that you should also look at slugging percentage and fielding percentage and other stats that measure how the player is performing in different ways. If you look at On Base Percentage ONLY you get weird results like Ben Zobrist is better than ARod and Miguel Cabrera. I don't care if the inventor of On Base Percentage says that this is the ONLY measure that should be used to rank a player, as an analyst I believe that I can get some benefit from looking at On Base Percentage but that it doesn't tell a complete story and must be put into perspective with other independent performance evaluators.

Can you see where what I said is different than what you attribute to me? Your analogy suggested that I'm somehow making excuses for Berri by trying to qualify the circumstances and ways in which to use his stat. That's not at all what I'm saying. What I'm saying is that we can and should apply his stat to every situation ALONG WITH other independent measures, and by understanding the strengths and weaknesses of each measure we use we can weed out the anomalous data and minimize the ridiculous results. And if several independent measures that look at different aspects of the game and covers a fully representative space (i.e. box score stats and non-box score stats, individual and team impact, efficiency and production, etc.) and these measures all tell me the same thing...if the conclusion is counterintuitive, I at least have to consider that maybe my intuition should be re-examined. Doesn't necessarily mean I'm wrong, but that at least it's worth thinking about. That's all I'm saying.
Creator of the Hoops Lab: tinyurl.com/mpo2brj
Contributor to NylonCalculusDOTcom
Contributor to TYTSports: https://www.youtube.com/playlist?list=PLTbFEVCpx9shKEsZl7FcRHzpGO1dPoimk
Follow on Twitter: @ProfessorDrz
wreck
Banned User
Posts: 512
And1: 0
Joined: Oct 29, 2008

Re: Is pau statistically more important than kobe? 

Post#13 » by wreck » Sun Nov 8, 2009 4:14 pm

NetsForce wrote:All I know is that the Lakers wouldn't be 5-1 if Pau was playing and Kobe wasn't.


You're right. They'd most likely be undefeated. Maybe you should look up the Lakers' record since Gasol has joined the team. Then again, most people have (inaccurately) attributed the improvement to Bryant.
wreck
Banned User
Posts: 512
And1: 0
Joined: Oct 29, 2008

Re: Is pau statistically more important than kobe? 

Post#14 » by wreck » Sun Nov 8, 2009 4:18 pm

Wile E. Coyote wrote:
So the Lakers would be better if Kobe were injured right now and Gasol wasn't?


Most likely. The triangle offense is ran more efficiently through the post. All you have to do is look at the efficiency stats + the W/L records and history backs me up. Throughout the years, the Lakers have always suffered more when their #1 post option was out of the lineup, Shaq O'Neal, Lamar Odom, Andrew Bynum, and now Pau Gasol compared to when Bryant was out.

I know Bryant fans don't like to hear that, because the statistics show Bryant is the most overrated player the NBA has seen in a long time.
Doctor MJ
Senior Mod
Senior Mod
Posts: 41,143
And1: 9,150
Joined: Mar 10, 2005
Location: Treys are for 'Kicks!...and Steph Curry
   

Re: Is pau statistically more important than kobe? 

Post#15 » by Doctor MJ » Sun Nov 8, 2009 7:03 pm

TrueLAfan wrote:Why is why I agree with your statement about use of statistical analysis “in conjunction with your own observations and knowledge.”  But that statement should be applied to and by designers of statistical models as well.  People who have developed a system that tells them that Jason Kidd was worth 260% more wins to the 2009 Dallas Mavericks than Dirk Nowitzki should not be asking, “What does that tell you about the players?”  They should be looking at the system and asking, “That can’t be right.  We need to fix this.”
 
But that doesn’t happen often—or, at least, often enough.  People become wedded to their modelling and analytic tools. 


This. I don't get frustrated by a statistician creating a tool that's imperfect (they're all imperfect). The same approach that we use which factors in various stats and non-stats is what any statistician should be using, and if they don't, they're going to hurt the statistical community when their credibility is shot.
In another world...
Your Longshoremen of Long Beach...

And now introducing...
The Song of Westeros...
LeBron James
John Stockton
Bobby Jones
Rudy Gobert
Khris Middleton
Connie Hawkins
Otto Porter
Ryan Anderson
User avatar
rrravenred
Retired Mod
Retired Mod
Posts: 5,903
And1: 418
Joined: Feb 24, 2006
Location: Pulling at the loose threads of arguments since 2006

Re: Is pau statistically more important than kobe? 

