PDA

View Full Version : Can somebody please explain "WIN SHARES" to me?



elledaddy
08-09-2013, 12:53 PM
I'll admit that while I don't care for advance stats, I won't ignore them either. I read a post where some one was talking about Beno Udrih's "win shares" and it peeked my enterest. So here are my two request/questions....

1. Explain win shares to me( not just post the definition of it). I want to know if ppl that use this to back their stance actually KNOW what it means or are they just looking on the web for " Top win shares in NBA"


2. This one may answer itself in the process but--------
Wouldnt players on better teams automatically have higher win shares than players on poorer teams?

Example( help me if Im wrong)....Wouldn't Lebron James ( current stats) have less win shares if he was on a 20 win Orl Magic team than ( lets say) Luol Deng( current stats) on a 60 win Chi Bulls team?<-----------------hypothetical

2-ONE-5
08-09-2013, 12:58 PM
i hate advanced stats espeically in bball. Too many people think they know it all by just looking up advanced stats and never seeing a player play or understand the situation that player is in.

mike_noodles
08-09-2013, 01:08 PM
Nobody can explain it. It's a made up stat where you can always insert your favorite player in the top ten.

4milesperday
08-09-2013, 01:13 PM
It's the dumbest statistics ever. It's pretty much your contribution to a win...but largely dependent on your teammates.

ILLUSIONIST^248
08-09-2013, 01:16 PM
If you question any advance stats the great and powerful FC shall strike you down.

1-800-STFU
08-09-2013, 01:22 PM
i hate advanced stats espeically in bball. Too many people think they know it all by just looking up advanced stats and never seeing a player play or understand the situation that player is in.

Nobody can explain it. It's a made up stat where you can always insert your favorite player in the top ten.

It's the dumbest statistics ever. It's pretty much your contribution to a win...but largely dependent on your teammates.

If you question any advance stats the great and powerful FC shall strike you down.

Why are people so hostile towards new ideas?

jayjay33
08-09-2013, 01:23 PM
Worst thing ever happened to basketball.

Chronz
08-09-2013, 01:27 PM
Its important to distinguish between OWS (Offensive WS) vs DWS for the very reason you mentioned. DWS has a strong team component that may (improperly) assign defensive value to a players Win Share. But as far as OWS is concerned, 2 players with identical stats on entirely different teams will accrue the same offensive value.

So yes, if a player is on a great defensive team, he will get credit for it, right or wrong. So if you see a guy with a higher/lower rate than you would expect, ask yourself if you think hes gotten enough praise/blame for his teams defense.

YoungOne
08-09-2013, 01:27 PM
i hate advanced stats espeically in bball. Too many people think they know it all by just looking up advanced stats and never seeing a player play or understand the situation that player is in.

this.

the best way to analyze a player is still the good old scouting..

Chronz
08-09-2013, 01:29 PM
this.

the best way to analyze a player is still the good old scouting..

What if we dont trust what you think your eyes are watching?

D-Leethal
08-09-2013, 01:29 PM
Just another flawed stat that dudes use to judge individuals in a team sport where your success is directly tied to your teammates ability and actions.

D-Leethal
08-09-2013, 01:31 PM
Its important to distinguish between OWS (Offensive WS) vs DWS for the very reason you mentioned. DWS has a strong team component that may (improperly) assign defensive value to a players Win Share. But as far as OWS is concerned, 2 players with identical stats on entirely different teams will accrue the same offensive value.

So yes, if a player is on a great defensive team, he will get credit for it, right or wrong. So if you see a guy with a higher/lower rate than you would expect, ask yourself if you think hes gotten enough praise/blame for his teams defense.

Can you elaborate on how the bold is possible?

D-Leethal
08-09-2013, 01:32 PM
Are you saying the only part of basketball your teammates directly impact your individual success and production is on defense?

NYKnickFanatic
08-09-2013, 01:34 PM
i hate advanced stats espeically in bball. Too many people think they know it all by just looking up advanced stats and never seeing a player play or understand the situation that player is in.

Agreed.

NYCkid12
08-09-2013, 01:34 PM
to take it a step further, can someone explain PER....I Know the definition but how does it get determined? what formula is used?

YoungOne
08-09-2013, 01:34 PM
What if we dont trust what you think your eyes are watching?

you dont have to trust my opinion, but I'd rather take the opinion of an expirienced nba scout than believe in some "advanced" stats that never tell the entire story

D-Leethal
08-09-2013, 01:35 PM
What if we dont trust what you think your eyes are watching?

Than you disagree. There is no defined "proof" of anything with these stats anyway. You can poke holes in every single of these stats and were supposed to trust them instead?

KG2TB
08-09-2013, 01:41 PM
Advanced stats have a place but like anything it's about context. There's no one holy grail you can look to. Like any good researcher, you have to find and compare numerous sources and the same holds true with stats. You have to look first and foremost with your eyes and a good basketball mind. There's intangibles that play a role in regards to value. For me, beyond the eye test, I generally look at TS%, eFG%, PER, 3 PT%, FT% and feel good about evaluating a player. However, the eye test is the most important because if you never really watch a player, stats could be misleading or not tell the whole story. You need all of those and some people will say you eve need more.

Chronz
08-09-2013, 01:42 PM
Can you elaborate on how the bold is possible?
You want me to look up the formula? Its because the team component in OWS is incredibly small, to the point where its a complete non factor.

Chronz
08-09-2013, 01:47 PM
to take it a step further, can someone explain PER....I Know the definition but how does it get determined? what formula is used?
Old fashioned linear weights. Outdated but the most popular it seems. WARP, WS, AWS and RAPM are all more insightful.

tp13baby
08-09-2013, 01:48 PM
TS% and eFG% as well as PER are probably what judge off of if I use stats. Rebounds and points are the biggest joke of stats in my opinion.

D-Leethal
08-09-2013, 01:48 PM
You want me to look up the formula? Its because the team component in OWS is incredibly small, to the point where its a complete non factor.

Every players production (or points produced in this case) is a product of their team component. The team component is everywhere in basketball - its a freakin' team sport, and not a team sport of individual matchups like baseball. All 5 guys feed off each other every second they are out there. You can't isolate players in a vaccum like you want to and I know it hurts you stat heads on the inside because you know its true.

Chronz
08-09-2013, 01:51 PM
you dont have to trust my opinion, but I'd rather take the opinion of an expirienced nba scout than believe in some "advanced" stats that never tell the entire story

yea but you said "when it comes down to analyzing a player", I took that to mean you were actually the one breaking things down.

When it comes down to believing what other people have to say, you can believe whoever you want, we all have that right, but say you were forced into thinking for yourself and explaining why you hold your belief.

The point Im making is that there is no BEST way, we all have our way, we all try to mix both subjective/objective elements of the game, the only thing I know is the WORST way to evaluate the game is to ignore either end entirely.

Chronz
08-09-2013, 01:57 PM
Every players production (or points produced in this case) is a product of their team component.

You're not getting it. Heres what U bolded:

But as far as OWS is concerned, 2 players with identical stats on entirely different teams will accrue the same offensive value.

The underlined is what you were asking about. Spare me your diatribe about stats bro, Im not interested in that crusade. Im trying to stick to pointing out specific individuals nowadays, not liking your generalization of stat savvy posters. If you're going to quote me, make sure you stay on topic.


If you wish to ask me a DIFFERENT question about the state of stats, feel free. That would be a refreshing change from you just lumping people into your predetermined category. People change, its funny, if you look up a thread on this very same subject from 2 years ago, you would see just how much. I was more of a dick then.

KnickaBocka.44
08-09-2013, 02:09 PM
What if we dont trust what you think your eyes are watching?

Then you can use stats to back your claim. But using stats as a basis for your claim is wrong.

D-Leethal
08-09-2013, 02:17 PM
You're not getting it. Heres what U bolded:

But as far as OWS is concerned, 2 players with identical stats on entirely different teams will accrue the same offensive value.

The underlined is what you were asking about. Spare me your diatribe about stats bro, Im not interested in that crusade. Im trying to stick to pointing out specific individuals nowadays, not liking your generalization of stat savvy posters. If you're going to quote me, make sure you stay on topic.


If you wish to ask me a DIFFERENT question about the state of stats, feel free. That would be a refreshing change from you just lumping people into your predetermined category. People change, its funny, if you look up a thread on this very same subject from 2 years ago, you would see just how much. I was more of a dick then.

Sorry, I don't equate OWS to Offensive Value. I took that as you saying two players with identical stats on different teams have the same offensive value, not have the same OWS. You can provide more value to an offense than points, direct assists and offensive rebounds.

And I was here, I know how much of a dick you were back then, on your high horse fighting for these almighty stats. I am just happy people can see past the high horse and see the glaring flaws in the metrics you have been preaching for so long. Come to think of it, I actually blame you for the Guppy's of the world, even if you actually know how to use these stats properly. Those dudes are like your disciples, took your high horse as gospel and it spread like the plague around here.

Chronz
08-09-2013, 02:18 PM
Then you can use stats to back your claim. But using stats as a basis for your claim is wrong.
I get that.

What I dont get is the urge for people to bring this up in a thread revolving around a specific STAT. lol...

its someone trying to understand a stat. It should go without saying that we all know the limitations inherent in ALL stats. That doesn't prevent the league and its managers from tracking stats. Did anyone aside from me even attempt to answer the Q?

Im the only one on topic here.

Chronz
08-09-2013, 02:23 PM
Sorry, I don't equate OWS to Offensive Value.
Offensive value is defined by whatever metric is being discussed. Thats all I was getting at.


And I was here, I know how much of a dick you were back then, on your high horse fighting for these almighty stats. I am just happy people can see past the high horse and see the glaring flaws in the metrics you have been preaching for so long. Come to think of it, I actually blame you for the Guppy's of the world, even if you actually know how to use these stats properly. Those dudes are like your disciples, took your high horse as gospel and it spread like the plague around here.
lol you cant be serious, im going to write most of this in non serious (aka no caps) internet speech.

and feel free to bring up these glaring flaws whenever pertinent to a debate, using a terms like flawed to describe stats isn't something people didn't acknowledge before entering the field as an interest/hobby/profession whatever the case may be. again tho, not taking you seriously right now. dont ever blame me for guppy.

D-Leethal
08-09-2013, 02:33 PM
I'm just busting balls dude. I will say the biggest flaw is people simply not knowing how to properly supplement an opinion with them, provide enough proper contextual support to them, or acknowledge the flaws when you don't factor in the team aspect that directly affects all these numbers. Too many dudes pull them out their *** to form an opinion and it effectively rendered this forum ****ed for awhile. Seems like people are starting to come around to their senses again. Lets see if it sticks now that Guppy is no longer banned.

I do find it funny that there probably is not one guy here who can properly break down the stat and provide anything more than a broad description.

How exactly does this stat determine how many wins you produce? Nobody needs to bust out the formula but at least provide detail. To me, it sounds stupid as hell on the surface and the more I read into it the I confirm that initial thought.

Hawkeye15
08-09-2013, 02:43 PM
http://www.basketball-reference.com/about/ws.html

D-Leethal
08-09-2013, 02:44 PM
And for all the guys who love to boast "Why you think all these teams have analytics departments?"

I'd be willing to bet they use them to evaluate within the context of their own team, and they understand the flaws in crossing them over to other teams. It makes a whole world of sense to hire analytics departments to find out your best lineup combos, your hot spots for each player, who thrives next to who, who gives the team a boost, who struggles next to who, what type of pace you should play etc...

They sure as hell don't use them to rank/compare/rate players from different teams thats for sure. These stats have a long way to go until you can do that without a million grains of salt.

elledaddy
08-09-2013, 02:48 PM
Offensive value is defined by whatever metric is being discussed. Thats all I was getting at.


lol you cant be serious, im going to write most of this in non serious (aka no caps) internet speech.

and feel free to bring up these glaring flaws whenever pertinent to a debate, using a terms like flawed to describe stats isn't something people didn't acknowledge before entering the field as an interest/hobby/profession whatever the case may be. again tho, not taking you seriously right now. dont ever blame me for guppy.



I'll ask you because you seem to follow metrics closely. You said that players with similar stats on different teams would basically result in the same or similar WS or OWS. But when I look at OWS top 20 players last year, 18 made the playoffs. Also 19 of the top 20 on DWS made the playoffs. And WS( OWS + DWS) of the TOP 20, all 20 made the playoffs. The same goes for WS per 48. Actually all of the top 20 came from the top 12 NBA teams.I cant see how team wins doesnt factor heavily in this science period, whether it is OWS, DWS or WS. Not one player in the entire NBA, whose team didnt make the playoffs was able to produce more( or have more ) WS than Tyson Chandler? That just seems wrong to me..

2-ONE-5
08-09-2013, 02:51 PM
I'm just busting balls dude. I will say the biggest flaw is people simply not knowing how to properly supplement an opinion with them, provide enough proper contextual support to them, or acknowledge the flaws when you don't factor in the team aspect that directly affects all these numbers. Too many dudes pull them out their *** to form an opinion and it effectively rendered this forum ****ed for awhile. Seems like people are starting to come around to their senses again. Lets see if it sticks now that Guppy is no longer banned.
I do find it funny that there probably is not one guy here who can properly break down the stat and provide anything more than a broad description.

How exactly does this stat determine how many wins you produce? Nobody needs to bust out the formula but at least provide detail. To me, it sounds stupid as hell on the surface and the more I read into it the I confirm that initial thought.

this dude is the worst with advanced stats.

ILLUSIONIST^248
08-09-2013, 02:56 PM
Its important to distinguish between OWS (Offensive WS) vs DWS for the very reason you mentioned. DWS has a strong team component that may (improperly) assign defensive value to a players Win Share. But as far as OWS is concerned, 2 players with identical stats on entirely different teams will accrue the same offensive value.

So yes, if a player is on a great defensive team, he will get credit for it, right or wrong. So if you see a guy with a higher/lower rate than you would expect, ask yourself if you think hes gotten enough praise/blame for his teams defense.

The mighty FC has graced us with his presence.

John Walls Era
08-09-2013, 03:12 PM
Advance stats are actually pretty useful. Though I don't agree when the so called nerds (so called because they aren't nerd --- that would imply they are as smart as they think when they just regurgitate what others made up) think its the be all end all.

Advance stats are like another side of a 'cube'. Its useful to get a slight hand up against vegas oddsmaker (very slight if used correctly to notice trends).

Chronz
08-09-2013, 03:14 PM
I'm just busting balls dude. I will say the biggest flaw is people simply not knowing how to properly supplement an opinion with them, provide enough proper contextual support to them, or acknowledge the flaws when you don't factor in the team aspect that directly affects all these numbers.
Opinions can vary on how much it effects them, but you still need a starting point for comparison. Even then the subjective element of your argument is heightened by prior evidence.



I do find it funny that there probably is not one guy here who can properly break down the stat and provide anything more than a broad description.
How exactly does this stat determine how many wins you produce? Nobody needs to bust out the formula but at least provide detail. To me, it sounds stupid as hell on the surface and the more I read into it the I confirm that initial thought.
Well, that depends on what you already know. Are you familiar with the pythagorean method and how it pertains to basketball? Its not that stupid to me but it depends on how you look at the game.

John Walls Era
08-09-2013, 03:16 PM
And for all the guys who love to boast "Why you think all these teams have analytics departments?"

I'd be willing to bet they use them to evaluate within the context of their own team, and they understand the flaws in crossing them over to other teams. It makes a whole world of sense to hire analytics departments to find out your best lineup combos, your hot spots for each player, who thrives next to who, who gives the team a boost, who struggles next to who, what type of pace you should play etc...

They sure as hell don't use them to rank/compare/rate players from different teams thats for sure. These stats have a long way to go until you can do that without a million grains of salt.

Bingo.

This is about as plausible a reason as there should be. Its asinine to not used all the tools available to get a leg up or at least improve the team.

JeremiahWing
08-09-2013, 03:18 PM
Advanced stats in basketball are just silly. You can't isolate the action enough to quantify everything, like you can and should in baseball.

Chronz
08-09-2013, 03:30 PM
I'll ask you because you seem to follow metrics closely. You said that players with similar stats on different teams would basically result in the same or similar WS or OWS. But when I look at OWS top 20 players last year, 18 made the playoffs. Also 19 of the top 20 on DWS made the playoffs. And WS( OWS + DWS) of the TOP 20, all 20 made the playoffs. The same goes for WS per 48. Actually all of the top 20 came from the top 12 NBA teams.I cant see how team wins doesnt factor heavily in this science period, whether it is OWS, DWS or WS. Not one player in the entire NBA, whose team didnt make the playoffs was able to produce more( or have more ) WS than Tyson Chandler? That just seems wrong to me..
Well the best players due tend to make the playoffs but yes, its an efficiency based stat and Tyson is statistically the most efficient player on a Per Possession basis. I dont know if thats your cup of tea in terms of an all-in-1 stat but I can assure you its not because of a team component in the stat itself.

Just compare the stats of 18 and 19 in OWS and their respective teams.

Ryan Anderson (18) vs Bosh (19). Bosh was on a contender, Anderson was on the lowly Pelicans. Why similar OWS?

Because both scored roughly 16PPG, they did so on the exact same efficiency scale (114 O-RTG) in roughly the same amount of minutes, slight advantage to Ryan in usage% and 50+minutes gave him the edge.

What makes Bosh WS superior is the +3.4 boost he gets from Miami's top-8 defense vs (+.8DWS) from a 28th defense for Anderson.

IndyRealist
08-09-2013, 03:37 PM
And for all the guys who love to boast "Why you think all these teams have analytics departments?"

I'd be willing to bet they use them to evaluate within the context of their own team, and they understand the flaws in crossing them over to other teams. It makes a whole world of sense to hire analytics departments to find out your best lineup combos, your hot spots for each player, who thrives next to who, who gives the team a boost, who struggles next to who, what type of pace you should play etc...

They sure as hell don't use them to rank/compare/rate players from different teams thats for sure. These stats have a long way to go until you can do that without a million grains of salt.

Well, what did you expect? No one that you should take seriously says stats are the end all, be all of basketball analysis. Yet there are so many people, many in this thread, who say "stats are worthless, watch a game" which directly contradicts what you just said. Stats are to be taken in context, but when one suggests that someone's favorite player isn't as good as they thought he was, suddenly all stats are evil, blah blah blah.

No one likes being told they are wrong. But some perfectly reasonable people say that Kobe was the best player of the last decade. Other perfectly reasonable people say Tim Duncan. Others say Lebron. They can't all be right. Yet these people all watched games, all reasonably assessed what they saw, and all came to completely different conclusions.

Advanced metrics exist for two reasons:

One, you can't watch every game, it's physically impossible. There are 1230 games each season not including playoffs. Most people don't even watch 10% of that. So how can you say that the few games you watched weren't' an aberration, and that the player in question isn't significantly better or worse than you believe? And out of the games you watched last year, what was the 5th offensive play in the 5th game? Exactly. Human memory is flawed, you glean generalities about what you've seen and apply them as a rule. That's why there's a guy in another thread saying Brandon Jennings is an all-star caliber player, probably because the only thing he knows about Jennings is one game his rookie year, and that he was drafted in the lottery. If he bothered to look at his FG%, he'd probably say, "Who the hell drafted this guy in the lottery?" or at the very least "Who the hell gave this guy $25M?"

Two, people give weight to different observations, and they generally get it wrong. Know what the two main driving factors for a player's first non-rookie contract are? Points per game and draft position. Rookie of the year? Points per game and draft position. MVP? Points per game and team wins. 6th man of the year? Points per game. DOES ANYONE HERE THINK POINTS PER GAME IS THE BEST WAY TO JUDGE A PLAYER? Hell no. But that's exactly what we use to judge players, and that's what advanced metrics contradict. When people say that they don't look at stats they watch the game, they are flat out lying. They're simply looking at grossly misleading stats, and then incorporating that into their view of the player.

No one should ever, ever say to ignore what you see and just look at stats. Just like no one should ever say to ignore stats and just watch the game. Both are foolish.

FYL_McVeezy
08-09-2013, 03:38 PM
Advanced stats have a place but like anything it's about context. There's no one holy grail you can look to. Like any good researcher, you have to find and compare numerous sources and the same holds true with stats. You have to look first and foremost with your eyes and a good basketball mind. There's intangibles that play a role in regards to value. For me, beyond the eye test, I generally look at TS%, eFG%, PER, 3 PT%, FT% and feel good about evaluating a player. However, the eye test is the most important because if you never really watch a player, stats could be misleading or not tell the whole story. You need all of those and some people will say you eve need more.

:clap:
/thread

Tony_Starks
08-09-2013, 04:04 PM
My issue with advanced stats are the people that use them to hype their favorite player or discredit a player they just flat out dont like. Takes the credibility out of it.

bearadonisdna
08-09-2013, 04:10 PM
It seems like advanced stats in basketball took off when Lebron entered the league because there had to be some stat that said he was better than Jordan.

KnickaBocka.44
08-09-2013, 04:15 PM
I get that.

What I dont get is the urge for people to bring this up in a thread revolving around a specific STAT. lol...

its someone trying to understand a stat. It should go without saying that we all know the limitations inherent in ALL stats. That doesn't prevent the league and its managers from tracking stats. Did anyone aside from me even attempt to answer the Q?

Im the only one on topic here.

In fairness, you did ask the question, "What if we don't trust what you think your eyes are seeing?", or something to that affect. To which I, and others, responded with varying responses. So you kind of helped derail it in a way. Not giving you ****, just sayin...


Offensive value is defined by whatever metric is being discussed. Thats all I was getting at.


lol you cant be serious, im going to write most of this in non serious (aka no caps) internet speech.

and feel free to bring up these glaring flaws whenever pertinent to a debate, using a terms like flawed to describe stats isn't something people didn't acknowledge before entering the field as an interest/hobby/profession whatever the case may be. again tho, not taking you seriously right now. dont ever blame me for guppy.

I had an encounter with him recently and I'm glad to see you don't subscribe to the same philosophy as he does :laugh2: .

In all seriousness though, he is the kind of poster that shoves stats down your throat and makes people dislike advanced metrics. He uses them as gospel and tells you that if you disagree then you're wrong, because "that is what actually happened".

macc
08-09-2013, 04:44 PM
I'm not big on advanced stats. I much prefer the eye test myself. In saying that, I really don't fully understand most advanced stats so it's tough for me to have a full opinion on them. In some circumstances I think people can tend to over analyze some things.

Though I'm not ignorant to advanced stats. I'm not going to bash them simply because I don't understand them, which is what I think alot of people do. I just personally think the eye test is still the best way to judge any particular player. There are just some things that impact a game that a stat doesn't cover. For instance...

When VC was in Orlando that year. The year prior he was averaging over 20 ppg. He was obviously not a prime VC but at the same time he was def capable of making plays (as he did). Now picture a scenario, 20 sec left, 4th quarter, tie game, Magic with the ball. The other team knows that most likely Vince Carter was going to be the one taking the shot. So they would adjust their defense for this. Reguardless of the fact VC was having one of the worst years of his career, it's not like the other team was thinking "well his PER is down this year so let's not worry about him on this last play." So tell me where the stat is that shows a players impact on a court whereas the other team has to put a defense in place that stops mainly just one guy.

The point I'm making is reguardless if VC makes that final shot or not, he impacts the game in a different way that doesn't show up on a stat sheet.

bagwell368
08-09-2013, 05:22 PM
i hate advanced stats espeically in bball. Too many people think they know it all by just looking up advanced stats and never seeing a player play or understand the situation that player is in.

Hmmmm...

What about if the observer is a ESPN bred fan boy and doesn't know much about the game? The "evaluations" they make from that are likely to be less reliable than stats. I'd estimate generously 20% of the posters here do not suffer from some to a great extent from this disease.

What if it's a West Coast team that's seldom seen by an East Coast fan (2 team games, and maybe 1-2 other games on TNT)? Still think you have a grasp of a player you might see play 40 minutes a year?

What about historic comparisons. I'm the only person on PSD that's a regular that saw Bill Russell in/near his prime (the other 4 that claimed to be didn't have near enough associated knowledge/memories to prove their claims). You sure you want to go with written accounts and a few clips to talk about Russell in an authoritative manner?

Win Shares by the way (I'm sure someone pointed it out) is biased by being on a team that wins a lot more or a lot less than average. Better to take a look at the Win Shares a player has a percentage of the team total vs some other player in such cases. Not to mention ORtg and DRtg.

Clearly the game is more enjoyable and understandable the more ways you can look at it. I can look at it as a fan, former player, amateur historian, stat guy, and coach. It might not be the best background here, but, it's pretty good.

elledaddy
08-09-2013, 05:44 PM
It seems like advanced stats in basketball took off when Lebron entered the league because there had to be some stat that said he was better than Jordan.


Interesting way to look at it.

IKnowHoops
08-09-2013, 05:47 PM
Every players production (or points produced in this case) is a product of their team component. The team component is everywhere in basketball - its a freakin' team sport, and not a team sport of individual matchups like baseball. All 5 guys feed off each other every second they are out there. You can't isolate players in a vaccum like you want to and I know it hurts you stat heads on the inside because you know its true.

As much as I agree with you, you do realize that using this rational(the right rational I may add) makes it impossible to say Jordan was better than Lebron because of his Rings. A single player has better control of his OWS than he does over whether he wins a ring. Thats just the most obvious one. But you have also erased all of Jordan's scoring titles as well because points don't matter either because of how envolved team is. When "5 guys are feeding off each other every second they are out there" then along with OWS, scoring, rebounding, steals, asst, are all so heavily reliant on the other four guys out on the floor with you, reacting, to your every faint, every solitary second, hoping in and out of passing lanes, boxing out your man, reading your eyes, tipping you the rebound, help defense on the strong side, reading your cut to the basket, screening your man, etc, etc. Infinite possibility because of the players around you make it impossible for any stat to tell people which player is more efficient and or better at performing said stat. Is this what you are implying?

IKnowHoops
08-09-2013, 05:55 PM
Nothing is 100% for sure, not even the eye test. But everyone agrees that Lebron and Durant are 1-2 in the NBA. Its no coincidence that Lebron and Durant are 1-2 in PER, OWS, and any other stat that tries to define the totality of a players worth. The advanced stats are pretty on. There are often anomalies like Brook Lopez being top 5-6 in PER but for the most part there will be 1-2 anomalies in every 20 names and if you know anything about basketball, they will be very easy to spot. Otherwise its pretty spot on when comparing players that are playing similar minutes.

kozelkid
08-09-2013, 05:58 PM
i hate advanced stats espeically in bball. Too mancombine think they know it all by just looking up advanced stats and never seeing a player play or understand the situation that player is in.

this.

the best way to analyze a player is still the good old scouting..

It's a shame that people continue to believe that it can only be one option or the other instead acknowledging the fact that both schools of thought can and should be used together in analyzing basketball or any other sport for that matter.

IKnowHoops
08-09-2013, 06:20 PM
Sorry, I don't equate OWS to Offensive Value. I took that as you saying two players with identical stats on different teams have the same offensive value, not have the same OWS. You can provide more value to an offense than points, direct assists and offensive rebounds.

And I was here, I know how much of a dick you were back then, on your high horse fighting for these almighty stats. I am just happy people can see past the high horse and see the glaring flaws in the metrics you have been preaching for so long. Come to think of it, I actually blame you for the Guppy's of the world, even if you actually know how to use these stats properly. Those dudes are like your disciples, took your high horse as gospel and it spread like the plague around here.

Common PSD is not that deep.

IKnowHoops
08-09-2013, 06:35 PM
I'm just busting balls dude. I will say the biggest flaw is people simply not knowing how to properly supplement an opinion with them, provide enough proper contextual support to them, or acknowledge the flaws when you don't factor in the team aspect that directly affects all these numbers. Too many dudes pull them out their *** to form an opinion and it effectively rendered this forum ****ed for awhile. Seems like people are starting to come around to their senses again. Lets see if it sticks now that Guppy is no longer banned.

I do find it funny that there probably is not one guy here who can properly break down the stat and provide anything more than a broad description.

How exactly does this stat determine how many wins you produce? Nobody needs to bust out the formula but at least provide detail. To me, it sounds stupid as hell on the surface and the more I read into it the I confirm that initial thought.

Dude, its not like its calculus. Its easy to understand but it lengthy and long. It has to be to be accurate. It has to take everything into account.

Check it. Its straight forward on how this stat works.

http://www.basketball-reference.com/about/ws.html
Calculating Win Shares

I. Introduction

Stealing a page from baseball's Bill James, I decided to attempt to calculate basketball Win Shares. This article will describe how I came up with the Win Shares system for basketball. If you believe that any attempt to attribute team success to individual players is an abomination, then read no further, as this article will be of no interest to you.

II. What is a Win Share?

Bill James developed his system such that one win is equivalent to three Win Shares. My system deviates from James's system in three key ways:

In James's system, one win is equivalent to three Win Shares. In my system, one win is equivalent to one Win Share.
James made team Win Shares directly proportional to team wins. In his system, a baseball team that wins 80 games will have exactly 240 Win Shares, a baseball team that wins 90 games will have exactly 270 Win Shares, etc. In my system, a basketball team that wins 50 games will have about 50 Win Shares, give or take.
James did not allow for the possibility of negative Win Shares. In his system, the fewest number of Win Shares a player can have is zero. In my system, a player can have negative Win Shares. I justify this by thinking about it in the following way: a player with negative Win Shares was so poor that he essentially took away wins that his teammates had generated.
III. Crediting Offensive Win Shares to Players

A. 1977-78 to present NBA

Offensive Win Shares are credited to players based on Dean Oliver's points produced and offensive possessions. The formulas are quite detailed, so I would point you to Oliver's book Basketball on Paper for complete details. The process for crediting Offensive Win Shares is outlined below (using LeBron James of the 2008-09 Cleveland Cavaliers as an example):

Calculate points produced for each player. In 2008-09, James had an estimated 2345.9 points produced.
Calculate offensive possessions for each player. James had an estimated 1928.1 offensive possessions in 2008-09.
Calculate marginal offense for each player. Marginal offense is equal to (points produced) - 0.92 * (league points per possession) * (offensive possessions). For James this is 2345.9 - 0.92 * 1.083 * 1928.1 = 424.8. Note that this formula may produce a negative result for some players.
Calculate marginal points per win. Marginal points per win reduces to 0.32 * (league points per game) * ((team pace) / (league pace)). For the 2008-09 Cavaliers this is 0.32 * 100.0 * (88.7 / 91.7) = 30.95.
Credit Offensive Win Shares to the players. Offensive Win Shares are credited using the following formula: (marginal offense) / (marginal points per win). James gets credit for 424.8 / 30.95 = 13.73 Offensive Win Shares.
B. 1973-74 to 1976-77 NBA

The NBA did not track player turnovers until the 1977-78 season, and player turnovers are needed to calculate player possessions. However, the NBA did track turnovers at the team level from 1973-74 to 1976-77. Since player turnovers are the only thing holding us back from using the method outlined above, I have chosen to estimate player turnovers for this time period. Player turnovers are estimated as follows (using Kareem Abdul-Jabbar of the 1976-77 Los Angeles Lakers as an example):

Obtain an initial estimate of the player's turnovers. To do this use the following formula:
-0.0005075172 * (minutes played) * (player age)
- 0.0873982755 * (field goals)
+ 0.0925506598 * (field goal attempts)
+ 0.1566322510 * (free throw attempts)
+ 0.0449241773 * (total rebounds)
+ 0.2321637159 * (assists)
+ 0.2040169400 * (personal fouls)
Note that if this number is less than zero, then it should be rounded up to zero. Plugging Abdul-Jabbar's statistics into the formula above we get an estimate of 280.316 turnovers.
Find the sum of estimated turnovers for the players on the given team. The sum for the players on the 1976-77 Lakers is 1448.057.
Calculate the player's share of this total. Abdul-Jabbar's share of the team total is 280.316 / 1448.057 = 0.194.
Multiply the team's turnovers (adjusted for team turnovers) by the player's share. As mentioned, the NBA tracked turnovers at the team level in these seasons. However, the team totals include team turnovers (i.e., turnovers that are not attributed to an individual player). Thus, we multiply the team's turnovers by 0.985, then multiply this adjusted figure by the player's share. For Abdul-Jabbar this is 1538 * 0.985 * 0.194 = 293.9, which we round up to 294.
Now that we have this estimate, the method above is used to complete the calculation of Offensive Win Shares.

C. 1946-47 to 1948-49 BAA and 1949-50 to 1972-73 NBA

Because so many statistics are missing prior to the 1973-74 season (offensive rebounds, turnovers, etc.), we will not use Oliver's points produced and offensive possessions for this time period, although the basic framework will remain the same. Here is the process for crediting Offensive Win Shares prior to the 1973-74 season (using Oscar Robertson of the 1964-65 Cincinnati Royals as an example):

Calculate the player's modified points. The formula is:
2.0 * (field goals) * (1 - ((team assists) / (team field goals)))
+ 1.5 * (field goals) * ((team assists) / (team field goals))
+ 1.0 * (free throws)
+ 0.5 * (assists)
Plugging Robertson's statistics into the formula above we get 2495.93 modified points.
Calculate the player's modified shot attempts. The formula is:
1.00 * (field goals) * (1 - ((team assists) / (team field goals)))
+ 0.50 * (field goals) * ((team assists) / (team field goals))
+ 1.00 * ((field goal attempts) - (field goals))
+ 0.44 * (free throw attempts)
+ 0.50 * (assists)
Plugging Robertson's statistics into the formula above we get 2246.85 modified shot attempts.
Calculate league points per shot attempt. League points per shot attempt is equal to (league points) / (league field goal attempts + 0.44 * (league free throw attempts)). For the 1964-65 NBA this is 79641 / (71882 + 0.44 * 25604) = 0.9578.
Calculate marginal offense for each player. Marginal offense is equal to (modified points) - 0.92 * (league points per shot attempt) * (modified shot attempts). For Robertson this is 2495.93 - 0.92 * 0.9578 * 2246.85 = 515.06. Note that this formula may produce a negative result for some players.
Calculate marginal points per win. Marginal points per win reduces to 0.16 * (team points per game + opponent points per game). For the 1964-65 Royals this is 0.16 * (114.2 + 111.9) = 36.176.
Credit Offensive Win Shares to the players. Offensive Win Shares are credited using the following formula: (marginal offense) / (marginal points per win). Robertson gets credit for 516.06 / 36.176 = 14.27 Offensive Win Shares.
IV. Crediting Defensive Win Shares to Players

A. 1973-74 to present NBA

Crediting Defensive Win Shares to players is based on Dean Oliver's Defensive Rating. Defensive Rating is an estimate of the player's points allowed per 100 defensive possessions (please see Oliver's book for further details). Here is a description of the process (once again using LeBron James in 2008-09 as an example):

Calculate the Defensive Rating for each player. James's Defensive Rating in 2008-09 was 99.1.
Calculate marginal defense for each player. Marginal defense is equal to (player minutes played / team minutes played) * (team defensive possessions) * (1.08 * (league points per possession) - ((Defensive Rating) / 100)). For James this is (3054 / 19780) * 7341 * ((1.08 * 1.083) - (99.1 / 100)) = 202.5. Note that this formula may produce a negative result for some players.
Calculate marginal points per win. Marginal points per win reduces to 0.32 * (league points per game) * ((team pace) / (league pace)). For the 2008-09 Cavaliers this is 0.32 * 100.0 * (88.7 / 91.7) = 30.95.
Credit Defensive Win Shares to the players. Defensive Win Shares are credited using the following formula: (marginal defense) / (marginal points per win). James gets credit for 202.5 / 30.95 = 6.54 Defensive Win Shares.
B. 1951-52 to 1972-73 NBA

Prior to the 1973-74 season, the NBA did not track defensive rebounds, steals, or blocks, so allocating defensive credit is a difficult task. Nevertheless, here is the process for crediting Defensive Win Shares in those seasons (once again using Robertson in 1964-65 as an example):

Calculate team marginal defense. Team marginal defense is equal to 1.08 * (league points per shot attempt) * (team field goal attempts + 0.44 * (team free throw attempts)) - (opponent points). If you're wondering why we're using team shot attempts as opposed to opponent shot attempts, the answer is (a) we don't have opponent shot attempts prior to 1970-71 and (b) the system works better using team shot attempts. For the 1964-65 Royals we get 1.08 * 0.9578 * (7797 + 0.44 * 2866) - 8952 = 417.854.
Calculate the player's share of the team's marginal defense. The player's share of the team's marginal defense is equal to 0.25 * ((minutes played) / (team minutes played)) + 0.5 * ((total rebounds) / (team total rebounds)) + 0.25 * ((assists) / (team assists)). How did I get those weights? Modern Defensive Win Shares are most dependent on minutes played, defensive rebounds, steals, and blocks. I regressed DWS on those stats and then found the relative importance of each regressor (approximately 25% for minutes played, 35% for defensive rebounds, 25% for steals, and 15% for blocks). Since those defensive statistics are not available for past seasons, I used total rebounds as a proxy for defensive rebounds and blocks; and assists as a proxy for steals. A couple more notes: (1) prior to the 1964-65 season, team minutes played were not an official statistic, so for those seasons estimate the team's minutes played using the formula 5 * 48 * (team games) + 125; and (2) prior to the 1967-68 season, team total rebounds included team rebounds, so to account for this multiply the team total by 0.875. Getting back to our example, Robertson's share on the 1964-65 Royals is equal to 0.25 * (3421 / 19325) + 0.5 * (674 / (0.875 * 5387)) + 0.25 * (861 / 1843) = 0.2325.
Calculate marginal defense for each player. Marginal defense is equal to (team marginal defense) * (player share). For Robertson this is 417.854 * 0.2325 = 97.151. Note that this formula may produce a negative result for some players.
Calculate marginal points per win. Marginal points per win reduces to 0.16 * (team points per game + opponent points per game). For the 1964-65 Royals this is 0.16 * (114.2 + 111.9) = 36.176.
Credit defensive Win Shares to the players. Defensive Win Shares are credited using the following formula: (marginal defense) / (marginal points per win). Robertson gets credit for 97.151 / 36.176 = 2.69 Defensive Win Shares.
B. 1950-51 NBA

Prior to the 1951-52 season, the NBA did not track minutes played, so allocating defensive credit is an even more difficult task. Nevertheless, here is the process for crediting Defensive Win Shares in the 1950-51 season (using George Mikan as an example):

Calculate team marginal defense. Team marginal defense is equal to 1.08 * (league points per shot attempt) * (team field goal attempts + 0.44 * (team free throw attempts)) - (opponent points). If you're wondering why we're using team shot attempts as opposed to opponent shot attempts, the answer is (a) we don't have opponent shot attempts prior to 1970-71 and (b) the system works better using team shot attempts. For the 1950-51 Minneapolis Lakers we get 1.08 * 0.8553 * (5590 + 0.44 * 1989) - 5264 = 708.023.
Calculate the player's share of the team's marginal defense. The player's share of the team's marginal defense is equal to 0.25 * ((field goal attempts) / (team field goal attempts)) + 0.5 * ((total rebounds) / (team total rebounds)) + 0.25 * ((assists) / (team assists)). How did I get those weights? Modern Defensive Win Shares are most dependent on minutes played, defensive rebounds, steals, and blocks. I regressed DWS on those stats and then found the relative importance of each regressor (approximately 25% for minutes played, 35% for defensive rebounds, 25% for steals, and 15% for blocks). Since those defensive statistics are not available for past seasons, I used field goal attempts as a proxy for minutes played; total rebounds as a proxy for defensive rebounds and blocks; and assists as a proxy for steals. Note that prior to the 1967-68 season, team total rebounds included team rebounds, so to account for this multiply the team total by 0.875. Getting back to our example, Mikan's share on the 1950-51 Lakers is equal to 0.25 * (1584 / 5590) + 0.5 * (958 / (0.875 * 3049)) + 0.25 * (208 / 1408) = 0.2873.
Calculate marginal defense for each player. Marginal defense is equal to (team marginal defense) * (player share). For Mikan this is 708.023 * 0.2873 = 203.415. Note that this formula may produce a negative result for some players.
Calculate marginal points per win. Marginal points per win reduces to 0.16 * (team points per game + opponent points per game). For the 1950-61 Lakers this is 0.16 * (82.8 + 77.4) = 25.632.
Credit defensive Win Shares to the players. Defensive Win Shares are credited using the following formula: (marginal defense) / (marginal points per win). Mikan gets credit for 203.415 / 25.632 = 7.94 Defensive Win Shares.
B. 1946-47 to 1948-49 BAA and 1949-50 NBA

Prior to the 1950-51 season, the NBA did not track total rebounds, so allocating defensive credit is an almost impossible task. Nevertheless, here is the process for crediting Defensive Win Shares in those seasons (using Bob Feerick in 1946-47 as an example):

Calculate team marginal defense. Team marginal defense is equal to 1.08 * (league points per shot attempt) * (team field goal attempts + 0.44 * (team free throw attempts)) - (opponent points). If you're wondering why we're using team shot attempts as opposed to opponent shot attempts, the answer is (a) we don't have opponent shot attempts prior to 1970-71 and (b) the system works better using team shot attempts. For the 1946-47 Washington Capitols we get 1.08 * 0.6528 * (5794 + 0.44 * 1391) - 3836 = 680.412.
Calculate the player's share of the team's marginal defense. The player's share of the team's marginal defense is equal to 0.25 * ((field goal attempts) / (team field goal attempts)) + 0.5 * ((free throw attempts) / (team free throw attempts)) + 0.25 * ((assists) / (team assists)). How did I get those weights? Modern Defensive Win Shares are most dependent on minutes played, defensive rebounds, steals, and blocks. I regressed DWS on those stats and then found the relative importance of each regressor (approximately 25% for minutes played, 35% for defensive rebounds, 25% for steals, and 15% for blocks). Since those defensive statistics are not available for past seasons, I used field goal attempts as a proxy for minutes played; personal fouls as a proxy for defensive rebounds and blocks; and assists as a proxy for steals. Getting back to our example, Feerick's share on the 1946-47 Capitols is equal to 0.25 * (908 / 5794) + 0.5 * (142 / 1144) + 0.25 * (69 / 378) = 0.1469.
Calculate marginal defense for each player. Marginal defense is equal to (team marginal defense) * (player share). For Feerick this is 680.412 * 0.1469 = 99.953. Note that this formula may produce a negative result for some players.
Calculate marginal points per win. Marginal points per win reduces to 0.16 * (team points per game + opponent points per game). For the 1950-61 Lakers this is 0.16 * (73.8 + 63.9) = 22.032.
Credit defensive Win Shares to the players. Defensive Win Shares are credited using the following formula: (marginal defense) / (marginal points per win). Feerick gets credit for 99.953 / 22.032 = 4.54 Defensive Win Shares.
V. Putting It All Together

The final step of the process is to add Offensive Win Shares to Defensive Win Shares. In our examples, LeBron James total in 2008-09 is 13.73 + 6.54 = 20.27 Win Shares and Oscar Robertson total in 1964-65 is 14.27 + 2.69 = 16.96 Win Shares.

VI. Does This Work?

Because this metric is designed to estimate a player's contribution in terms of wins, it makes sense to see if the sum of player Win Shares for a particular team closely matches the team win total. For the 2008-09 Cavaliers the sum of player Win Shares is 67.9, while the team win total is 66, an error of 66 - 67.9 = -1.9 wins. For the 1964-65 Royals the sum of player Win Shares is 43.5, while the team total is 48, an error of 48 - 43.5 = 4.5 wins. These errors are actually close to the "typical" error; looking at all NBA teams since the 1962-63 season (the last season we have complete player splits), the average absolute error is 2.74 wins and the root mean squared error is 3.41 wins.