Post#16 » by rrravenred » Sun Nov 8, 2009 11:24 pm

True, in your opinion what's wrong with the system apart from the eyeball test? What do you believe is the evidence which either challenges or invalidates WP as a stat?

I suppose the corollary to that is how wedded you are to your interpretation of basketball and your respective weighting of particular players. Are you looking at your perceptions with the same critical eye you are at WP?

(NB, I'm not necessarily defending WP as a stat, just querying the objective basis for your criticism)
ElGee wrote:You, my friend, have shoved those words into my mouth, which is OK because I'm hungry.


Got fallacy?
Doctor MJ
Senior Mod
Senior Mod
Posts: 41,143
And1: 9,150
Joined: Mar 10, 2005
Location: Treys are for 'Kicks!...and Steph Curry
   

Re: Is pau statistically more important than kobe? 

Post#17 » by Doctor MJ » Sun Nov 8, 2009 11:33 pm

rrravenred wrote:True, in your opinion what's wrong with the system apart from the eyeball test? What do you believe is the evidence which either challenges or invalidates WP as a stat?

I suppose the corollary to that is how wedded you are to your interpretation of basketball and your respective weighting of particular players. Are you looking at your perceptions with the same critical eye you are at WP?

(NB, I'm not necessarily defending WP as a stat, just querying the objective basis for your criticism)


Honestly it's been quite a while since I've analyzed Berri. There are a whole slew of guys coming up with their own stats, and when Berri came out with his metric I looked at it, and concluded it was significantly weaker than most of the other metrics from guys you've never heard of. That Berri's become a big enough name that people are specifically asking about him here is amusing to me.

If you've taken a look at his stuff and concluded it makes some sense to you, I'm fine with that as long as you use other factors as well. Basically his stuff is in the same phylum as some other metrics, and while it may or may not be the best, all of them are decent enough to be useful.
In another world...
Your Longshoremen of Long Beach...

And now introducing...
The Song of Westeros...
LeBron James
John Stockton
Bobby Jones
Rudy Gobert
Khris Middleton
Connie Hawkins
Otto Porter
Ryan Anderson
NetsForce
Banned User
Posts: 20,711
And1: 24
Joined: Dec 27, 2006

Re: Is pau statistically more important than kobe? 

Post#18 » by NetsForce » Sun Nov 8, 2009 11:39 pm

wreck wrote:You're right. They'd most likely be undefeated. Maybe you should look up the Lakers' record since Gasol has joined the team. Then again, most people have (inaccurately) attributed the improvement to Bryant.


Of course the Lakers are going to be between when Bryant AND Gasol play... Though if you could only have one of the two on the current Lakers team you'd be a fool to take Pau over Kobe... I mean newsflash pre Kobe Bryant Pau Gasol racked up a total of 0 playoff wins as his team's #1 option, that's not exactly the stuff of legends.
User avatar
rrravenred
Retired Mod
Retired Mod
Posts: 5,903
And1: 418
Joined: Feb 24, 2006
Location: Pulling at the loose threads of arguments since 2006

Re: Is pau statistically more important than kobe? 

Post#19 » by rrravenred » Sun Nov 8, 2009 11:41 pm

Oh sure... a lot of the metrics agree with each other (in a broad categorisation of player quality) sense. The interesting bit to me on Berri is the valuation of raw scoring numbers in an economic sense and how well this meshes with the effect on team wins.

Since MJ, it's a perimeter player's league (in the sense of marketing), and I just find it interesting how MJ-type play is remunerated beyond apparent value to the team. The main critiques I've seen of Berri is how he deifies shooting efficiency and defensive rebounding. Hollinger, OTOH, deifies shooting, period. ;)

What other metrics would you recommend as providing a different slant on player performance?
ElGee wrote:You, my friend, have shoved those words into my mouth, which is OK because I'm hungry.


Got fallacy?
RockTHECasbah
Lead Assistant
Posts: 4,526
And1: 6
Joined: Oct 05, 2007

Re: Is pau statistically more important than kobe? 

Post#20 » by RockTHECasbah » Mon Nov 9, 2009 4:22 am

What makes Pau so efficient in that system? Part of it is Kobe breaking down the defense, even if it doesn't end up in a basket for Pau.
Know anyone who is disabled? has an addiction?
Image
HandyTax - Your Canadian Disability Tax Credit Consultants
http://www.handytax.ca

Return to Statistical Analysis