VII. Feedback

If you have any comments or questions about the Win Shares methodology, please send me some feedback.

VIII. Revision History

Version 4.0

Extended Win Shares back to the 1946-47 BAA season.
Version 3.1

Changed the calculation of marginal points per win. Prior to this update the league average marginal points per win was used for all players, but now the pace-adjusted league average is used instead.
Version 3.0

Extended Win Shares back to the 1951-52 NBA season.
Version 2.0

Changed the ratio of Win Shares to team wins from 3:1 to 1:1.
Removed the adjustment that forced team Win Shares to add up to team wins.
Modified the formulas to allow for the possibility of negative Win Shares.
Version 1.0

Initial release.

FOXHOUND
08-09-2013, 06:55 PM
Here's my problem with WS, just one small example but one that I think speaks volumes.

Tyson Chandler - 2011-12
Offensive Win Shares - 5.8 (8th)
Defensive Win Shares - 3.6 (11th)

So, apparently, Tyson Chandler was a better offensive contributor to wins than he was on defense during his DPOY year. To make matters worse he was top 5 in Offensive Win Shares during the 2012-13 all year before he got injured and missed 20 games, which ended up dropping him down to 12th overall.

The End

FOXHOUND
08-09-2013, 06:58 PM
Nothing is 100% for sure, not even the eye test. But everyone agrees that Lebron and Durant are 1-2 in the NBA. Its no coincidence that Lebron and Durant are 1-2 in PER, OWS, and any other stat that tries to define the totality of a players worth. The advanced stats are pretty on. There are often anomalies like Brook Lopez being top 5-6 in PER but for the most part there will be 1-2 anomalies in every 20 names and if you know anything about basketball, they will be very easy to spot. Otherwise its pretty spot on when comparing players that are playing similar minutes.

Or that just might be because LeBron James and Kevin Durant are in fact the two best players in the NBA, so any formula based on stats will always lead to them being at the top. A formula can't have obvious and glaring flaws, like the Lopez example you gave, but then be a great indicator whenever you feel like it.

If the science is broke then the formula is broke, thus making it an unreliable source of knowledge.

IKnowHoops
08-09-2013, 07:24 PM
Or that just might be because LeBron James and Kevin Durant are in fact the two best players in the NBA, so any formula based on stats will always lead to them being at the top. A formula can't have obvious and glaring flaws, like the Lopez example you gave, but then be a great indicator whenever you feel like it.

If the science is broke then the formula is broke, thus making it an unreliable source of knowledge.

Same can be said for the eye test or any measure of any kind.

IndyRealist
08-09-2013, 07:43 PM
Nothing is 100% for sure, not even the eye test. But everyone agrees that Lebron and Durant are 1-2 in the NBA. Its no coincidence that Lebron and Durant are 1-2 in PER, OWS, and any other stat that tries to define the totality of a players worth. The advanced stats are pretty on. There are often anomalies like Brook Lopez being top 5-6 in PER but for the most part there will be 1-2 anomalies in every 20 names and if you know anything about basketball, they will be very easy to spot. Otherwise its pretty spot on when comparing players that are playing similar minutes.

The eye test isn't even close to 100%, which is why advanced metrics were developed. There is a whole branch of psychology devoted to how human perception is flawed and how we make illogical judgments based on our emotions (like our favorite player). If scouts, coaches, managers, and GMs could accurately gauge a given player only by watching them play, then why were Kwame Brown and Andrea Bargnani #1 picks? Why was Darko Milicic picked over Dwayne Wade? Why did Jeremy Lin (a passable PG, whatever you think of him) go undrafted while Jonny Flynn went #6? The eye test gets it wrong all the time, and analysts need help. That's where metrics come in. Not to replace scouts and video, but augment them.

The comment about PER is spot on. The question isn't "Is PER valid?" the question is, "WHY does PER overrate some players and underrate others?" Just like WS, or WP, or RAPM, or any of the other acronyms, they are designed to show you a certain perspective. That doesn't make them "the Truth". Just like the eye test.

IndyRealist
08-09-2013, 07:50 PM
Or that just might be because LeBron James and Kevin Durant are in fact the two best players in the NBA, so any formula based on stats will always lead to them being at the top. A formula can't have obvious and glaring flaws, like the Lopez example you gave, but then be a great indicator whenever you feel like it.

If the science is broke then the formula is broke, thus making it an unreliable source of knowledge.

If you understand how formulas for modeling a population are created, then you would understand that there are always outliers. It's a mathematical approximation. That does not disprove the formula any more than saying the calculus for tracking weather formations is wrong because a raincloud veered over your house and dropped an inch of rain.

bagwell368
08-09-2013, 08:19 PM
Here's my problem with WS, just one small example but one that I think speaks volumes.

Tyson Chandler - 2011-12
Offensive Win Shares - 5.8 (8th)
Defensive Win Shares - 3.6 (11th)

So, apparently, Tyson Chandler was a better offensive contributor to wins than he was on defense during his DPOY year. To make matters worse he was top 5 in Offensive Win Shares during the 2012-13 all year before he got injured and missed 20 games, which ended up dropping him down to 12th overall.

The End

Chandler gets a ton of points for his shooting percentages. His TS% has been #1 3 years in a row. He doesn't seem to be a great offensive player according to guys that worship guys that shoot outside and have lots of moves. Chandler is effective, very effective and that's what Win Shares is about, not style.

By the way DPOY is subjective award, prone to all sorts of mistakes by the voters.

Chandler got hurt last year, that cuts down his value in Win Shares which is a counting stat. If you want to look at the "rate" go look at WS/48.

Chandler has had injury problems ever since 2007-2008. Calling out last year is strange since it's the 2nd most games he's played in the last 5 years.

Don't blame the stat because you don't understand it.

ztilzer31
08-09-2013, 08:30 PM
Here's my problem with WS, just one small example but one that I think speaks volumes.

Tyson Chandler - 2011-12
Offensive Win Shares - 5.8 (8th)
Defensive Win Shares - 3.6 (11th)

So, apparently, Tyson Chandler was a better offensive contributor to wins than he was on defense during his DPOY year. To make matters worse he was top 5 in Offensive Win Shares during the 2012-13 all year before he got injured and missed 20 games, which ended up dropping him down to 12th overall.

The End

I think that's what you got to take into context. Advanced stats are helpful but they don't tell the whole story.

Just learn how to utilize them.

ManRam
08-09-2013, 08:34 PM
Just another flawed stat that dudes use to judge individuals in a team sport where your success is directly tied to your teammates ability and actions.

I'd perhaps be inclined to believe this statement if there was even the slightest possibility that you actually understood what win shares, or the other "flawed stats" that you hate so much, really meant.

Kevj77
08-09-2013, 08:42 PM
Win Share to me is the most flawed of the advanced stats because of the team aspect of basketball. TS%, USG%, PER, eFG% and TOV% are all good stats though.

Advanced stats work much better in baseball.

Guppyfighter
08-09-2013, 08:54 PM
i hate advanced stats espeically in bball. Too many people think they know it all by just looking up advanced stats and never seeing a player play or understand the situation that player is in.


"I failed my high school stats class."

Guppyfighter
08-09-2013, 08:56 PM
this.

the best way to analyze a player is still the good old scouting..

Tell that to the Heat: http://espn.go.com/blog/truehoop/miamiheat/post/_/id/4356/how-advanced-stats-changed-chris-boshs-game

And the Spurs: http://blog.mysanantonio.com/spursnation/2013/03/05/revenge-of-the-nerds-how-advanced-stats-took-over-the-nba/

Guppyfighter
08-09-2013, 08:57 PM
you dont have to trust my opinion, but I'd rather take the opinion of an expirienced nba scout than believe in some "advanced" stats that never tell the entire story

James Kerti. Professional NBA scout, uses advanced stats.




And for all the guys who love to boast "Why you think all these teams have analytics departments?"

I'd be willing to bet they use them to evaluate within the context of their own team, and they understand the flaws in crossing them over to other teams. It makes a whole world of sense to hire analytics departments to find out your best lineup combos, your hot spots for each player, who thrives next to who, who gives the team a boost, who struggles next to who, what type of pace you should play etc...

They sure as hell don't use them to rank/compare/rate players from different teams thats for sure. These stats have a long way to go until you can do that without a million grains of salt.

Teams absolutely rank players and judge them relatively.


Especially for drafting. Ever seen a draft board?

ManRam
08-09-2013, 08:57 PM
Tell that to the Heat: http://espn.go.com/blog/truehoop/miamiheat/post/_/id/4356/how-advanced-stats-changed-chris-boshs-game

And the Spurs: http://blog.mysanantonio.com/spursnation/2013/03/05/revenge-of-the-nerds-how-advanced-stats-took-over-the-nba/

And pretty much every other team in basketball...

Guppyfighter
08-09-2013, 09:01 PM
And pretty much every other team in basketball...

I figured they'd prefer seeing the two teams who met in the finals use them.

Guppyfighter
08-09-2013, 09:04 PM
I find it funny when people try to tell you what these stats can't measure, but often times do measure those exact things.

IKnowHoops
08-09-2013, 09:20 PM
Tell that to the Heat: http://espn.go.com/blog/truehoop/miamiheat/post/_/id/4356/how-advanced-stats-changed-chris-boshs-game

And the Spurs: http://blog.mysanantonio.com/spursnation/2013/03/05/revenge-of-the-nerds-how-advanced-stats-took-over-the-nba/

Any rebuttal to this is poppycock.

FOXHOUND
08-09-2013, 09:26 PM
[/B]

Same can be said for the eye test or any measure of any kind.

The eye test is not even close to being a reliable judge of ability on it's own. It's completely tied to whoever is giving it, their perceptions on what they are seeing and whether they admit it or not the emotion tied to whatever player they are currently watching.

I don't dislike all advanced stats and I actually think a good amount of them are very useful. The formulated ones just seem silly, misused and overused. There's simply no way you can put number values on basic stats and make it mean something of substance. Basic stats themselves are flawed, so using them as a base of a formula will just give you a flawed foundation.

FOXHOUND
08-09-2013, 09:33 PM
If you understand how formulas for modeling a population are created, then you would understand that there are always outliers. It's a mathematical approximation. That does not disprove the formula any more than saying the calculus for tracking weather formations is wrong because a raincloud veered over your house and dropped an inch of rain.

Interesting analogy given there, considering that tracking weather formations is well understood to have it's flaws. It's pretty much impossible to predict the weather correctly, which is why so many people think the weather man on TV is wrong.

The main difference here is basketball is much more tied to things outside of one player. If you wanted to make a comparison to weather formation tracking methods then maybe advanced team stats would be more accurate. There's also the fact that not all basic stats are created equal, and those basic stats is what these formulas draw from.

justinnum1
08-09-2013, 09:34 PM
Funny how fans of players who have poor advanced stats, always try to devalue the stat.

FOXHOUND
08-09-2013, 09:41 PM
Chandler gets a ton of points for his shooting percentages. His TS% has been #1 3 years in a row. He doesn't seem to be a great offensive player according to guys that worship guys that shoot outside and have lots of moves. Chandler is effective, very effective and that's what Win Shares is about, not style.

By the way DPOY is subjective award, prone to all sorts of mistakes by the voters.

Chandler got hurt last year, that cuts down his value in Win Shares which is a counting stat. If you want to look at the "rate" go look at WS/48.

Chandler has had injury problems ever since 2007-2008. Calling out last year is strange since it's the 2nd most games he's played in the last 5 years.

Don't blame the stat because you don't understand it.

As a Knicks fan I have defended Chandler contributions on offense more than my fair share. I understand full well that while he is extremely limited he is at least very effective at the things he does provide for an offense, and that they are all valuable. Offensive rebounding, setting picks, put backs, his good PnR game, his high and 1 conversion rate etc.

At the same time, as valuable and underrated as those things can be, there is no way he is a top 10 contributor to wins on offense in the NBA at any time in his career. There is also no way that his contributions on offense have ever matched or exceeded what he brings on defense. As subjective as an award as DPOY can be there is absolutely nothing subjective in saying that Tyson Chandler is a far better defensive player than he is an offensive player.

Your post is a good example of advanced stats gone wrong. Tyson Chandler is a good offensive role player, and probably underrated at that. He is not even remotely close to a top 10 contributor to wins on offense.

FOXHOUND
08-09-2013, 09:43 PM
Funny how fans of players who have poor advanced stats, always try to devalue the stat.

I'm a huge Tyson Chandler fan, but I'm not going to sit here and say him leading the NBA in Offensive Rating the past THREE seasons now means that he's some sort of secret offensive weapon or something. Playing with Dirk and Melo are the reasons he has led in that stat more than anything.

Teeboy1487
08-09-2013, 09:49 PM
I'm not a fan of Advanced statistics because they do not tell the whole story. I prefer the eye test and that's actually watching the game and seeing how well a player plays for myself. Now, I do value PER and Efficiency, but that's about it. Honestly, when measuring stats for a career, I even think averages are flawed when you take into account for age and injuries. That's for another thread though.

D-Leethal
08-09-2013, 10:38 PM
James Kerti. Professional NBA scout, uses advanced stats.





Teams absolutely rank players and judge them relatively.


Especially for drafting. Ever seen a draft board?

Of course they are going to use them, it would be stupid not to. You quoted my post saying teams will use them within the context of their own roster and system to evaluate their own players in a multitude of different ways.

That makes sense, your analyzing players who are all playing within the same situation. The issue is when you try to use those stats across the board to analyze players in completely different situations. The articles you posted just reiterated my post you quoted, and did absolutely nothing to discredit it.

And I'm sure scouts do use advanced stats, and I'm sure thats about 1/5th of their evaluation when it comes to draft night, and I'm sure they do not let the numbers form their opinion for them, like posters such as yourself do on a daily basis.

D-Leethal
08-09-2013, 10:42 PM
Player A in situation A can post phenomenal stats. Player B in situation B could post crappier stats. That is not enough to say Player A is better than Player B. You swap situations and player B could all of a sudden post the better advanced statline.

Your stats are directly affected by your teammates - you know it is a team sport right? Thats precisely where the flaws lie. Every situation is different, and your statline is directly affected by your situation.

D-Leethal
08-09-2013, 10:49 PM
I think we can all agree advanced stats are best used as a supplemental tool. The issue lies when dudes like Guppy start saying things like Novak is a comparable defender to Battier and using their defensive ratings as 'proof', without acknowledging the glaring flaws in that stat, or the supreme idiocy in that statement. Stats IMO should never tell you something that your eye test doesn't agree with. I mean Bagwell literally said Chandler was a better offensive player than Melo. Posts like that are why I feel the need to raise awareness against these stats before it gets out of control, this forum has become 'if your a more efficient scorer and post a better advanced statline you are a better player'. Your efficiency for one is directly tied to the ability of your teammates. And two, efficient scoring is one iota in a stratosphere of basketball. Its a lot easier for some guys to score more efficiently than other guys. Why? Because of their teammates. It shouldn't be that hard to understand.

D-Leethal
08-09-2013, 10:52 PM
Hmmmm...

What about if the observer is a ESPN bred fan boy and doesn't know much about the game? The "evaluations" they make from that are likely to be less reliable than stats. I'd estimate generously 20% of the posters here do not suffer from some to a great extent from this disease.

What if it's a West Coast team that's seldom seen by an East Coast fan (2 team games, and maybe 1-2 other games on TNT)? Still think you have a grasp of a player you might see play 40 minutes a year?

What about historic comparisons. I'm the only person on PSD that's a regular that saw Bill Russell in/near his prime (the other 4 that claimed to be didn't have near enough associated knowledge/memories to prove their claims). You sure you want to go with written accounts and a few clips to talk about Russell in an authoritative manner?

Win Shares by the way (I'm sure someone pointed it out) is biased by being on a team that wins a lot more or a lot less than average. Better to take a look at the Win Shares a player has a percentage of the team total vs some other player in such cases. Not to mention ORtg and DRtg.

Clearly the game is more enjoyable and understandable the more ways you can look at it. I can look at it as a fan, former player, amateur historian, stat guy, and coach. It might not be the best background here, but, it's pretty good.

Your the guy who said Melo was the 6th best offensive player on the Knicks. You are the problem, not the solution. You are the only taking these stats to a level they are not even close to being ready to be at yet.

D-Leethal
08-09-2013, 10:59 PM
I'd perhaps be inclined to believe this statement if there was even the slightest possibility that you actually understood what win shares, or the other "flawed stats" that you hate so much, really meant.

Its not all that difficult dude, don't give yourself too much credit there. Its even easier to understand the flaws.

So are you saying you don't agree with my statement, or just trying to be a douche?

How about you give us a nice detailed description of how win shares accurately portrays how much a player contributes to his teams wins?

Instead of being a prick, how about you counter my argument instead of resorting to "derr you don't even know what the stats mean bro!"

D-Leethal
08-09-2013, 11:02 PM
Funny, I haven't heard one person actually find a way to discredit what I have been saying all day in this thread. I'm actually waiting for one of the knowledgable posters here to tell me where MY logic is flawed. Instead you have guys like Manram resorting to petty name calling and Chronz resorting to 'I have no interest in this debate'.

elledaddy
08-09-2013, 11:11 PM
Dude, its not like its calculus. Its easy to understand but it lengthy and long. It has to be to be accurate. It has to take everything into account.

Check it. Its straight forward on how this stat works.

http://www.basketball-reference.com/about/ws.html
Calculating Win Shares

I. Introduction

Stealing a page from baseball's Bill James, I decided to attempt to calculate basketball Win Shares. This article will describe how I came up with the Win Shares system for basketball. If you believe that any attempt to attribute team success to individual players is an abomination, then read no further, as this article will be of no interest to you.

II. What is a Win Share?

Bill James developed his system such that one win is equivalent to three Win Shares. My system deviates from James's system in three key ways:

In James's system, one win is equivalent to three Win Shares. In my system, one win is equivalent to one Win Share.
James made team Win Shares directly proportional to team wins. In his system, a baseball team that wins 80 games will have exactly 240 Win Shares, a baseball team that wins 90 games will have exactly 270 Win Shares, etc. In my system, a basketball team that wins 50 games will have about 50 Win Shares, give or take.
James did not allow for the possibility of negative Win Shares. In his system, the fewest number of Win Shares a player can have is zero. In my system, a player can have negative Win Shares. I justify this by thinking about it in the following way: a player with negative Win Shares was so poor that he essentially took away wins that his teammates had generated.
III. Crediting Offensive Win Shares to Players

A. 1977-78 to present NBA

Offensive Win Shares are credited to players based on Dean Oliver's points produced and offensive possessions. The formulas are quite detailed, so I would point you to Oliver's book Basketball on Paper for complete details. The process for crediting Offensive Win Shares is outlined below (using LeBron James of the 2008-09 Cleveland Cavaliers as an example):

Calculate points produced for each player. In 2008-09, James had an estimated 2345.9 points produced.
Calculate offensive possessions for each player. James had an estimated 1928.1 offensive possessions in 2008-09.
Calculate marginal offense for each player. Marginal offense is equal to (points produced) - 0.92 * (league points per possession) * (offensive possessions). For James this is 2345.9 - 0.92 * 1.083 * 1928.1 = 424.8. Note that this formula may produce a negative result for some players.
Calculate marginal points per win. Marginal points per win reduces to 0.32 * (league points per game) * ((team pace) / (league pace)). For the 2008-09 Cavaliers this is 0.32 * 100.0 * (88.7 / 91.7) = 30.95.
Credit Offensive Win Shares to the players. Offensive Win Shares are credited using the following formula: (marginal offense) / (marginal points per win). James gets credit for 424.8 / 30.95 = 13.73 Offensive Win Shares.
B. 1973-74 to 1976-77 NBA

The NBA did not track player turnovers until the 1977-78 season, and player turnovers are needed to calculate player possessions. However, the NBA did track turnovers at the team level from 1973-74 to 1976-77. Since player turnovers are the only thing holding us back from using the method outlined above, I have chosen to estimate player turnovers for this time period. Player turnovers are estimated as follows (using Kareem Abdul-Jabbar of the 1976-77 Los Angeles Lakers as an example):

Obtain an initial estimate of the player's turnovers. To do this use the following formula:
-0.0005075172 * (minutes played) * (player age)
- 0.0873982755 * (field goals)
+ 0.0925506598 * (field goal attempts)
+ 0.1566322510 * (free throw attempts)
+ 0.0449241773 * (total rebounds)
+ 0.2321637159 * (assists)
+ 0.2040169400 * (personal fouls)
Note that if this number is less than zero, then it should be rounded up to zero. Plugging Abdul-Jabbar's statistics into the formula above we get an estimate of 280.316 turnovers.
Find the sum of estimated turnovers for the players on the given team. The sum for the players on the 1976-77 Lakers is 1448.057.
Calculate the player's share of this total. Abdul-Jabbar's share of the team total is 280.316 / 1448.057 = 0.194.
Multiply the team's turnovers (adjusted for team turnovers) by the player's share. As mentioned, the NBA tracked turnovers at the team level in these seasons. However, the team totals include team turnovers (i.e., turnovers that are not attributed to an individual player). Thus, we multiply the team's turnovers by 0.985, then multiply this adjusted figure by the player's share. For Abdul-Jabbar this is 1538 * 0.985 * 0.194 = 293.9, which we round up to 294.
Now that we have this estimate, the method above is used to complete the calculation of Offensive Win Shares.

C. 1946-47 to 1948-49 BAA and 1949-50 to 1972-73 NBA

Because so many statistics are missing prior to the 1973-74 season (offensive rebounds, turnovers, etc.), we will not use Oliver's points produced and offensive possessions for this time period, although the basic framework will remain the same. Here is the process for crediting Offensive Win Shares prior to the 1973-74 season (using Oscar Robertson of the 1964-65 Cincinnati Royals as an example):

Calculate the player's modified points. The formula is:
2.0 * (field goals) * (1 - ((team assists) / (team field goals)))
+ 1.5 * (field goals) * ((team assists) / (team field goals))
+ 1.0 * (free throws)
+ 0.5 * (assists)
Plugging Robertson's statistics into the formula above we get 2495.93 modified points.
Calculate the player's modified shot attempts. The formula is:
1.00 * (field goals) * (1 - ((team assists) / (team field goals)))
+ 0.50 * (field goals) * ((team assists) / (team field goals))
+ 1.00 * ((field goal attempts) - (field goals))
+ 0.44 * (free throw attempts)
+ 0.50 * (assists)
Plugging Robertson's statistics into the formula above we get 2246.85 modified shot attempts.
Calculate league points per shot attempt. League points per shot attempt is equal to (league points) / (league field goal attempts + 0.44 * (league free throw attempts)). For the 1964-65 NBA this is 79641 / (71882 + 0.44 * 25604) = 0.9578.
Calculate marginal offense for each player. Marginal offense is equal to (modified points) - 0.92 * (league points per shot attempt) * (modified shot attempts). For Robertson this is 2495.93 - 0.92 * 0.9578 * 2246.85 = 515.06. Note that this formula may produce a negative result for some players.
Calculate marginal points per win. Marginal points per win reduces to 0.16 * (team points per game + opponent points per game). For the 1964-65 Royals this is 0.16 * (114.2 + 111.9) = 36.176.
Credit Offensive Win Shares to the players. Offensive Win Shares are credited using the following formula: (marginal offense) / (marginal points per win). Robertson gets credit for 516.06 / 36.176 = 14.27 Offensive Win Shares.
IV. Crediting Defensive Win Shares to Players

A. 1973-74 to present NBA

Crediting Defensive Win Shares to players is based on Dean Oliver's Defensive Rating. Defensive Rating is an estimate of the player's points allowed per 100 defensive possessions (please see Oliver's book for further details). Here is a description of the process (once again using LeBron James in 2008-09 as an example):

Calculate the Defensive Rating for each player. James's Defensive Rating in 2008-09 was 99.1.
Calculate marginal defense for each player. Marginal defense is equal to (player minutes played / team minutes played) * (team defensive possessions) * (1.08 * (league points per possession) - ((Defensive Rating) / 100)). For James this is (3054 / 19780) * 7341 * ((1.08 * 1.083) - (99.1 / 100)) = 202.5. Note that this formula may produce a negative result for some players.
Calculate marginal points per win. Marginal points per win reduces to 0.32 * (league points per game) * ((team pace) / (league pace)). For the 2008-09 Cavaliers this is 0.32 * 100.0 * (88.7 / 91.7) = 30.95.
Credit Defensive Win Shares to the players. Defensive Win Shares are credited using the following formula: (marginal defense) / (marginal points per win). James gets credit for 202.5 / 30.95 = 6.54 Defensive Win Shares.
B. 1951-52 to 1972-73 NBA

Prior to the 1973-74 season, the NBA did not track defensive rebounds, steals, or blocks, so allocating defensive credit is a difficult task. Nevertheless, here is the process for crediting Defensive Win Shares in those seasons (once again using Robertson in 1964-65 as an example):

Calculate team marginal defense. Team marginal defense is equal to 1.08 * (league points per shot attempt) * (team field goal attempts + 0.44 * (team free throw attempts)) - (opponent points). If you're wondering why we're using team shot attempts as opposed to opponent shot attempts, the answer is (a) we don't have opponent shot attempts prior to 1970-71 and (b) the system works better using team shot attempts. For the 1964-65 Royals we get 1.08 * 0.9578 * (7797 + 0.44 * 2866) - 8952 = 417.854.
Calculate the player's share of the team's marginal defense. The player's share of the team's marginal defense is equal to 0.25 * ((minutes played) / (team minutes played)) + 0.5 * ((total rebounds) / (team total rebounds)) + 0.25 * ((assists) / (team assists)). How did I get those weights? Modern Defensive Win Shares are most dependent on minutes played, defensive rebounds, steals, and blocks. I regressed DWS on those stats and then found the relative importance of each regressor (approximately 25% for minutes played, 35% for defensive rebounds, 25% for steals, and 15% for blocks). Since those defensive statistics are not available for past seasons, I used total rebounds as a proxy for defensive rebounds and blocks; and assists as a proxy for steals. A couple more notes: (1) prior to the 1964-65 season, team minutes played were not an official statistic, so for those seasons estimate the team's minutes played using the formula 5 * 48 * (team games) + 125; and (2) prior to the 1967-68 season, team total rebounds included team rebounds, so to account for this multiply the team total by 0.875. Getting back to our example, Robertson's share on the 1964-65 Royals is equal to 0.25 * (3421 / 19325) + 0.5 * (674 / (0.875 * 5387)) + 0.25 * (861 / 1843) = 0.2325.
Calculate marginal defense for each player. Marginal defense is equal to (team marginal defense) * (player share). For Robertson this is 417.854 * 0.2325 = 97.151. Note that this formula may produce a negative result for some players.
Calculate marginal points per win. Marginal points per win reduces to 0.16 * (team points per game + opponent points per game). For the 1964-65 Royals this is 0.16 * (114.2 + 111.9) = 36.176.
Credit defensive Win Shares to the players. Defensive Win Shares are credited using the following formula: (marginal defense) / (marginal points per win). Robertson gets credit for 97.151 / 36.176 = 2.69 Defensive Win Shares.
B. 1950-51 NBA

Prior to the 1951-52 season, the NBA did not track minutes played, so allocating defensive credit is an even more difficult task. Nevertheless, here is the process for crediting Defensive Win Shares in the 1950-51 season (using George Mikan as an example):

Calculate team marginal defense. Team marginal defense is equal to 1.08 * (league points per shot attempt) * (team field goal attempts + 0.44 * (team free throw attempts)) - (opponent points). If you're wondering why we're using team shot attempts as opposed to opponent shot attempts, the answer is (a) we don't have opponent shot attempts prior to 1970-71 and (b) the system works better using team shot attempts. For the 1950-51 Minneapolis Lakers we get 1.08 * 0.8553 * (5590 + 0.44 * 1989) - 5264 = 708.023.
Calculate the player's share of the team's marginal defense. The player's share of the team's marginal defense is equal to 0.25 * ((field goal attempts) / (team field goal attempts)) + 0.5 * ((total rebounds) / (team total rebounds)) + 0.25 * ((assists) / (team assists)). How did I get those weights? Modern Defensive Win Shares are most dependent on minutes played, defensive rebounds, steals, and blocks. I regressed DWS on those stats and then found the relative importance of each regressor (approximately 25% for minutes played, 35% for defensive rebounds, 25% for steals, and 15% for blocks). Since those defensive statistics are not available for past seasons, I used field goal attempts as a proxy for minutes played; total rebounds as a proxy for defensive rebounds and blocks; and assists as a proxy for steals. Note that prior to the 1967-68 season, team total rebounds included team rebounds, so to account for this multiply the team total by 0.875. Getting back to our example, Mikan's share on the 1950-51 Lakers is equal to 0.25 * (1584 / 5590) + 0.5 * (958 / (0.875 * 3049)) + 0.25 * (208 / 1408) = 0.2873.
Calculate marginal defense for each player. Marginal defense is equal to (team marginal defense) * (player share). For Mikan this is 708.023 * 0.2873 = 203.415. Note that this formula may produce a negative result for some players.
Calculate marginal points per win. Marginal points per win reduces to 0.16 * (team points per game + opponent points per game). For the 1950-61 Lakers this is 0.16 * (82.8 + 77.4) = 25.632.
Credit defensive Win Shares to the players. Defensive Win Shares are credited using the following formula: (marginal defense) / (marginal points per win). Mikan gets credit for 203.415 / 25.632 = 7.94 Defensive Win Shares.
B. 1946-47 to 1948-49 BAA and 1949-50 NBA

Prior to the 1950-51 season, the NBA did not track total rebounds, so allocating defensive credit is an almost impossible task. Nevertheless, here is the process for crediting Defensive Win Shares in those seasons (using Bob Feerick in 1946-47 as an example):

Calculate team marginal defense. Team marginal defense is equal to 1.08 * (league points per shot attempt) * (team field goal attempts + 0.44 * (team free throw attempts)) - (opponent points). If you're wondering why we're using team shot attempts as opposed to opponent shot attempts, the answer is (a) we don't have opponent shot attempts prior to 1970-71 and (b) the system works better using team shot attempts. For the 1946-47 Washington Capitols we get 1.08 * 0.6528 * (5794 + 0.44 * 1391) - 3836 = 680.412.
Calculate the player's share of the team's marginal defense. The player's share of the team's marginal defense is equal to 0.25 * ((field goal attempts) / (team field goal attempts)) + 0.5 * ((free throw attempts) / (team free throw attempts)) + 0.25 * ((assists) / (team assists)). How did I get those weights? Modern Defensive Win Shares are most dependent on minutes played, defensive rebounds, steals, and blocks. I regressed DWS on those stats and then found the relative importance of each regressor (approximately 25% for minutes played, 35% for defensive rebounds, 25% for steals, and 15% for blocks). Since those defensive statistics are not available for past seasons, I used field goal attempts as a proxy for minutes played; personal fouls as a proxy for defensive rebounds and blocks; and assists as a proxy for steals. Getting back to our example, Feerick's share on the 1946-47 Capitols is equal to 0.25 * (908 / 5794) + 0.5 * (142 / 1144) + 0.25 * (69 / 378) = 0.1469.
Calculate marginal defense for each player. Marginal defense is equal to (team marginal defense) * (player share). For Feerick this is 680.412 * 0.1469 = 99.953. Note that this formula may produce a negative result for some players.
Calculate marginal points per win. Marginal points per win reduces to 0.16 * (team points per game + opponent points per game). For the 1950-61 Lakers this is 0.16 * (73.8 + 63.9) = 22.032.
Credit defensive Win Shares to the players. Defensive Win Shares are credited using the following formula: (marginal defense) / (marginal points per win). Feerick gets credit for 99.953 / 22.032 = 4.54 Defensive Win Shares.
V. Putting It All Together

The final step of the process is to add Offensive Win Shares to Defensive Win Shares. In our examples, LeBron James total in 2008-09 is 13.73 + 6.54 = 20.27 Win Shares and Oscar Robertson total in 1964-65 is 14.27 + 2.69 = 16.96 Win Shares.

VI. Does This Work?

Because this metric is designed to estimate a player's contribution in terms of wins, it makes sense to see if the sum of player Win Shares for a particular team closely matches the team win total. For the 2008-09 Cavaliers the sum of player Win Shares is 67.9, while the team win total is 66, an error of 66 - 67.9 = -1.9 wins. For the 1964-65 Royals the sum of player Win Shares is 43.5, while the team total is 48, an error of 48 - 43.5 = 4.5 wins. These errors are actually close to the "typical" error; looking at all NBA teams since the 1962-63 season (the last season we have complete player splits), the average absolute error is 2.74 wins and the root mean squared error is 3.41 wins.

VII. Feedback

If you have any comments or questions about the Win Shares methodology, please send me some feedback.

VIII. Revision History

Version 4.0

Extended Win Shares back to the 1946-47 BAA season.
Version 3.1

Changed the calculation of marginal points per win. Prior to this update the league average marginal points per win was used for all players, but now the pace-adjusted league average is used instead.
Version 3.0

Extended Win Shares back to the 1951-52 NBA season.
Version 2.0

Changed the ratio of Win Shares to team wins from 3:1 to 1:1.
Removed the adjustment that forced team Win Shares to add up to team wins.
Modified the formulas to allow for the possibility of negative Win Shares.
Version 1.0

Initial release.

This is why I wrote in the OP, dont just post the definition of win shares. Who the hell wrote this? It says that whoever wrote this took it from somebody who did it for baseball then modified it for Basketball. But its his( whoever wrote this) own calculation. Who the hell is he and who made him qualified

Tony_Starks
08-09-2013, 11:13 PM
I think we can all agree advanced stats are best used as a supplemental tool. The issue lies when dudes like Guppy start saying things like Novak is a comparable defender to Battier and using their defensive ratings as 'proof', without acknowledging the glaring flaws in that stat, or the supreme idiocy in that statement. Stats IMO should never tell you something that your eye test doesn't agree with. I mean Bagwell literally said Chandler was a better offensive player than Melo. Posts like that are why I feel the need to raise awareness against these stats before it gets out of control, this forum has become 'if your a more efficient scorer and post a better advanced statline you are a better player'. Your efficiency for one is directly tied to the ability of your teammates. And two, efficient scoring is one iota in a stratosphere of basketball. Its a lot easier for some guys to score more efficiently than other guys. Why? Because of their teammates. It shouldn't be that hard to understand.

Another example is Dwight Howard. Looking at advanced stats you would think he's some kind great offensive player. Actually watching him play you see that he's extremely offensively limited and not the type of player you run a offense through.....

All-In
08-09-2013, 11:18 PM
Weather you like it or not advanced stats is a part of the game…. advanced stats can say that a certain player’s PPP is a 1.11 on P&R ballhandler plays but a .67 on P&R rollman plays….so in certain situation’s when that player is in a P&R the opponent will be more aware to “ice” or “blu” the P&R so the play would be pushed towards the baseline….players use them all the time to tell them where their more efficient from…opponents weak spots…if its able to help with a teams strategy it can help determine how good a player is......some of them not all.....some adavnced stats are pointless some are great.....watching the game helps…but having large sample sized numbers can be more convincing sometimes

D-Leethal
08-09-2013, 11:41 PM
Weather you like it or not advanced stats is a part of the game…. advanced stats can say that a certain player’s PPP is a 1.11 on P&R ballhandler plays but a .67 on P&R rollman plays….so in certain situation’s when that player is in a P&R the opponent will be more aware to “ice” or “blu” the P&R so the play would be pushed towards the baseline….players use them all the time to tell them where their more efficient from…opponents weak spots…if its able to help with a teams strategy it can help determine how good a player is......some of them not all.....some adavnced stats are pointless some are great.....watching the game helps…but having large sample sized numbers can be more convincing sometimes

The bold is straight up completely wrong.

D-Leethal
08-09-2013, 11:47 PM
I ran the same test for shooting that I ran for rebounds. For the 2008-09 season, I ran regression for each of the five positions. Each row of the regression was a single team for that year, and I checked how each position's shooting (measured by eFG%) affected the average of the other four positions (the simple average, not weighted by attempts).

It turns out that there is a strong positive correlation in shooting percentage among teammates. If one teammate shoots accurately, the rest of the team gets carried along.

Here are the numbers (updated, see end of post):

PG: slope 0.30, correlation 0.63
SG: slope 0.40, correlation 0.62
SF: slope 0.26, correlation 0.27
PF: slope 0.28, correlation 0.27
-C: slope 0.27, correlation 0.43

To read one line off the chart: for every one percentage point increase in shooting percentage by the SF (say, from 47% to 48%), you saw an increase of 0.26% in each of his teammates (say, from 47% to 47.26%).

The coefficients are a lot more important than they look at first glance, because they represent a change in the average of all four teammates. Suppose all five teammates took the same number of shots (which they don't, but never mind right now). That means that when the SF makes one extra field goal, each teammate also makes an extra 0.26, for a team team total of 1.04 extra field goals.

That's a huge effect.

And, it makes sense, if my logic is right. Suppose you have a team where everyone has a talent of .450, but then you get a new guy on the team (player X) with a talent of .550. You're going to want him to shoot more often than the other players. For instance, if X and another guy are equally open for a roughly equal shot, you're going to want to give the ball to X. Even if Y is a little more open than X, you'll figure that X will still outshoot Y -- maybe not .550 to .450, but, in this situation, maybe .500 to .450. So X gets the ball more often.

But, then, the defense will concentrate a little more on X, and a little less on the .450 guys. That means X might see his percentage drop from .550 to .500, say. But the extra attention to X creates more open shots for the .450 guys, and they improve to (say) .480 each.

Most of the new statistics simply treat FG% as if it's solely the achievement of the player taking the shot, when, it seems, it is very significantly influenced by his teammates.
.

D-Leethal
08-09-2013, 11:58 PM
Dan Rosenbaum:


But do not interpret me to be saying that statistical analysis has no place in basketball. Instead the point I am trying to make is that basketball people are right to be skeptical of statistical analysis, because analyses based upon an overly simple model of the game of basketball often can be more misleading than useful.

Guppyfighter
08-10-2013, 12:40 AM
Everything D-Leethal saying is wrong and it hurts to read how he thinks teams operate.


Mysynergysports makes 13,000,000 a year off of every NBA teams. You think teams are paying that much to not make conclusions?


I think you really do not understand how accurate advanced stats and how much NBA teams use them. And they use better ones then we have access too.

I am also not sure how people are saying all advanced stats are based on simple models. TS is not based on a simple model, there was regression analysis done.

He's basically arguing advanced stats aren't as accurate as there actual correlation rates because of how he FEELS about how certain players grade out. What you feel doesn't ****ing matter. If the stats didn't show stuff with accuracy than you'd be able to falsify them. If you can't be bothered to try to falsify the stats your argument is moot.

Guppyfighter
08-10-2013, 12:44 AM
I ran the same test for shooting that I ran for rebounds. For the 2008-09 season, I ran regression for each of the five positions. Each row of the regression was a single team for that year, and I checked how each position's shooting (measured by eFG%) affected the average of the other four positions (the simple average, not weighted by attempts).

It turns out that there is a strong positive correlation in shooting percentage among teammates. If one teammate shoots accurately, the rest of the team gets carried along.

Here are the numbers (updated, see end of post):

PG: slope 0.30, correlation 0.63
SG: slope 0.40, correlation 0.62
SF: slope 0.26, correlation 0.27
PF: slope 0.28, correlation 0.27
-C: slope 0.27, correlation 0.43

To read one line off the chart: for every one percentage point increase in shooting percentage by the SF (say, from 47% to 48%), you saw an increase of 0.26% in each of his teammates (say, from 47% to 47.26%).

The coefficients are a lot more important than they look at first glance, because they represent a change in the average of all four teammates. Suppose all five teammates took the same number of shots (which they don't, but never mind right now). That means that when the SF makes one extra field goal, each teammate also makes an extra 0.26, for a team team total of 1.04 extra field goals.

That's a huge effect.

And, it makes sense, if my logic is right. Suppose you have a team where everyone has a talent of .450, but then you get a new guy on the team (player X) with a talent of .550. You're going to want him to shoot more often than the other players. For instance, if X and another guy are equally open for a roughly equal shot, you're going to want to give the ball to X. Even if Y is a little more open than X, you'll figure that X will still outshoot Y -- maybe not .550 to .450, but, in this situation, maybe .500 to .450. So X gets the ball more often.

But, then, the defense will concentrate a little more on X, and a little less on the .450 guys. That means X might see his percentage drop from .550 to .500, say. But the extra attention to X creates more open shots for the .450 guys, and they improve to (say) .480 each.

Most of the new statistics simply treat FG% as if it's solely the achievement of the player taking the shot, when, it seems, it is very significantly influenced by his teammates.

TIL D-Leethal thinks using advanced stats to argue against the usage of advanced stats make sense.

We already knew this. We already know the effect teammates and usage has on efficiency. That's not a problem with the stats we are using.

Chronz
08-10-2013, 01:04 AM
Funny, I haven't heard one person actually find a way to discredit what I have been saying all day in this thread. I'm actually waiting for one of the knowledgable posters here to tell me where MY logic is flawed. Instead you have guys like Manram resorting to petty name calling and Chronz resorting to 'I have no interest in this debate'.
I never said that. And I've responded to your posts. R u drunk ?

bearadonisdna
08-10-2013, 01:05 AM
Weather you like it or not advanced stats is a part of the game…. advanced stats can say that a certain player’s PPP is a 1.11 on P&R ballhandler plays but a .67 on P&R rollman plays….so in certain situation’s when that player is in a P&R the opponent will be more aware to “ice” or “blu” the P&R so the play would be pushed towards the baseline….players use them all the time to tell them where their more efficient from…opponents weak spots…if its able to help with a teams strategy it can help determine how good a player is......some of them not all.....some adavnced stats are pointless some are great.....watching the game helps…but having large sample sized numbers can be more convincing sometimes

When the stats are too complicated is where the problem is. And where it becomes deceptive. Something like ur example is more of a percentage of what a player does per possession. thats what u were using right? point per possession. Thats not even all that advanced as more just good scouting and percentages that u would expect from professional sports leagues.
When trying to diagnose tendencies and stuff like that, Point per possession in situations does seem useful. You would expect professionals to use stuff like that. That formula for Win shares seems too damn complicated to be practical.

el hidalgo
08-10-2013, 01:34 AM
advanced stats were created by john hollinger to discredit kobe and credit lebron

Crackadalic
08-10-2013, 01:52 AM
Win shared are helpful to a degree but it has a few flaws especially defensively. 2 season ago I believe Carlos boozer led he's team in defensive win shares. Says a lot. I think stats like rapm are more helpful when it comes to rating players

Guppyfighter
08-10-2013, 03:06 AM
It's flaw is grading defense. It's spot on with offense.

Pacerlive
08-10-2013, 08:42 AM
I think we can all agree advanced stats are best used as a supplemental tool. The issue lies when dudes like Guppy start saying things like Novak is a comparable defender to Battier and using their defensive ratings as 'proof', without acknowledging the glaring flaws in that stat, or the supreme idiocy in that statement. Stats IMO should never tell you something that your eye test doesn't agree with. I mean Bagwell literally said Chandler was a better offensive player than Melo. Posts like that are why I feel the need to raise awareness against these stats before it gets out of control, this forum has become 'if your a more efficient scorer and post a better advanced statline you are a better player'. Your efficiency for one is directly tied to the ability of your teammates. And two, efficient scoring is one iota in a stratosphere of basketball. Its a lot easier for some guys to score more efficiently than other guys. Why? Because of their teammates. It shouldn't be that hard to understand.
It almost sounds like you want percent assisted to hold a negative value in some advance metric in evaluating a player offensive production.

nycericanguy
08-10-2013, 08:49 AM
I've never really found a good stat for defense, is there one out there that's not largely team dependent?

Also I don't even understand how defense can be judged and charted individually. There are so many things that go into defense that can not be equated to a number.

So I stick to the more basic advanced stats offensively. Like TS, EFG and ORTG. But even those have to be taken in context of course. Those stats will tell you Novak & Chandler are offensive monsters, but in reality they are just two guys that need to be spoon fed wide open shots in order to score.

What's interesting to me is that Deng was voted ahead of Gallo, even though Gallo's advanced stats blow away Deng. Heck even his basic stats are about the same, except Gallo needed a lot less shots and a lot less minutes to achieve his.

3RDASYSTEM
08-10-2013, 09:31 AM
i hate advanced stats espeically in bball. Too many people think they know it all by just looking up advanced stats and never seeing a player play or understand the situation that player is in.

I don't hate advance stats ut I agree with your view on playing and people not understanding the situation(or just choose to ignore it flat out)

its the situation I be telling them about IVERSON having to shoot 25x a game or more, playing with 3ppg scorers will leave you no choice

had actual former HOF'ers came out with this stuff then i'd view it differently but not people who feel titled to do something just because they have watched the game for years

its just like I didn't get the whole JORDAN shooting 50pct from the field being a big deal because he had killer game, but then that's when I realize they are trying to separate JORDAN from others with advanced stats that are not needed

I didn't need to see JORDAN PER his first 7yrs nor his last 3 yrs with the BULLS to see he had dipped off as a player PER wise, because it showed on the floor

that's why I don't care for advanced stats, it shows a good player but not the situation/gameplan/pressure which is really what it comes down to, making plays and winning, hand in hand

its like we all(or most) knew that bean is a overrating 'clutch' player and when they leaked out the stats it showed right? well I been saying it for years and I don't never need to see no stat to justify it, watching pretty much sums it all up, playing also helps the understanding drastically, coaching and scouting and trainer also are big time helpers

watching for years makes you a frequent viewer and fanatic of the game

this PER/win share **** remind me how people use to come to games dressed in the RODMAN colors with hair and stuff, sort of like how BIRDMAN is now today just a bunch of colorful **** that a few handful will grasp onto and run with(see the few RODMAN-BIRDMAN triplets in the crowd at games)

all the best players usually have high PER unless situation was playing with a weak cast support

ALCINDOR didn't need a win ring to be on WILT's level, he was there day1, WILT was old when ALCINDOR came into league so how could WILT match the young lion, same way I looked at DREAM to SHAQ? he couldn't be he could bang with him and make it a tit for tat like contest because he was once that young lion, same way I looked at DR J impact/game in ABA to the NBA in comparison to MAGIC-BIRD, same way I looked at KIDD to G.HILL to IVERSON when he came into league and went against oldman JORDAN same way I looked at BRON and a few others, a win share or PER can change nothing of my view I have towards those type of players, and more I didn't mention

3RDASYSTEM
08-10-2013, 09:58 AM
What if we dont trust what you think your eyes are watching?

The eyes never lie, don't you know and understand a lot of these owners don't play for most part and end up having final decision over the gm? that's where the problem is, they don't they let the gm do they job

but that is no doubt the best way to analyze a player, busts and bad picks and bad signings happen in all eras of NBA history, see KONCAK and DUDLEY for past ridiculous deals, its a part of the game, bad deals and player busts = drama and storylines

now the trick is what kind of player are you scouting for to fit your team or are you taking best player avail or best player you think with the fake highest 'ceiling', my blueprint approach would be similar to the 96' BULLS or 80's/00' lakers or this latest HEAT team, the best/most dominant are usually the easiest to weed out but also hard to get...see BRON-JORDAN-SHAQ for examples

it makes it a lot easier to scout for the team needs with 1-2 franchise players already in your pocket

old school scouting is way just like pretty much everything in life is, ask LARRY BROWN and a few other nba legends

I just spoke with a actual coach the other day and he said scouting was the best way of evaluating a player, even better in big game meaningful situations

macc
08-10-2013, 10:06 AM
A GM needs to put together his own money ball team and we'll see how far they go. Would be interesting to see to say the least.

IKnowHoops
08-10-2013, 10:30 AM
This is why I wrote in the OP, dont just post the definition of win shares. Who the hell wrote this? It says that whoever wrote this took it from somebody who did it for baseball then modified it for Basketball. But its his( whoever wrote this) own calculation. Who the hell is he and who made him qualified

Hey, if you can't understand this by looking at the formula, then you can't. I looked at the formulas first. Then once I understand how the formula works, then I can either agree or disagree with it. It has nothing to do with who made it. I looked at the formula, and said, OK that makes sense. Try and understand the formula before you disagree. Like I said before its long. So it will take a while to read through and understand each formula for each phase of a players game. But at the end of the day, everyone gets the same formula. So you plug in everyones different number into a fixed formula. Again nothing is 100% but there is a lot of accuracy. When you have two players like TMac and Kobe, where people are split down the middle on who was better in there prime, advanced stats are a great tool to tell you who was the more effective player. The eye test will split you down the middle, the advanced stats will prove who actually played better. May not tell you who was better, but it will tell you who played better on there respective teams.

Pierzynski4Prez
08-10-2013, 10:42 AM
this.

the best way to analyze a player is still the good old scouting..

So when you watch the game are you watching every player on the court and keeping track of when they do what every time down the court? Or is your scouting report something like "well there was that one game, or few games, where he was hot against so and so team so he is better then so and so player. Plus he averages more ppg than the other guy so he's definitely better."

Advanced stats keep track of those things you look for and calculate basically that players scouting report for you so you don't have to rate a player off you trying to calculate every players every move for 82 games in your head.

FOXHOUND
08-10-2013, 12:02 PM
It's flaw is grading defense. It's spot on with offense.

So you agree that Tyson Chandler is a perennial top 10 offensive contributor in the NBA? Good to know.

meloman1592
08-10-2013, 12:12 PM
It's ********. Most advanced stats are besides PER. just watch the game

Gators123
08-10-2013, 12:26 PM
It's ********. Most advanced stats are besides PER. just watch the game

30 teams x 82 games a year x 13-15 players on every roster.

How many minutes did you watch Jeff Adrien play last year?

Clippersfan86
08-10-2013, 12:52 PM
Uh oh... Guppy got out of his cage. You guys unleashed the mother of all ranting into this thread. :laugh:

Bookey
08-10-2013, 12:53 PM
**** Advanced Stats

el hidalgo
08-10-2013, 01:12 PM
It'd be interesting to see the demographics of those who do and don't believe in advanced stats

Crackadalic
08-10-2013, 01:22 PM
I don't understand how people can't use both the eye test and advance stats to evaluate a player. Both seem flawed in its own way but paints a bigger picture when both are used

Gators123
08-10-2013, 01:23 PM
It'd be interesting to see the demographics of those who do and don't believe in advanced stats

I'm pretty sure every NBA team uses them :shrug:

elledaddy
08-10-2013, 01:42 PM
Hey, if you can't understand this by looking at the formula, then you can't. I looked at the formulas first. Then once I understand how the formula works, then I can either agree or disagree with it. It has nothing to do with who made it. I looked at the formula, and said, OK that makes sense. Try and understand the formula before you disagree. Like I said before its long. So it will take a while to read through and understand each formula for each phase of a players game. But at the end of the day, everyone gets the same formula. So you plug in everyones different number into a fixed formula. Again nothing is 100% but there is a lot of accuracy. When you have two players like TMac and Kobe, where people are split down the middle on who was better in there prime, advanced stats are a great tool to tell you who was the more effective player. The eye test will split you down the middle, the advanced stats will prove who actually played better. May not tell you who was better, but it will tell you who played better on there respective teams.


Ok you understand the formula but the formula is not based on facts. Look at the first sentence.....

1.In James's system, one win is equivalent to three Win Shares. In my system, one win is equivalent to one Win Share.

So if whoever wrote this said " in my system" one win is equivalent to six win shares, the formula would still basically make sense BUT that doesnt mean it should be used. All he did was piggy back off the original formula used for an entire different sport and adjusted it to his liking, then made the formula support what he was trying to accomplish. Whats stopping you( and I mean YOU, not man in general) from doing the excact same thing but just changing the one or two things about it and calling it a completely different metric system? Since you understand the formula( and I believe you do) how would you explain it to someone in "layman's terms"?
Use your own words and briefly describe it.( If you feel up to it, ofcourse)


PS.... This wasnt about " advanced stats" in general, just Win shares. So based off this formula even if Kobe and TMAC had similar numbers in their Prime, Kobe's team won more games so his win shares would be higher. It actually would even be close.

Just my 2 cents on it

IKnowHoops
08-10-2013, 02:14 PM
Ok you understand the formula but the formula is not based on facts. Look at the first sentence.....

1.In James's system, one win is equivalent to three Win Shares. In my system, one win is equivalent to one Win Share.

So if whoever wrote this said " in my system" one win is equivalent to six win shares, the formula would still basically make sense BUT that doesnt mean it should be used. All he did was piggy back off the original formula used for an entire different sport and adjusted it to his liking, then made the formula support what he was trying to accomplish. Whats stopping you( and I mean YOU, not man in general) from doing the excact same thing but just changing the one or two things about it and calling it a completely different metric system? Since you understand the formula( and I believe you do) how would you explain it to someone in "layman's terms"?
Use your own words and briefly describe it.( If you feel up to it, ofcourse)


PS.... This wasnt about " advanced stats" in general, just Win shares. So based off this formula even if Kobe and TMAC had similar numbers in their Prime, Kobe's team won more games so his win shares would be higher. It actually would even be close.

Just my 2 cents on it

Ok, do understand this. In the case of counting a win as 1 win shares or 3 win shares, it makes absolutely no difference because all it does is in effect give a wider range of scores that still have the exact same relationship.

for example.
Lets say 1 win equals 1 win share
Heat 50 wins = 50 win shares
Bulls 40 wins = 40 win shares

1 win = 3 win shares
heat 50 wins = 150 win shares
Bulls 40 wins = 120 win shares

1 win = 10 win shares
Heat 50 wins = 500 win shares
Bulls 40 wins = 400 win shares

See it does not matter what you times it by because 50 to 40 is = 150 to 120 is = 500 to 400. The ratio is the same, the only thing that changed was the point system. Feel me?

JeremiahWing
08-10-2013, 02:16 PM
Individual players aren't solely responsible for their numbers in basketball. Unlike baseball players who hit and field by themselves, basketball players rely on good passes and screens to get open looks, good help defense, good transition play, etc, etc, etc. Again, the idea that all of the circumstances of individual can be quantified is a stupid idea on its face.

Guppyfighter
08-10-2013, 04:03 PM
So you agree that Tyson Chandler is a perennial top 10 offensive contributor in the NBA? Good to know.

If you don't think offensive rebounds, screens, and all the little things help a team score more points you are dumb.

He dunks the ball and gets putback (low percentage shots) at an extremely high rate while keeping possessions alive at an elite level.

Having an elite offensive rebounder, the best in the league in fact, is so important to an offense whether you believe that or not.

Guppyfighter
08-10-2013, 04:03 PM
Individual players aren't solely responsible for their numbers in basketball. Unlike baseball players who hit and field by themselves, basketball players rely on good passes and screens to get open looks, good help defense, good transition play, etc, etc, etc. Again, the idea that all of the circumstances of individual can be quantified is a stupid idea on its face.


Advanced stats keep track of all that.

Guppyfighter
08-10-2013, 04:05 PM
Uh oh... Guppy got out of his cage. You guys unleashed the mother of all ranting into this thread. :laugh:

Big talk coming from someone who thinks Chris Paul let his team down in the playoffs.

The_Jamal
08-10-2013, 04:43 PM
So you agree that Tyson Chandler is a perennial top 10 offensive contributor in the NBA? Good to know.

Yes, because Tyson's efficiency is historically great. He's 5th all time in TS%, he's 1st all-time in eFG% and he's 10th all-time in ORBG%. In essence, he's the most reliable 10 PPG in NBA history, he doesn't need any offensive ball to do it and he's one of the best in the NBA at creating extra offensive possesions. There's an incredible amount of value he brings offensively

WadeKobe
08-10-2013, 05:02 PM
Yes, because Tyson's efficiency is historically great. He's 5th all time in TS%, he's 1st all-time in eFG% and he's 10th all-time in ORBG%. In essence, he's the most reliable 10 PPG in NBA history, he doesn't need any offensive ball to do it and he's one of the best in the NBA at creating extra offensive possesions. There's an incredible amount of value he brings offensively

:clap:

skierdude44
08-10-2013, 05:46 PM
Individual players aren't solely responsible for their numbers in basketball. Unlike baseball players who hit and field by themselves, basketball players rely on good passes and screens to get open looks, good help defense, good transition play, etc, etc, etc. Again, the idea that all of the circumstances of individual can be quantified is a stupid idea on its face.

Baseball events don't occur in a vacuum either. A pitcher's effectiveness is tied to the range of his defenders, a hitter is more effective at driving in runs if men are frequently on base in front of him, pitchers attack hitters differently with men on base then with the bases empty, etc. It's as much of a team game as basketball it's just a different game; and advanced stats are just as applicable in either game.


If you don't think offensive rebounds, screens, and all the little things help a team score more points you are dumb.

He dunks the ball and gets putback (low percentage shots) at an extremely high rate while keeping possessions alive at an elite level.

Having an elite offensive rebounder, the best in the league in fact, is so important to an offense whether you believe that or not.

Just to expand on this point for a second. I think we can all agree that basketball is a sport of efficiency. More than any other sport everything has a time limit on it. There's time limits on how long it takes you to in bound the ball, to bring it to the front court, to get off a shot, to how long you can camp out in the lane, everything. There's also a fixed amount of time for a game, unlike baseball where there is no clock. When looking at the sport through the lens of efficiency you can see the value of a player like Chandler who makes the most of his opportunities, doesn't do things that would hurt his team's offensive efficiency because he's trying to play beyond his offensive capabilities, and does little things that add extra possessions that afford his team more opportunities to score (as Guppyfighter noted above). Conversely, a player like J.R. Smith is going to score more points than Chandler because he takes more shots, but he's not more efficient because he frequently takes ill-advised and/or contested shots. In many ways, Chandler balances out Smith's inefficiency by cleaning up his misses and at times providing Smith with a second opportunity to score.

It's important to understand what exactly a particular stat is attempting to measure when determining it's value. If you look at offensive win shares and assume it means who's a better scorer, of course you're going to disagree with it's assumption that Chandler is a better scorer than Smith. Instead, what it's actually measuring is how much Chandler's offensive total offensive contributions (the rate at which he scores when attempting shots, the second chances he creates threw offensive rebounds and tap outs, the way he impacts teammates shots by creating space with screens, etc) contribute to his team's wins. Under that premise, if you use the eye-test you'd agree that he's a more efficient offensive player (not a better scorer, more efficient offensive player - there's a huge difference) than his teammate J.R. Smith who is known to take ill-advised shots relatively frequently, thus giving away offensive possessions.

So yeah, if you never watched a basketball game before and assumed that offensive win shares solely measured who was better at scoring you'd assume that Chandler was a better scorer than J.R. Smith. Then, if you actually watched a game you'd pretty quickly see that Smith is the better scorer of the two and believe the stat was bogus. In that instances though, the issue isn't with the stat but rather your perception of the stat. So overall, yes it's best to use your observations and stats in unison and realize that each have their limitations. If you watch Knicks games regularly you know that Chandler does a lot of little things to help keep the offense running smoothly, and if you look at offensive win shares with an open mind and understand the stat then you can see exactly to what extent Chandler's contributions positively impact his team's offense.

3RDASYSTEM
08-10-2013, 06:19 PM
I don't understand how people can't use both the eye test and advance stats to evaluate a player. Both seem flawed in its own way but paints a bigger picture when both are used

Eye test from a non player or player? majority of pct is non players coming up this type of flawed science(like in todays world of flawed science)

im speaking on using the eye test and playing in game situations, I've played in good and avg leagues where I had to drop 18ppg(stacked squad for that league) or 28ppg in other league, I had it way way easier avg 18ppg since I was like wide open and would hit like 4-6 threes per game so I think the eye test is vastly superior

but im just speaking for me and my little circle

JeremiahWing
08-10-2013, 06:25 PM
Baseball events don't occur in a vacuum either. A pitcher's effectiveness is tied to the range of his defenders, a hitter is more effective at driving in runs if men are frequently on base in front of him, pitchers attack hitters differently with men on base then with the bases empty, etc. It's as much of a team game as basketball it's just a different game; and advanced stats are just as applicable in either game.



Just to expand on this point for a second. I think we can all agree that basketball is a sport of efficiency. More than any other sport everything has a time limit on it. There's time limits on how long it takes you to in bound the ball, to bring it to the front court, to get off a shot, to how long you can camp out in the lane, everything. There's also a fixed amount of time for a game, unlike baseball where there is no clock. When looking at the sport through the lens of efficiency you can see the value of a player like Chandler who makes the most of his opportunities, doesn't do things that would hurt his team's offensive efficiency because he's trying to play beyond his offensive capabilities, and does little things that add extra possessions that afford his team more opportunities to score (as Guppyfighter noted above). Conversely, a player like J.R. Smith is going to score more points than Chandler because he takes more shots, but he's not more efficient because he frequently takes ill-advised and/or contested shots. In many ways, Chandler balances out Smith's inefficiency by cleaning up his misses and at times providing Smith with a second opportunity to score.

It's important to understand what exactly a particular stat is attempting to measure when determining it's value. If you look at offensive win shares and assume it means who's a better scorer, of course you're going to disagree with it's assumption that Chandler is a better scorer than Smith. Instead, what it's actually measuring is how much Chandler's offensive total offensive contributions (the rate at which he scores when attempting shots, the second chances he creates threw offensive rebounds and tap outs, the way he impacts teammates shots by creating space with screens, etc) contribute to his team's wins. Under that premise, if you use the eye-test you'd agree that he's a more efficient offensive player (not a better scorer, more efficient offensive player - there's a huge difference) than his teammate J.R. Smith who is known to take ill-advised shots relatively frequently, thus giving away offensive possessions.

So yeah, if you never watched a basketball game before and assumed that offensive win shares solely measured who was better at scoring you'd assume that Chandler was a better scorer than J.R. Smith. Then, if you actually watched a game you'd pretty quickly see that Smith is the better scorer of the two and believe the stat was bogus. In that instances though, the issue isn't with the stat but rather your perception of the stat. So overall, yes it's best to use your observations and stats in unison and realize that each have their limitations. If you watch Knicks games regularly you know that Chandler does a lot of little things to help keep the offense running smoothly, and if you look at offensive win shares with an open mind and understand the stat then you can see exactly to what extent Chandler's contributions positively impact his team's offense.

This reply is a COMPLETE fail, as advanced statistics in baseball have been able to isolate a pitcher's performance without regard for the defense (ERA is a very weak stat, and no one really pays it much regard anymore), and the same can be said for offensive players, as NO ONE with any credibility uses RBI has a legit statistic. Advanced metrics in baseball have done an amazing job isolating individual control in regards to performance in a way the NBA could never do, due to the nature of the game.

Try again.

Chronz
08-10-2013, 06:40 PM
The eyes never lie
.... http://ct.fra.bz/ol/fz/sw/i56/5/4/22/frabz-DAFUQ-e04e7c.jpg

Clippersfan86
08-10-2013, 09:06 PM
Big talk coming from someone who thinks Chris Paul let his team down in the playoffs.

Quote me lol. Never have once said CP3 let us down come playoff time, at least not this year. Chris Paul was the ONLY one who stepped up in the playoffs. Anyways continue on with the rant, it's fun.

bagwell368
08-10-2013, 09:23 PM
Your the guy who said Melo was the 6th best offensive player on the Knicks.

No such thing. I said he was 6th in offensive efficiency. If you understood what I said maybe you wouldn't be attempting to defame me with this rubbish.

I have argued that Melo should shoot less and pass more. His team w/o him included is .518 FG vs Melo's .502. Melo holding back on his worst 4 shots of any given game is liable to result in better efficiency/scoring for his team. The fact that he will throw more passes may keep the D off of him a bit more, which could raise his. Not taking his worst 4 shots (not missed shots - worst shots) would likely do the same. His misses are turnovers when not rebounded, and his team is unlikely to hustle as hard with him holding the ball and shooting the ball so much.

As far as your prior argument that the team can't play without him, they played 37.1% of the season w/o him on the floor. Do you propose they only shot .300 when he wasn't in the game?


You are the problem, not the solution. You are the only taking these stats to a level they are not even close to being ready to be at yet.

Big pronouncements regarding things one doesn't understand well have an unfortunate side effect on that persons reputation I find.

bagwell368
08-10-2013, 09:44 PM
advanced stats were created by john hollinger to discredit kobe and credit lebron

Have any proof for that?

C-Wick925
08-10-2013, 10:01 PM
Uhh when you share the win duh

JeremiahWing
08-10-2013, 10:12 PM
Can't even stand up and make your case, just slink away with a fey attempt at a passive aggressive blow it off line...

Pathetic.

Dude I've made my case, and I'll talk to other calm and respectful posters about the topic. You're a little too angry and bitter for me.

JeremiahWing
08-10-2013, 10:18 PM
I've reported your many personal attacks, and will now be on my ignore list. God speed.

IndyRealist
08-10-2013, 10:20 PM
Interesting analogy given there, considering that tracking weather formations is well understood to have it's flaws. It's pretty much impossible to predict the weather correctly, which is why so many people think the weather man on TV is wrong.

The main difference here is basketball is much more tied to things outside of one player. If you wanted to make a comparison to weather formation tracking methods then maybe advanced team stats would be more accurate. There's also the fact that not all basic stats are created equal, and those basic stats is what these formulas draw from.

I chose that comparison specifically because it's not always accurate. There are outliers, that does not disprove the math. It's much better than a guy looking up at the clouds and saying, "yup, it's gonna rain."

The fact that all basic stats are not equal is EXACTLY why we need advanced stats. Just about the most useless basic stat is points per game, yet that is 90% of what people use to judge a player. It's been shown that PPG is the predominate factor in ROY voting, MVP voting, all-star voting, 6th man voting, etc. And aside from all-star, that voting is done by supposed experts who should know better. Advanced metrics exist because people don't know how to properly value the basic ones.

Take PER for instance. In it's most basic form, it takes each box score stat and applied a weight to it, making a point more valuable than an assist, for instance. Box score metrics differ in how much weight they give each stat, which is where the you get the outliers. PER rates Brook Lopez over Tim Duncan, James Harden, Kobe Bryant, etc. and rates Amar'e Stoudamire as the 13th best player last year, while rating Klay Thompson 214th. Using a stat means understanding it's limitations. Just like predicting the weather.

+/- based stats are a whole different animal, and going into that would triple the size of this post.

IndyRealist
08-10-2013, 10:43 PM
Individual players aren't solely responsible for their numbers in basketball. Unlike baseball players who hit and field by themselves, basketball players rely on good passes and screens to get open looks, good help defense, good transition play, etc, etc, etc. Again, the idea that all of the circumstances of individual can be quantified is a stupid idea on its face.

Then why, when is a player traded or moves to another team, does a player's stats generally stay close to the same? A solid percentage of PSDers thought that James Harden would fail miserably as the #1 guy in Houston, without Durant and Westbrook getting him open. Not only did he not fall on his face, but he thrived despite having a hodgepodge team trying to get used to him.

The idea that baseball is an individual sport is a misnomer. The hitter hits solo against the defense, yes. The team's offense is greatly dependent on other players getting on base, teammate's ability to steal bases, where he hits in the lineup, etc. The defense is not just the pitcher throwing the ball. It's the ability of the catcher to read the hitter and opponents on base, the ability of the outfielders to catch fly balls, the ability of the basemen to cover their assignments, etc. Baseball is not individual in the slightest.

Conversely, of the major American sports basketball players have the LEAST variation in stats from year to year, from changes in teammates, from changes in coaches and systems, etc. (adjusted for aging curve). That suggests that the majority of a player's basic box score, that player is mostly responsible for. The difference between baseball and basketball, is that in baseball it's been SHOWN that a team had success following a stats based system. In basketball, teams are guarding their stat systems like Fort Knox. No one knows how much credit to give to stats, because no team wants to advertise the fact and lose their advantage.

WadeKobe
08-10-2013, 11:05 PM
Then why, when is a player traded or moves to another team, does a player's stats generally stay close to the same? A solid percentage of PSDers thought that James Harden would fail miserably as the #1 guy in Houston, without Durant and Westbrook getting him open. Not only did he not fall on his face, but he thrived despite having a hodgepodge team trying to get used to him.

The idea that baseball is an individual sport is a misnomer. The hitter hits solo against the defense, yes. The team's offense is greatly dependent on other players getting on base, teammate's ability to steal bases, where he hits in the lineup, etc. The defense is not just the pitcher throwing the ball. It's the ability of the catcher to read the hitter and opponents on base, the ability of the outfielders to catch fly balls, the ability of the basemen to cover their assignments, etc. Baseball is not individual in the slightest.

Conversely, of the major American sports basketball players have the LEAST variation in stats from year to year, from changes in teammates, from changes in coaches and systems, etc. (adjusted for aging curve). That suggests that the majority of a player's basic box score, that player is mostly responsible for. The difference between baseball and basketball, is that in baseball it's been SHOWN that a team had success following a stats based system. In basketball, teams are guarding their stat systems like Fort Knox. No one knows how much credit to give to stats, because no team wants to advertise the fact and lose their advantage.

Thank you for bringing this up. I am so tired of the worn out "system", "teammates", etc. it simply doesn't match the data.

Pablonovi
08-10-2013, 11:13 PM
I don't understand how people can't use both the eye test and advance stats to evaluate a player. Both seem flawed in its own way but paints a bigger picture when both are used

Yeah the Eye Test sure MUST be flawed; afterall, no two witnesses to anything ever see it exactly the same way; much less most "homer-istic" fans. And, at least for any stat I've seen to date, it is NOT COMPLETE ENOUGH in itself, thus IS partially flawed at best. BUT the two together (as un-biased an Eye Test as one can manage PLUS the best Offensive Stat PLUS the best Defensive Stat; or some such combo) should be better than either separately.

I grew up / was raised in a kind of "anti-bias" atmosphere generally; particular vis-à-vis the NBA and its teams and their players. And yet, watching any game "completely un-biased" is a real challenge (if not impossible); because other personally-idiocynatic biases exist. For me, for example, I just love numbers; so in any game, if any player starts racking up good numbers at any stat, I tend to root for them to keep racking them up higher). And that will tend to bias my overall take on the game.

P.S. I've recently seen some sharp internet analyzers attempt to "improve" the PER by removing the (inadequate) defensive component(s) and claiming that this reworked stat is quite useful as a measure of Offensive performance.

jerellh528
08-10-2013, 11:37 PM
Dont' worry about it. It's not used to determine a players greatness. Just another useless stat created by people who do nothing else but think of new stats to create.

b@llhog24
08-10-2013, 11:51 PM
Dont' worry about it. It's not used to determine a players greatness. Just another useless stat created by people who do nothing else but think of new stats to create.

Did you think this was witty or something? It wasn't.

bagwell368
08-11-2013, 06:10 AM
Dont' worry about it. It's not used to determine a players greatness. Just another useless stat created by people who do nothing else but think of new stats to create.

WS has flaws, some of which I've pointed out before, but if you look at consensus subjective all time great player lists side by side with Win Shares totals and Win Shares / 48 - the correlation is fairly high.

I live on the East Coast, and my memory only goes back to the mid 60's for the NBA, so do tell us if I wanted to discuss/compare players from before 1965 or players on less popular West Coast teams I might see for a handful of minutes per year, I should ONLY use the eye test or ESPN highlights and TNT talking heads to make up my mind about players? Is that what you propose?

skierdude44
08-11-2013, 08:26 AM
This reply is a COMPLETE fail, as advanced statistics in baseball have been able to isolate a pitcher's performance without regard for the defense (ERA is a very weak stat, and no one really pays it much regard anymore), and the same can be said for offensive players, as NO ONE with any credibility uses RBI has a legit statistic. Advanced metrics in baseball have done an amazing job isolating individual control in regards to performance in a way the NBA could never do, due to the nature of the game.

Try again.


Then why, when is a player traded or moves to another team, does a player's stats generally stay close to the same? A solid percentage of PSDers thought that James Harden would fail miserably as the #1 guy in Houston, without Durant and Westbrook getting him open. Not only did he not fall on his face, but he thrived despite having a hodgepodge team trying to get used to him.

The idea that baseball is an individual sport is a misnomer. The hitter hits solo against the defense, yes. The team's offense is greatly dependent on other players getting on base, teammate's ability to steal bases, where he hits in the lineup, etc. The defense is not just the pitcher throwing the ball. It's the ability of the catcher to read the hitter and opponents on base, the ability of the outfielders to catch fly balls, the ability of the basemen to cover their assignments, etc. Baseball is not individual in the slightest.

Conversely, of the major American sports basketball players have the LEAST variation in stats from year to year, from changes in teammates, from changes in coaches and systems, etc. (adjusted for aging curve). That suggests that the majority of a player's basic box score, that player is mostly responsible for. The difference between baseball and basketball, is that in baseball it's been SHOWN that a team had success following a stats based system. In basketball, teams are guarding their stat systems like Fort Knox. No one knows how much credit to give to stats, because no team wants to advertise the fact and lose their advantage.

Indy's reply is exactly the point I was making. Baseball is just as much of a team sport as any other team sport. When talking about stats, it's important to realize what each stat is attempting to measure and use it in the proper context. You brought up ERA in baseball for example, which is useful only in the context of what it proposes to measure, the average earned runs a pitcher gives up per 9 innings. Because baseball is a team game, it is affected by the range, arm strength, and other abilities of the defenders playing behind him. So if a pitcher has an ERA of 4.00 we know that on average over 9 innings pitching with his team's normal defensive alignment behind him the pitcher will give up 4 earned runs. That's all it says however, and it's not a measure of his true talent because if he pitched on a better or worse defensive team his ERA would be lower or higher. FIP (fielding independent pitching) only considers things a pitcher can control (such as walks, strikeouts, and home runs) therefore a pitcher's FIP would remain constant regardless of the quality of the defense behind him. FIP is a much better true talent indicator as it isolates only what is directly in the pitcher's control so it's a stat that would be used when evaluating whether to acquire a certain pitcher.

The point is, regardless of the sport, different stats measure different things regardless of how simple or complex the stat is. It would be inane and misguided to look at one and only one stat to form a judgment on a player's total worth because that stat isn't trying to measure his entire worth in one metric, just one aspect of his game. The idea that basketball is some tightly interwoven game where it is impossible to pick apart individual contributions and their value via stats is misguided. In fact, if that were true than you also wouldn't be able to pick apart individual contributions and evaluate individual talent via the eye test either because how would you know who was more responsible for the outcome. Case in point, the ball handler comes across a screen and shoots a jumper, the shot goes in... Did he make the shot because he's a good shooter, or because the screen got him open and aided in his shot making ability? What would your eye test tell you about that scenario? Probably not much, but over time you could chart how accurate a jump shooter was shooting from the same spot on the floor with and without the screen and compare the percentages to figure out what effect the screen has on the outcome. Furthermore, you could chart the percentages with the same shooter coming off screens from different teammates and see which teammate was the best screener and by approximately how much. There's no way you could do that solely with the eye test alone as theirs no way you could keep track of every shot and every screener without some math involved.

Clippersfan86
08-11-2013, 10:40 AM
Lalalalalalalalala means I love youuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuu.

Hustla23
08-11-2013, 11:48 AM
Indy's reply is exactly the point I was making. Baseball is just as much of a team sport as any other team sport. When talking about stats, it's important to realize what each stat is attempting to measure and use it in the proper context. You brought up ERA in baseball for example, which is useful only in the context of what it proposes to measure, the average earned runs a pitcher gives up per 9 innings. Because baseball is a team game, it is affected by the range, arm strength, and other abilities of the defenders playing behind him. So if a pitcher has an ERA of 4.00 we know that on average over 9 innings pitching with his team's normal defensive alignment behind him the pitcher will give up 4 earned runs. That's all it says however, and it's not a measure of his true talent because if he pitched on a better or worse defensive team his ERA would be lower or higher. FIP (fielding independent pitching) only considers things a pitcher can control (such as walks, strikeouts, and home runs) therefore a pitcher's FIP would remain constant regardless of the quality of the defense behind him. FIP is a much better true talent indicator as it isolates only what is directly in the pitcher's control so it's a stat that would be used when evaluating whether to acquire a certain pitcher.

The point is, regardless of the sport, different stats measure different things regardless of how simple or complex the stat is. It would be inane and misguided to look at one and only one stat to form a judgment on a player's total worth because that stat isn't trying to measure his entire worth in one metric, just one aspect of his game. The idea that basketball is some tightly interwoven game where it is impossible to pick apart individual contributions and their value via stats is misguided. In fact, if that were true than you also wouldn't be able to pick apart individual contributions and evaluate individual talent via the eye test either because how would you know who was more responsible for the outcome. Case in point, the ball handler comes across a screen and shoots a jumper, the shot goes in... Did he make the shot because he's a good shooter, or because the screen got him open and aided in his shot making ability? What would your eye test tell you about that scenario? Probably not much, but over time you could chart how accurate a jump shooter was shooting from the same spot on the floor with and without the screen and compare the percentages to figure out what effect the screen has on the outcome. Furthermore, you could chart the percentages with the same shooter coming off screens from different teammates and see which teammate was the best screener and by approximately how much. There's no way you could do that solely with the eye test alone as theirs no way you could keep track of every shot and every screener without some math involved.

Everyone in here needs to read this post. Spot ****ing on.

jerellh528
08-11-2013, 05:20 PM
Did you think this was witty or something? It wasn't.

Nahhhh, it's a basic response. Although you must think it's somewhat witty to even bring up that notion.

lol, please
08-11-2013, 05:24 PM
I'll admit that while I don't care for advance stats, I won't ignore them either. I read a post where some one was talking about Beno Udrih's "win shares" and it peeked my enterest. So here are my two request/questions....

1. Explain win shares to me( not just post the definition of it). I want to know if ppl that use this to back their stance actually KNOW what it means or are they just looking on the web for " Top win shares in NBA"


2. This one may answer itself in the process but--------
Wouldnt players on better teams automatically have higher win shares than players on poorer teams?

Example( help me if Im wrong)....Wouldn't Lebron James ( current stats) have less win shares if he was on a 20 win Orl Magic team than ( lets say) Luol Deng( current stats) on a 60 win Chi Bulls team?<-----------------hypothetical
Explain winshares to me, because my reading comprehension is sub par, thus I am inept at doing research on my own, and therefore I need you to break this down for me like a child.

That's what I got from the OP.

Guppyfighter
08-11-2013, 08:42 PM
Explain winshares to me, because my reading comprehension is sub par, thus I am inept at doing research on my own, and therefore I need to you break this down for me like a child.

That's what I got from the OP.

Not as bad as scoots opinion on winshares.

3RDASYSTEM
08-11-2013, 09:23 PM
Everyone in here needs to read this post. Spot ****ing on.

or maybe you should read all mine

I have been saying this and now all you see are the 'eye test'

but I said you would have to also play the game at some level, especially HS and college, to higher

like I understand if a player sets a screen really well or gets away with something, that comes with watching the game and playing a lot adds on to you know how of the situation and what to watch for because its all pretty much the same sets run over and over, now who can execute the best is what it is all about

the game started with AUERBACH and others having the best players via eye test and it will always be and end like that but I respect the science of something that has to be so confusing for most, its a reflection of our society

IKnowHoops
08-12-2013, 01:15 AM
or maybe you should read all mine

I have been saying this and now all you see are the 'eye test'

but I said you would have to also play the game at some level, especially HS and college, to higher

like I understand if a player sets a screen really well or gets away with something, that comes with watching the game and playing a lot adds on to you know how of the situation and what to watch for because its all pretty much the same sets run over and over, now who can execute the best is what it is all about

the game started with AUERBACH and others having the best players via eye test and it will always be and end like that but I respect the science of something that has to be so confusing for most, its a reflection of our society

lol

lol, please
08-12-2013, 01:29 AM
Not as bad as scoots opinion on winshares.

:laugh:

RollingWave
08-12-2013, 01:55 AM
I'll admit that while I don't care for advance stats, I won't ignore them either. I read a post where some one was talking about Beno Udrih's "win shares" and it peeked my enterest. So here are my two request/questions....

1. Explain win shares to me( not just post the definition of it). I want to know if ppl that use this to back their stance actually KNOW what it means or are they just looking on the web for " Top win shares in NBA"


2. This one may answer itself in the process but--------
Wouldnt players on better teams automatically have higher win shares than players on poorer teams?

Example( help me if Im wrong)....Wouldn't Lebron James ( current stats) have less win shares if he was on a 20 win Orl Magic team than ( lets say) Luol Deng( current stats) on a 60 win Chi Bulls team?<-----------------hypothetical

The actual calculation is pretty complicated, but in short, it can be broken up into 2 parts, offense and defense.

offensively, the calculation is a bit more reliable, it's essentially your efficiency x your usage x your secondary value. if a player uses the ball around the same, then the higher the efficiency the better. which makes sense obviously.

if the players are of the same efficiency, then it may vary a bit, if that efficiency isn't very good, then the lower the usage the better (because possessions going to other guys probably end up with better results .) if that efficiency is above average, then the higher the usage the better.


Defensively, it's more complicated, and generally less reliable, because the value is 80% tied to the overall team defensive results, the logic kinda make sense in that your 1/5 of the guy standing on the court, but in practice the problem is that different positions effect defense in very disproportional ways, obviously for one thing, Center's defense usually make or break most team. where as a good defensive PG isn't going to be the difference between a good or bad defensive team.

The actual calculation also relies heavily on counting stats like rebounds and blocks / steals. which is not very ideal.


But overall, it is an interesting stat, it is useful as long as you are aware of the caveat, and tend to be fairly good indicator as long as roles are the same. and it is actually pretty stable in practice, for example, check out Lebron's winshare in his last 3 years in Cleveland to these 3 years in Miami. (in per 48 terms.)

(From this year back)

.322
.298
.244
.299
.318
.242

So you see, not only was it all within a pretty similar range, but the difference of changing teammates haven't really influenced it much at all. but this is mostly because Lebron's the lead guy, it's HE that influence others, not the other way around.



But as to your questions: yes players on better teams tend to have better winshare, though in reality that isn't a totally wrong thing either given that.... well... if your winning.. of course your doing better...

But obviously the key here is the role players, what winshare tend to do is overrate decent role players on great teams. for example, these were the role players who were significantly above average in ws/48 last year on the heat (average is .100)

Birdman
Miller
Chalmers
Battier
Allen

Now, some of these are not like the others, Birdman has always been extremely good in the same role, and Allen was a star very recently, but MIllier / Battier / Chalmers are good example of guys who's value is really dependent on the team, but one should note that this is only true if those are DECENT role players, for example, Norris Cole's WS/48 is garbage.


Generally from observing winshare versus actual game, my conclusion of types of player winshare tend to over / underrate.

Overrate : role players on great teams (example, Mario Chalmers) / solid reserves in general (example, Chris Copeland ) / offensive players with questionable defensive games. (example, Ryan Anderson.)

Underrate : good but not great players on bad teams. (example, Jrue Holiday) / average starters on none great teams (too many to name, generally this is in comparason to solid reserves ) / great defensive players without much of an offensive game. (Example, Omer Asik)


It has plenty of flaws, but it does catches a lot of interesting things, and the weakness are not difficult to understand. it's not the end all stat, but it is a good starting point, then adjust for other factors like team / role etc.

Chronz
08-12-2013, 10:56 AM
So D-Lee got suspended. Cant say Im surprised with the rampage he was attempting to go on the last few days.



Dont' worry about it. It's not used to determine a players greatness. Just another useless stat created by people who do nothing else but think of new stats to create.

You mean the kind of guys who work in the industry you love simply watching ..... yea sucks to be them. Who wouldn't want to work and better understand the field you devote your free time watching.

Pablonovi
08-12-2013, 10:58 AM
The actual calculation is pretty complicated, but in short, it can be broken up into 2 parts, offense and defense.

offensively, the calculation is a bit more reliable, it's essentially your efficiency x your usage x your secondary value. if a player uses the ball around the same, then the higher the efficiency the better. which makes sense obviously.

if the players are of the same efficiency, then it may vary a bit, if that efficiency isn't very good, then the lower the usage the better (because possessions going to other guys probably end up with better results .) if that efficiency is above average, then the higher the usage the better.


Defensively, it's more complicated, and generally less reliable, because the value is 80% tied to the overall team defensive results, the logic kinda make sense in that your 1/5 of the guy standing on the court, but in practice the problem is that different positions effect defense in very disproportional ways, obviously for one thing, Center's defense usually make or break most team. where as a good defensive PG isn't going to be the difference between a good or bad defensive team.

The actual calculation also relies heavily on counting stats like rebounds and blocks / steals. which is not very ideal.


But overall, it is an interesting stat, it is useful as long as you are aware of the caveat, and tend to be fairly good indicator as long as roles are the same. and it is actually pretty stable in practice, for example, check out Lebron's winshare in his last 3 years in Cleveland to these 3 years in Miami. (in per 48 terms.)

(From this year back)

.322
.298
.244
.299
.318
.242

So you see, not only was it all within a pretty similar range, but the difference of changing teammates haven't really influenced it much at all. but this is mostly because Lebron's the lead guy, it's HE that influence others, not the other way around.



But as to your questions: yes players on better teams tend to have better winshare, though in reality that isn't a totally wrong thing either given that.... well... if your winning.. of course your doing better...

But obviously the key here is the role players, what winshare tend to do is overrate decent role players on great teams. for example, these were the role players who were significantly above average in ws/48 last year on the heat (average is .100)

Birdman
Miller
Chalmers
Battier
Allen

Now, some of these are not like the others, Birdman has always been extremely good in the same role, and Allen was a star very recently, but MIllier / Battier / Chalmers are good example of guys who's value is really dependent on the team, but one should note that this is only true if those are DECENT role players, for example, Norris Cole's WS/48 is garbage.


Generally from observing winshare versus actual game, my conclusion of types of player winshare tend to over / underrate.

Overrate : role players on great teams (example, Mario Chalmers) / solid reserves in general (example, Chris Copeland ) / offensive players with questionable defensive games. (example, Ryan Anderson.)

Underrate : good but not great players on bad teams. (example, Jrue Holiday) / average starters on none great teams (too many to name, generally this is in comparason to solid reserves ) / great defensive players without much of an offensive game. (Example, Omer Asik)


It has plenty of flaws, but it does catches a lot of interesting things, and the weakness are not difficult to understand. it's not the end all stat, but it is a good starting point, then adjust for other factors like team / role etc.

Hey Rolling Wave,
Thanx for this. I'd say that it's clearly the best answer to the question raised in the OP; and it only took us 151 posts to do this! HEHE. I agree with your analysis completely.
Again, thanx,
Pablo

WadeKobe
08-12-2013, 11:09 AM
The actual calculation is pretty complicated, but in short, it can be broken up into 2 parts, offense and defense.

offensively, the calculation is a bit more reliable, it's essentially your efficiency x your usage x your secondary value. if a player uses the ball around the same, then the higher the efficiency the better. which makes sense obviously.

if the players are of the same efficiency, then it may vary a bit, if that efficiency isn't very good, then the lower the usage the better (because possessions going to other guys probably end up with better results .) if that efficiency is above average, then the higher the usage the better.


Defensively, it's more complicated, and generally less reliable, because the value is 80% tied to the overall team defensive results, the logic kinda make sense in that your 1/5 of the guy standing on the court, but in practice the problem is that different positions effect defense in very disproportional ways, obviously for one thing, Center's defense usually make or break most team. where as a good defensive PG isn't going to be the difference between a good or bad defensive team.

The actual calculation also relies heavily on counting stats like rebounds and blocks / steals. which is not very ideal.


But overall, it is an interesting stat, it is useful as long as you are aware of the caveat, and tend to be fairly good indicator as long as roles are the same. and it is actually pretty stable in practice, for example, check out Lebron's winshare in his last 3 years in Cleveland to these 3 years in Miami. (in per 48 terms.)

(From this year back)

.322
.298
.244
.299
.318
.242

So you see, not only was it all within a pretty similar range, but the difference of changing teammates haven't really influenced it much at all. but this is mostly because Lebron's the lead guy, it's HE that influence others, not the other way around.



But as to your questions: yes players on better teams tend to have better winshare, though in reality that isn't a totally wrong thing either given that.... well... if your winning.. of course your doing better...

But obviously the key here is the role players, what winshare tend to do is overrate decent role players on great teams. for example, these were the role players who were significantly above average in ws/48 last year on the heat (average is .100)

Birdman
Miller
Chalmers
Battier
Allen

Now, some of these are not like the others, Birdman has always been extremely good in the same role, and Allen was a star very recently, but MIllier / Battier / Chalmers are good example of guys who's value is really dependent on the team, but one should note that this is only true if those are DECENT role players, for example, Norris Cole's WS/48 is garbage.


Generally from observing winshare versus actual game, my conclusion of types of player winshare tend to over / underrate.

Overrate : role players on great teams (example, Mario Chalmers) / solid reserves in general (example, Chris Copeland ) / offensive players with questionable defensive games. (example, Ryan Anderson.)

Underrate : good but not great players on bad teams. (example, Jrue Holiday) / average starters on none great teams (too many to name, generally this is in comparason to solid reserves ) / great defensive players without much of an offensive game. (Example, Omer Asik)


It has plenty of flaws, but it does catches a lot of interesting things, and the weakness are not difficult to understand. it's not the end all stat, but it is a good starting point, then adjust for other factors like team / role etc.

Norris Cole's WS48 is awful because... Well... Norris Cole is awful. And EVERY stat agrees.

RAPM
WS
PER
WP
ezPM

Cole is terrible.

elledaddy
08-12-2013, 11:26 AM
The actual calculation is pretty complicated, but in short, it can be broken up into 2 parts, offense and defense.

offensively, the calculation is a bit more reliable, it's essentially your efficiency x your usage x your secondary value. if a player uses the ball around the same, then the higher the efficiency the better. which makes sense obviously.

if the players are of the same efficiency, then it may vary a bit, if that efficiency isn't very good, then the lower the usage the better (because possessions going to other guys probably end up with better results .) if that efficiency is above average, then the higher the usage the better.


Defensively, it's more complicated, and generally less reliable, because the value is 80% tied to the overall team defensive results, the logic kinda make sense in that your 1/5 of the guy standing on the court, but in practice the problem is that different positions effect defense in very disproportional ways, obviously for one thing, Center's defense usually make or break most team. where as a good defensive PG isn't going to be the difference between a good or bad defensive team.

The actual calculation also relies heavily on counting stats like rebounds and blocks / steals. which is not very ideal.


But overall, it is an interesting stat, it is useful as long as you are aware of the caveat, and tend to be fairly good indicator as long as roles are the same. and it is actually pretty stable in practice, for example, check out Lebron's winshare in his last 3 years in Cleveland to these 3 years in Miami. (in per 48 terms.)

(From this year back)

.322
.298
.244
.299
.318
.242

So you see, not only was it all within a pretty similar range, but the difference of changing teammates haven't really influenced it much at all. but this is mostly because Lebron's the lead guy, it's HE that influence others, not the other way around.



But as to your questions: yes players on better teams tend to have better winshare, though in reality that isn't a totally wrong thing either given that.... well... if your winning.. of course your doing better...

But obviously the key here is the role players, what winshare tend to do is overrate decent role players on great teams. for example, these were the role players who were significantly above average in ws/48 last year on the heat (average is .100)

Birdman
Miller
Chalmers
Battier
Allen

Now, some of these are not like the others, Birdman has always been extremely good in the same role, and Allen was a star very recently, but MIllier / Battier / Chalmers are good example of guys who's value is really dependent on the team, but one should note that this is only true if those are DECENT role players, for example, Norris Cole's WS/48 is garbage.


Generally from observing winshare versus actual game, my conclusion of types of player winshare tend to over / underrate.

Overrate : role players on great teams (example, Mario Chalmers) / solid reserves in general (example, Chris Copeland ) / offensive players with questionable defensive games. (example, Ryan Anderson.)

Underrate : good but not great players on bad teams. (example, Jrue Holiday) / average starters on none great teams (too many to name, generally this is in comparason to solid reserves ) / great defensive players without much of an offensive game. (Example, Omer Asik)


It has plenty of flaws, but it does catches a lot of interesting things, and the weakness are not difficult to understand. it's not the end all stat, but it is a good starting point, then adjust for other factors like team / role etc.


Much Respect Sir. I made the original post and this was an excellent reply.

3RDASYSTEM
08-12-2013, 11:43 AM
The actual calculation is pretty complicated, but in short, it can be broken up into 2 parts, offense and defense.

offensively, the calculation is a bit more reliable, it's essentially your efficiency x your usage x your secondary value. if a player uses the ball around the same, then the higher the efficiency the better. which makes sense obviously.

if the players are of the same efficiency, then it may vary a bit, if that efficiency isn't very good, then the lower the usage the better (because possessions going to other guys probably end up with better results .) if that efficiency is above average, then the higher the usage the better.


Defensively, it's more complicated, and generally less reliable, because the value is 80% tied to the overall team defensive results, the logic kinda make sense in that your 1/5 of the guy standing on the court, but in practice the problem is that different positions effect defense in very disproportional ways, obviously for one thing, Center's defense usually make or break most team. where as a good defensive PG isn't going to be the difference between a good or bad defensive team.

The actual calculation also relies heavily on counting stats like rebounds and blocks / steals. which is not very ideal.


But overall, it is an interesting stat, it is useful as long as you are aware of the caveat, and tend to be fairly good indicator as long as roles are the same. and it is actually pretty stable in practice, for example, check out Lebron's winshare in his last 3 years in Cleveland to these 3 years in Miami. (in per 48 terms.)

(From this year back)

.322
.298
.244
.299
.318
.242

So you see, not only was it all within a pretty similar range, but the difference of changing teammates haven't really influenced it much at all. but this is mostly because Lebron's the lead guy, it's HE that influence others, not the other way around.



But as to your questions: yes players on better teams tend to have better winshare, though in reality that isn't a totally wrong thing either given that.... well... if your winning.. of course your doing better...

But obviously the key here is the role players, what winshare tend to do is overrate decent role players on great teams. for example, these were the role players who were significantly above average in ws/48 last year on the heat (average is .100)

Birdman
Miller
Chalmers
Battier
Allen

Now, some of these are not like the others, Birdman has always been extremely good in the same role, and Allen was a star very recently, but MIllier / Battier / Chalmers are good example of guys who's value is really dependent on the team, but one should note that this is only true if those are DECENT role players, for example, Norris Cole's WS/48 is garbage.


Generally from observing winshare versus actual game, my conclusion of types of player winshare tend to over / underrate.

Overrate : role players on great teams (example, Mario Chalmers) / solid reserves in general (example, Chris Copeland ) / offensive players with questionable defensive games. (example, Ryan Anderson.)

Underrate : good but not great players on bad teams. (example, Jrue Holiday) / average starters on none great teams (too many to name, generally this is in comparason to solid reserves ) / great defensive players without much of an offensive game. (Example, Omer Asik)


It has plenty of flaws, but it does catches a lot of interesting things, and the weakness are not difficult to understand. it's not the end all stat, but it is a good starting point, then adjust for other factors like team / role etc.

damn I must toot my own horn since I don't need none of those BRON numbers because its basically what I said seeing he has 2 mvps in 3yrs with HEAT and 2 in 7yrs with CAVS, same player just diff. allstars rolling with him, he's still the same stat filler/mvp

how is that so hard to see or figure out? well nevermind playing on some level at a high level really helps

I have been breaking this type of **** down since I told CHRONZ and other junior high level players that IVERSON was suffering from being efficient because of playing with two 3ppg scorers as his core and they said I was crazy and didn't know ****, and im thinking I have played with both allstar teams and ****** 2 scorer teams, I was ****** efficient(still ok) with me being 1 of the 2 scorer team(we lost in playoffs also) but we won it all with the allstar team and I shot no worse than 60pct from 3pt land, now how would it be any different on higher levels of play?

different level same ****

Pablonovi
08-12-2013, 11:58 AM
damn I must toot my own horn since I don't need none of those BRON numbers because its basically what I said seeing he has 2 mvps in 3yrs with HEAT and 2 in 7yrs with CAVS, same player just diff. allstars rolling with him, he's still the same stat filler/mvp

how is that so hard to see or figure out? well nevermind playing on some level at a high level really helps

I have been breaking this type of **** down since I told CHRONZ and other junior high level players that IVERSON was suffering from being efficient because of playing with two 3ppg scorers as his core and they said I was crazy and didn't know ****, and im thinking I have played with both allstar teams and ****** 2 scorer teams, I was ****** efficient(still ok) with me being 1 of the 2 scorer team(we lost in playoffs also) but we won it all with the allstar team and I shot no worse than 60pct from 3pt land, now how would it be any different on higher levels of play?

different level same ****

Hey 3rda...,
Please try NOT to take offense at what I'm about to say, ok?
Of all the people who post in either the NBA General Forum or the Lakers' Forum (the only two PSD forums I visit and/or post at); your posts are ALWAYS the most difficult for me to understand. Here, at this moment, I am NOT referring to the CONTENT of what you say; but only to the FORM of how you say it.

Perhaps, if you'd take a little more time on (reworking) your posts, you'd vastly improve everybody else's understanding of what you really want to say.

Guppyfighter
08-12-2013, 12:15 PM
I was thinking the same thing. First thing I thought was "I don't know what this guy is saying."

b@llhog24
08-12-2013, 12:21 PM
I was thinking the same thing. First thing I thought was "I don't know what this guy is saying."

Through skimming through his content, it somewhat seems as if he has some level of insight. If only he passed third grade English.

Chronz
08-12-2013, 12:22 PM
damn I must toot my own horn since I don't need none of those BRON numbers because its basically what I said seeing he has 2 mvps in 3yrs with HEAT and 2 in 7yrs with CAVS, same player just diff. allstars rolling with him, he's still the same stat filler/mvp

how is that so hard to see or figure out? well nevermind playing on some level at a high level really helps

I have been breaking this type of **** down since I told CHRONZ and other junior high level players that IVERSON was suffering from being efficient because of playing with two 3ppg scorers as his core and they said I was crazy and didn't know ****, and im thinking I have played with both allstar teams and ****** 2 scorer teams, I was ****** efficient(still ok) with me being 1 of the 2 scorer team(we lost in playoffs also) but we won it all with the allstar team and I shot no worse than 60pct from 3pt land, now how would it be any different on higher levels of play?

different level same ****
nah playa

Chronz
08-12-2013, 12:27 PM
I was thinking the same thing. First thing I thought was "I don't know what this guy is saying."

Game is game from day 1 = Players have no peak/prime, they are what they have proven to be from the onset.

He holds the contradictory opinion that players dont make other players better (it helps him with the Nash/Kidd vs AI argument) but he then holds the opinion that Iverson wasn't as efficient as he could have been had he had more support. In essence, hes arguing that other players would have made AI more efficient but that Nash doesn't do that for his teammates.

He argues that Iverson was still capable of playing star basketball beyond his Memphis days but will then use the age/decline excuse when you downplay his performance in Denver.

Sprinkle in some quotes from coach Chuck Daley slurping up AI with him ignoring the unanimous praise Kobe has gotten over his career from a multitude of coaches/peers and you have his agenda down to a T.

Pablonovi
08-12-2013, 01:24 PM
Game is game from day 1 = Players have no peak/prime, they are what they have proven to be from the onset.

He holds the contradictory opinion that players dont make other players better (it helps him with the Nash/Kidd vs AI argument) but he then holds the opinion that Iverson wasn't as efficient as he could have been had he had more support. In essence, hes arguing that other players would have made AI more efficient but that Nash doesn't do that for his teammates.

He argues that Iverson was still capable of playing star basketball beyond his Memphis days but will then use the age/decline excuse when you downplay his performance in Denver.

Sprinkle in some quotes from coach Chuck Daley slurping up AI with him ignoring the unanimous praise Kobe has gotten over his career from a multitude of coaches/peers and you have his agenda down to a T.

Hey High Horse,
Perhaps I was being a little on the "kind" side, focusing on his "form" rather than his "content".

It is virtually undeniable what (as you correctly point out here) is this guy's agenda: "All Praise To AI; All Condemnation To Kobe"; and he's got like 1% agreement from the rest of PSD NBA.

But still, wouldn't you agree that IF he'd clean up his (non-)English, at least people could better understand his raunch-homeristic arguments?

P.S. Unless he realizes he can't win straight-out; and so, tries to confuse us with his "gobbley-gook"-speak? Still, even so, I don't see anybody qualifying for these two criteria:
1) They understand what he has just said; and
2) They agree with him.

I'm busting a gut, what comic relief!

knicks_99
10-15-2013, 04:40 PM
Came on just to say advanced stats are freakin rubbish.

The eye test never fails unless you're ******** or something.

Stats are really just a microcosm of what's happening on the court, a reflection really. Like looking at your reflection in water - it's murky, unreliable, and could NEVER give an accurate portrait of what a player really "looks" like.

Baseball's different b/c it's a team sport where player production/worth is almost COMPLETELY dependent on the individual rather than team play.

Basketball is completely and utterly dependent on synergy, chemistry, style of play, etc.

It's funny really, b/c all the Jordan stan dummies will point to Win Shares, and yet they don't realize that Jordan was a ballhog whose "stats" were accumulated by essentially holding onto the ball and dictating literally every play, just like Lebron and Chris Paul. ANY player whose talented who has the ball in his hands 75% of the time is going to rack up numbers, and thus lead to high advanced metrics. It's called common sense. If you look at all the top Win Share guys of all time, there's a pattern. It doesn't necessarily lead to wins though. It should probably be renamed "Production-Shares".

Meanwhile, a guy like Bill Russell, who has 5 MVP's and 11 championships, 2 as a player-coach, will be overshadowed b/c advanced metrics SUCK and could NEVER EVER EVER accurately portray things like intangibles or a player's effect on team defense. How about hustle? Or leadership? Or clutchness? Does it take into account loose balls saved, or hockey assists, or a player's effect on team chemistry?

Let's be real, Win Shares and PER are vain, skin deep nerdy reflections of players that favors high offensive production guys. Whether or not that actually leads to WINS is a completely different story.

You know what IS a good advanced metric? WINS. lol

Why? B/c eventually the cream rises to the top. On a court with only 10 players, regardless of situation, EVENTUALLY the cream ALWAYS rises to the top and finds a way to win, by hook or by crook.

D-Leethal
10-15-2013, 05:03 PM
I think there are plenty of ways to use advanced metrics very effectively, but they are normally within your team concept and in the form of best combinations on the floor, highest success rate with certain plays within those combinations etc...

Trying to come up an all encompassing math formula to determine "whos the best" or rank players who are in completely different situations, systems, surrounded by completely different talent levels is stupid on the surface and even more flawed the deeper you dig. When you have to supplement these comparison stats with boatloads of context (which is usually derived from the eye test), whats the point of even using them? Doesn't that defeat the purpose of having the stat in the first place? Every single one of these stats has laughable outliers all over the place when your using them to rank/judge players and none of them have any ability to accurately portray the 5 as 1 team aspect or the effect those teammates around you have on the stats you put up - and that team oriented/synergistic dynamic is basically the essence of the game of basketball - and its COMPLETELY ignored within these all encompassing stats.

D-Leethal
10-15-2013, 05:08 PM
I think people take the new wave of analytics in the NBA the wrong way - nobody is using these stupid win shares stats for anything. They are setting up cameras, finding hot spots, studying player combinations, finding the most efficient plays. They aren't ranking NBA players using win shares or judging offensive prowess using ORTG and TS%. Analytics are becoming a major part of the game - just not the type of analytic analysis everyone here likes to prop up on a pedestal - and it seems the 'gurus' on here and the ones propping them up are the most lost when it comes to understanding that.

IndyRealist
10-15-2013, 06:29 PM
Came on just to say advanced stats are freakin rubbish.

The eye test never fails unless you're ******** or something.

.
Everything you wrote is nonsense. Human perception lies ALL THE TIME. Memory even moreso. What was the fifth play of the twenty third game you watched last year? Exactly. There's no way for you to remember that. There are 11 moving parts on the floor at any time, 10 players and the ball. The human mind can track 5, sometimes 6. You literally can't see everything going on at the same time. And that doesn't even get into emotional memory bias (why people think dunks are better than layups), confirmation bias (discounting as irrelevant things that contradict your preconceived notions),etc. There are literally whole textbooks on why what you wrote is wrong.

knicks_99
10-15-2013, 06:39 PM
Everything you wrote is nonsense. Human perception lies ALL THE TIME. Memory even moreso. What was the fifth play of the twenty third game you watched last year? Exactly. There's no way for you to remember that. There are 11 moving parts on the floor at any time, 10 players and the ball. The human mind can track 5, sometimes 6. You literally can't see everything going on at the same time. And that doesn't even get into emotional memory bias (why people think dunks are better than layups), confirmation bias (discounting as irrelevant things that contradict your preconceived notions),etc. There are literally whole textbooks on why what you wrote is wrong.

lmfao you took a single sentence of like 2 or 3 paragraphs of what I wrote and singled THAT out?

the only damn "nonsense" is what YOU just wrote. bias? no it's called COMMON DAMN SENSE. lol I can't keep track of 10 players playing? are you serious kid?

you're trying to say that people can't keep "track" of all the players on the floor? are you FOR REAL? MAYBE YOU CAN'T...but I can.

There's a difference between an objective opinion, and ones maybe YOU make. I think Lebron is a piece of crap, but I realize he's most likely the best player in the league.

Like I said, unless you're a ******, it's obvious whose valuable and whose not. Ranking players with stats isn't any less biased than using the eye test, b/c for every "stat" they use, there's like 20 they don't.

Like I said, if you had even BOTHERED to read ANYTHING else I said, hockey assists, running down loose balls, hustle plays, setting good picks, being a coach on the floor, etc. Stats can't and don't illustrate ANY of those things.

What's an assist? What's considered an offensive rebound? You're just another stat nerd who doesn't even realize stats are a man-made interpretation of what's going on on the floor, and their interpretation has CHANGED over the years. Which is why the eye test is and ALWAYS will be the ONLY true dependable resource, b/c unless you're an MIT mathematician, you nor I even knows who the hell transferred Win Shares to basketball, how they did it, with what bias.

Take away your "stats" and YOU'RE lost, but I'm not...probably b/c I've played organized ball before and you haven't, and the only thing that matters is doing w/e it takes to win, whether it's winning a tap, or saving a loose ball, or encouraging your teammates, or setting a hard screen, or making the right rotation on defense, or making the right pass, or knowing when to score and when not to, or making a timely hoop.

Hell, half the stats out there are dependent on the box score guys on the sidelines tracking the freakin game. 50 years ago an assist was impossible to get, rebounds were given willy nilly, and blocks weren't even tracked.

Nowadays EVERYTHING is an assist, rebounds are more accurate.

No 3 pt line back in the day, less practices, less coaching.

ALL that matters is that they did what they had to to win the game. END OF STORY. Stats lie, stats change, stats get warped, wins don't.

Synergy. Look it up kid.

D-Leethal
10-15-2013, 07:01 PM
I do think stats are necessary, but mostly for the simple fact you can't watch all the games. But they always should take a back seat to the eye test. Human perception lies all the time, but thats why you have DVR, game replays, youtube replays of games. If your looking for certain things to form a strong and well-founded opinion on, its not that hard to find yourself an eye test in the form of online game film. Nobody that watches James Harden thinks he is the 4th best player in the league. People that drool over his efficiency stats will though.

knicks_99
10-15-2013, 07:07 PM
I do think stats are necessary, but mostly for the simple fact you can't watch all the games. But they always should take a back seat to the eye test. Human perception lies all the time, but thats why you have DVR, game replays, youtube replays of games. If your looking for certain things to form a strong and well-founded opinion on, its not that hard to find yourself an eye test in the form of online game film. Nobody that watches James Harden thinks he is the 4th best player in the league. People that drool over his efficiency stats will though.


Lol exactly man, but I completely disagree that you need "stats". Meh, you need journalistic integrity and reports of what happened in games; but stats are just fodder. In fact without stats we'd pay MORE attention to what's REALLY going on in games and give a more descriptive account of things b/c we'd be looking at the ENTIRE game and letting it come to us rather than just what stats tell us to look at. We'd notice synergy. How players effect the energy of a game just by their presence...

I mean basic stats yea maybe, points rebounds, steals, blocks, FG%'s, the obvious ones. Anything after that is just masturbation. But even then stat interpretation has changed over the years. Pettit averaged 26 and 16at a time when guys played 42 mpg, and rebounds weren't accurately tracked, not to mention the rules were quite different...does that mean he's better than Karl Malone? Maybe, maybe not.

3RDASYSTEM
10-15-2013, 07:28 PM
Why are people so hostile towards new ideas?

A new idea or poppycock media propaganda?

no player current or ex needs to look at a win share to determine who is the best players to play now and back then and in between

its like JORDAN-WILT-BRON has posted the highest win shares or PER(same thing to me) so they say they are the best but to me they are the best by the impact/game they display on the hardwood, nothing more nothing less

its like todays people say his PER was this or that and im like who cares

for instance J KIDD was MVP caliber for NETS and on his last leg with the KNICKS, being 25yrs old mvp and 40yrs old vet leader is miles away from what KIDD once was but it doesn't take no PER or win shares for me to figure that out it shows in the hardwood

Same with BRON now, he's the same player to me that he was with CAVS, just plays in the post more and has 2 all star caliber players to lean on

3RDASYSTEM
10-15-2013, 07:40 PM
nah playa

Yea playaham

knicks_99
10-15-2013, 07:41 PM
A new idea or poppycock media propaganda?

no player current or ex needs to look at a win share to determine who is the best players to play now and back then and in between

its like JORDAN-WILT-BRON has posted the highest win shares or PER(same thing to me) so they say they are the best but to me they are the best by the impact/game they display on the hardwood, nothing more nothing less



right but i'd completely disagree and counter with bird, magic, and Russell having the actual most IMPACT on the floor all time. guys who could dominate a game with 10 shots.

you can't say the same for wilt or mj, and bron CAN dominate with only 10 shots but seems to never know when the hell to do it.

i'd add Duncan to that list too. Guy just played the right way...

fact is, we just listed the 6 or 7 goat. that's the point. IT'S OBVIOUS....

IndyRealist
10-15-2013, 09:19 PM
lmfao you took a single sentence of like 2 or 3 paragraphs of what I wrote and singled THAT out?
*snip.

Calling someone "kid" doesn't make your argument any better. Stop being a child. I'm guessing from your response you have no idea what cognitive biases are, let alone when you're suffering from them.

And I snipped your post because there wasn't anything worth repeating. Everything you said could be summed up by what I left.

Skip to 0:40 for the "Basketball Test":
http://www.youtube.com/watch?v=nwV1xJu8qqM


At Sloan Bill James talked about basketball (see here around the 26:45 mark). While he admitted he would not be entering basketball stats, he did recount a memory of an interesting game (paraphrased below):

In 2003 the Kansas City Jayhawks lost a title game to Maryland. In this game Juan Dixon scored 23 points. He was guarded primarily by future Chicago Bulls guard Kirk Hinrich. In the thirty-seven minutes Kirk Hinrich guarded Juan Dixon he only scored four points. For the six (or seven) possesions that Keith Langford guarded Juan Dixon he scored 19 points.

There are a few problems with the story (boxscore here):

Maryland didn’t play Kansas City in 2003, they played them in 2002.
They did not play them in the title game (it was in the Elite 8 Final Four).
Juan Dixon scored 33 points not 23
Hinrich only played 29 minutes (making guarding a player for 37 minutes difficult)
Dixon was already up to 10 points by the 11:40 mark of the first half.
What makes this tale interesting is that Bill James prefaced it by saying he watched it multiple times. He also tells the shocked listeners to find the video, implying he had confidence in his memory. This is not actually meant to be dig on Bill James. I consider Bill James to be one of the best people in sports statistics. But even he is not capable of properly remembering one basketball game (and that is with multiple viewings.) This is a simple point. We do not remember things as well as we’d like to think. In fact, we can point out that one of Bill James’ biggest contributions was to help make sure many people were keeping track of stats and using the right stats. He did not advocate all of us finding the several people that watch all of the games for each team and getting an assessment from them. As I mentioned above, while you are watching your attention is divided. There is too much information for you to possibly hold and your emotions (and instant replay) are making the decision of what to highlight for you. And even after all that, your brain isn’t going to be able to store all of that information perfectly.
http://wagesofwins.com/2012/03/22/you-watch-the-games-so-what/

http://wagesofwins.com/2012/07/14/bartenders-height-and-the-eye-test/

3RDASYSTEM
10-15-2013, 09:34 PM
Hey High Horse,
Perhaps I was being a little on the "kind" side, focusing on his "form" rather than his "content".

It is virtually undeniable what (as you correctly point out here) is this guy's agenda: "All Praise To AI; All Condemnation To Kobe"; and he's got like 1% agreement from the rest of PSD NBA.

But still, wouldn't you agree that IF he'd clean up his (non-)English, at least people could better understand his raunch-homeristic arguments?

P.S. Unless he realizes he can't win straight-out; and so, tries to confuse us with his "gobbley-gook"-speak? Still, even so, I don't see anybody qualifying for these two criteria:
1) They understand what he has just said; and
2) They agree with him.

I'm busting a gut, what comic relief!

LeBron James says he watches a lot of tape. Who does he look at?

Michael Jordan and ... Allen Iverson?

"I watch Jordan more than anybody, for sure," James told ESPN's Chris Broussard for a story in the upcoming issue of ESPN The Magazine. "But I'll watch tapes of AI, too. I don't take anything from AI. Well, I do -- his will. They say he was 6 feet, but AI was like 5-10½. Do we even want to say 160? 170 [pounds]? Do we even want to give him that much weight? And he played like a 6-8 2-guard. He was one of the greatest finishers we've ever seen. You could never question his heart. Ever. He gave it his all. AI was like my second-favorite player growing up, after MJ."

3RDASYSTEM
10-15-2013, 09:35 PM
Game is game from day 1 = Players have no peak/prime, they are what they have proven to be from the onset.

He holds the contradictory opinion that players dont make other players better (it helps him with the Nash/Kidd vs AI argument) but he then holds the opinion that Iverson wasn't as efficient as he could have been had he had more support. In essence, hes arguing that other players would have made AI more efficient but that Nash doesn't do that for his teammates.

He argues that Iverson was still capable of playing star basketball beyond his Memphis days but will then use the age/decline excuse when you downplay his performance in Denver.

Sprinkle in some quotes from coach Chuck Daley slurping up AI with him ignoring the unanimous praise Kobe has gotten over his career from a multitude of coaches/peers and you have his agenda down to a T.

LeBron James says he watches a lot of tape. Who does he look at?

Michael Jordan and ... Allen Iverson?

"I watch Jordan more than anybody, for sure," James told ESPN's Chris Broussard for a story in the upcoming issue of ESPN The Magazine. "But I'll watch tapes of AI, too. I don't take anything from AI. Well, I do -- his will. They say he was 6 feet, but AI was like 5-10½. Do we even want to say 160? 170 [pounds]? Do we even want to give him that much weight? And he played like a 6-8 2-guard. He was one of the greatest finishers we've ever seen. You could never question his heart. Ever. He gave it his all. AI was like my second-favorite player growing up, after MJ."

3RDASYSTEM
10-15-2013, 09:42 PM
CHRONZ slurps on BRON who props up IVERSON

basically saying the same thing I was telling the BRON slurper CHRONZ a

how can you not be in the paint alot and be one of the greatest finishers ever?

I love it how people like BRON mention his will and heart because they know his skill and athletic prowess is off the super charts

now add BRON to the equation with DALY and NOLAN RICHARDSON and LARRY BROWN and other slurpers, including the all nba duo of MCKIE and SNOW

the slurpers are unreal

even MUTOMBO said he cant name 4 players ever in same breath as IVERSON in his career

but I get it the slurpers just don't know how to rank a players game, from day 1 not year 15

3RDASYSTEM
10-15-2013, 09:43 PM
BRON said IVERSON played like a 6'8''(TMAC and BRON height) 2 guard

amazing

3RDASYSTEM
10-15-2013, 09:49 PM
CHRONZ as soon as you can name players that have been benched 6 months later after being top 3 in scoring title race to the mvp of the league and media darling then we can have a decent debate about ball

but trust me you wont be able to

I wonder if BRON didn't tow the company line would he come off the bench for the SF-PF version's of STUCKEY and CONLEY

3RDASYSTEM
10-15-2013, 09:50 PM
right but i'd completely disagree and counter with bird, magic, and Russell having the actual most IMPACT on the floor all time. guys who could dominate a game with 10 shots.

you can't say the same for wilt or mj, and bron CAN dominate with only 10 shots but seems to never know when the hell to do it.

i'd add Duncan to that list too. Guy just played the right way...

fact is, we just listed the 6 or 7 goat. that's the point. IT'S OBVIOUS....

Well thanks for making my point even stronger, we don't need to see those guys PER or WS either, they impact the team on that level of best of the best to ever do it, you can have a top 20 list or 25 and it will be legit top to bottom, well mines at least

knicks_99
10-15-2013, 09:57 PM
Calling someone "kid" doesn't make your argument any better. Stop being a child. I'm guessing from your response you have no idea what cognitive biases are, let alone when you're suffering from them.

First off the only godamn "child" is YOU. You didn't "snip" jack crap. You tucked your tail between your legs and chose to ignore my OBVIOUS faultless logic by turning a blind eye, which is typical of people avoiding an argument they can't win.

Second, gtfo here with this 'name calling" complaint when you literally JUST did it.

Third, YOU clearly didn't read anything I had to say....AGAIN. Wake me up when you decide to confront YOUR biases and my ACTUAL quotes...otherwise go away; I'm done feeding trolls. And I took psychology in h.s. and college, I'm quite aware of what you're talking about, and trying to apply a psychological phenomenon to WATCHING SPORTS WITH A OBJECTIVE EYE is the lamest, most asinine thing I've ever read in my life. Only an idiot would say that. If you can't critically evaluate players properly that means you know nothing about basketball, or are a homer stan with no life.

Grow up.


And I snipped your post .... bla bla bla SNIP

No I'm SNIPPING YOU. B/c you AREN'T responding to me. You're purposely ignoring 85% of what I said so you could repeat the EXACT same crap again.

How the hell would you know or not know what "I'm" or anyone else is fake "suffering" from, when I haven't even freakin told you my actual OPINION of any players yet. You're a pompous child, and the only damn one "suffering" from observational bias is YOU. You assume that you can't rely on YOUR observations...b/c they probably suck...seeing as how you probably don't have a keen eye or basketball IQ.

I do, however, and I'm quite confident I can watch a player and make an accurate judgement on them. I look at Russell and I see a guy whose overrated; and yet I can't compensate for why he's the one reoccurring theme in all those championships; either could Red. Obviously not enough of his games are available.

Point is, game respects game. Winners win. And that's all I need to know.

knicks_99
10-15-2013, 10:04 PM
Well thanks for making my point even stronger, we don't need to see those guys PER or WS either, they impact the team on that level of best of the best to ever do it, you can have a top 20 list or 25 and it will be legit top to bottom, well mines at least


I didn't make your "point" stronger bro, I agreed with part of it. I wouldn't narcissistically say "yours would atleast' when you think Wilt impacted the game more than Magic or Bird, that's just laughable. And it just makes you look confident with no reason to be. He is one of the greats, but only when Hannum got him to open up the floor and stop being so damn egotistical did he have significant impact for his TEAM.

I know for a fact my list would be fair and balanced b/c I'd probably grudgingly put MJ over Bird all time, not b/c I want to, or b/c I PERSONALLY believe it, but b/c he did have a little bit better success career wise. But as far as peaks go, those players you/I listed are all on par with eachother. I have no idea if your list would be good or not...for all I know it could be awful, or it could be great...

Extenuating circumstances dictate who has a little more or less success than others. As does mental makeup.

LAcowBOMBER
10-15-2013, 10:13 PM
Well, what did you expect? No one that you should take seriously says stats are the end all, be all of basketball analysis. Yet there are so many people, many in this thread, who say "stats are worthless, watch a game" which directly contradicts what you just said. Stats are to be taken in context, but when one suggests that someone's favorite player isn't as good as they thought he was, suddenly all stats are evil, blah blah blah.

No one likes being told they are wrong. But some perfectly reasonable people say that Kobe was the best player of the last decade. Other perfectly reasonable people say Tim Duncan. Others say Lebron. They can't all be right. Yet these people all watched games, all reasonably assessed what they saw, and all came to completely different conclusions.

Advanced metrics exist for two reasons:

One, you can't watch every game, it's physically impossible. There are 1230 games each season not including playoffs. Most people don't even watch 10% of that. So how can you say that the few games you watched weren't' an aberration, and that the player in question isn't significantly better or worse than you believe? And out of the games you watched last year, what was the 5th offensive play in the 5th game? Exactly. Human memory is flawed, you glean generalities about what you've seen and apply them as a rule. That's why there's a guy in another thread saying Brandon Jennings is an all-star caliber player, probably because the only thing he knows about Jennings is one game his rookie year, and that he was drafted in the lottery. If he bothered to look at his FG%, he'd probably say, "Who the hell drafted this guy in the lottery?" or at the very least "Who the hell gave this guy $25M?"

Two, people give weight to different observations, and they generally get it wrong. Know what the two main driving factors for a player's first non-rookie contract are? Points per game and draft position. Rookie of the year? Points per game and draft position. MVP? Points per game and team wins. 6th man of the year? Points per game. DOES ANYONE HERE THINK POINTS PER GAME IS THE BEST WAY TO JUDGE A PLAYER? Hell no. But that's exactly what we use to judge players, and that's what advanced metrics contradict. When people say that they don't look at stats they watch the game, they are flat out lying. They're simply looking at grossly misleading stats, and then incorporating that into their view of the player.

No one should ever, ever say to ignore what you see and just look at stats. Just like no one should ever say to ignore stats and just watch the game. Both are foolish.

I feel like this post and prior ones defining WS should have been the end of this thread

LAcowBOMBER
10-15-2013, 10:16 PM
First off the only godamn "child" is YOU. You didn't "snip" jack crap. You tucked your tail between your legs and chose to ignore my OBVIOUS faultless logic by turning a blind eye, which is typical of people avoiding an argument they can't win.

Second, gtfo here with this 'name calling" complaint when you literally JUST did it.

Third, YOU clearly didn't read anything I had to say....AGAIN. Wake me up when you decide to confront YOUR biases and my ACTUAL quotes...otherwise go away; I'm done feeding trolls. And I took psychology in h.s. and college, I'm quite aware of what you're talking about, and trying to apply a psychological phenomenon to WATCHING SPORTS WITH A OBJECTIVE EYE is the lamest, most asinine thing I've ever read in my life. Only an idiot would say that. If you can't critically evaluate players properly that means you know nothing about basketball, or are a homer stan with no life.

Grow up.



No I'm SNIPPING YOU. B/c you AREN'T responding to me. You're purposely ignoring 85% of what I said so you could repeat the EXACT same crap again.

How the hell would you know or not know what "I'm" or anyone else is fake "suffering" from, when I haven't even freakin told you my actual OPINION of any players yet. You're a pompous child, and the only damn one "suffering" from observational bias is YOU. You assume that you can't rely on YOUR observations...b/c they probably suck...seeing as how you probably don't have a keen eye or basketball IQ.

I do, however, and I'm quite confident I can watch a player and make an accurate judgement on them. I look at Russell and I see a guy whose overrated; and yet I can't compensate for why he's the one reoccurring theme in all those championships; either could Red. Obviously not enough of his games are available.

Point is, game respects game. Winners win. And that's all I need to know.

Do you watch 80% of all teams games? If not, you can't compare players to other players which is what determines how good a player really is. Even if you have seen every player play 20 games last year, you can't make an honest assessment of how good they are and how they compare to the rest of the league

SPURSFAN1
10-16-2013, 12:33 AM
All the people who hate advance stats root for garbage or perrenial garbage teams to make them feel less bad. Reasonable people are not closed minded. They take the advance stats and put context behind it. Their might be outliers in stats, but for the most part, they get it right or close to right. To outright say all advance stats are bad is just ignorant in my opinion. Nate Silver predicted every state correctly using advance stats in the 2012 election. Advance stats has a place in sports.

torocan
10-16-2013, 09:54 AM
Arguing the value of stats with someone who doesn't even acknowledge the failings and limitations of human observation and memory is utterly pointless. It's like trying to argue that the earth orbits the Sun with someone who uses the Ptolemeic model of the solar system as the basis of assumption.

As for Win Shares, Morey himself said he uses Win Shares, however he uses his OWN formula for determining win shares and uses a *much* deeper pool of statistical data as a basis of his calculations. Fans attempting statistical analysis of teams are operating very much in the dark. The statistical data we typically receive is extremely limited. This is not the same position that GM's like Daryl Morey is in, who uses highly granular data sets provided by professional data gathering systems like the SportsVu product. The difference is akin to attempting to design an atom bomb, except Daryl Morey is using super computers and fans are using a slide ruler. Sure, you might reach the same conclusions as Morey, but he'll do it far faster and have a much higher level of accuracy.

As for the general value of Statistics, they aren't meant to be definitive answers to anything, they're meant to be used as a way of calculating probability and correlations, in other words the likelihood that a player will perform at a given level and the probability of that level increasing and decreasing when placed on the floor with or against certain opponents or teams under specified conditions.

Statistics aren't supposed to be 100% predictive of singular outcomes, and anyone that uses that as an argument against statistics doesn't understand the value of statistics.

For example, I can line up a shot in pool and assign a probability of you making that shot based upon historical observation, however the statistic itself does not predict whether you will or will not make the next attempt, just the probability that you may succeed or fail.

As for Win Shares themselves, they are just one statistical model attempting to predict a certain aspect of athletic performance. Take it for what it is, understand the assumptions behind the model, and consider the limitations of the model and you'll understand the value of the statistic.

Taking a single statistical model and using it as a "proof" or "disproof" of anything without understanding the basis of the model does a disservice to everyone.

mightybosstone
10-16-2013, 10:14 AM
Arguing the value of stats with someone who doesn't even acknowledge the failings and limitations of human observation and memory is utterly pointless. It's like trying to argue that the earth orbits the Sun with someone who uses the Ptolemeic model of the solar system as the basis of assumption.

As for Win Shares, Morey himself said he uses Win Shares, however he uses his OWN formula for determining win shares and uses a *much* deeper pool of statistical data as a basis of his calculations. Fans attempting statistical analysis of teams are operating very much in the dark. The statistical data we typically receive is extremely limited. This is not the same position that GM's like Daryl Morey is in, who uses highly granular data sets provided by professional data gathering systems like the SportsVu product. The difference is akin to attempting to design an atom bomb, except Daryl Morey is using super computers and fans are using a slide ruler. Sure, you might reach the same conclusions as Morey, but he'll do it far faster and have a much higher level of accuracy.

As for the general value of Statistics, they aren't meant to be definitive answers to anything, they're meant to be used as a way of calculating probability and correlations, in other words the likelihood that a player will perform at a given level and the probability of that level increasing and decreasing when placed on the floor with or against certain opponents or teams under specified conditions.

Statistics aren't supposed to be 100% predictive of singular outcomes, and anyone that uses that as an argument against statistics doesn't understand the value of statistics.

For example, I can line up a shot in pool and assign a probability of you making that shot based upon historical observation, however the statistic itself does not predict whether you will or will not make the next attempt, just the probability that you may succeed or fail.

As for Win Shares themselves, they are just one statistical model attempting to predict a certain aspect of athletic performance. Take it for what it is, understand the assumptions behind the model, and consider the limitations of the model and you'll understand the value of the statistic.

Taking a single statistical model and using it as a "proof" or "disproof" of anything without understanding the basis of the model does a disservice to everyone.

I came a little reading this.

tredigs
10-16-2013, 10:32 AM
All the people who hate advance stats root for garbage or perrenial garbage teams to make them feel less bad. Reasonable people are not closed minded. They take the advance stats and put context behind it. Their might be outliers in stats, but for the most part, they get it right or close to right. To outright say all advance stats are bad is just ignorant in my opinion. Nate Silver predicted every state correctly using advance stats in the 2012 election. Advance stats has a place in sports.
Haha, while I agree with the sentiment, that's apples and oranges. Tad difference between extrapolating an outcome from a myriad of poll results and doing the same with something as dynamic as basketball.

Like Nate has talked about, guessing elections is a hell of a lot easier than people make it out to be. The media tell a different story in order to drum up coverage.

b@llhog24
10-16-2013, 11:05 AM
Arguing the value of stats with someone who doesn't even acknowledge the failings and limitations of human observation and memory is utterly pointless. It's like trying to argue that the earth orbits the Sun with someone who uses the Ptolemeic model of the solar system as the basis of assumption.

As for Win Shares, Morey himself said he uses Win Shares, however he uses his OWN formula for determining win shares and uses a *much* deeper pool of statistical data as a basis of his calculations. Fans attempting statistical analysis of teams are operating very much in the dark. The statistical data we typically receive is extremely limited. This is not the same position that GM's like Daryl Morey is in, who uses highly granular data sets provided by professional data gathering systems like the SportsVu product. The difference is akin to attempting to design an atom bomb, except Daryl Morey is using super computers and fans are using a slide ruler. Sure, you might reach the same conclusions as Morey, but he'll do it far faster and have a much higher level of accuracy.

As for the general value of Statistics, they aren't meant to be definitive answers to anything, they're meant to be used as a way of calculating probability and correlations, in other words the likelihood that a player will perform at a given level and the probability of that level increasing and decreasing when placed on the floor with or against certain opponents or teams under specified conditions.

Statistics aren't supposed to be 100% predictive of singular outcomes, and anyone that uses that as an argument against statistics doesn't understand the value of statistics.

For example, I can line up a shot in pool and assign a probability of you making that shot based upon historical observation, however the statistic itself does not predict whether you will or will not make the next attempt, just the probability that you may succeed or fail.

As for Win Shares themselves, they are just one statistical model attempting to predict a certain aspect of athletic performance. Take it for what it is, understand the assumptions behind the model, and consider the limitations of the model and you'll understand the value of the statistic.

Taking a single statistical model and using it as a "proof" or "disproof" of anything without understanding the basis of the model does a disservice to everyone.

Marry me. :love:

Jenceman
10-16-2013, 01:45 PM
I came a little reading this.

Psh, a little?

IndyRealist
10-16-2013, 02:51 PM
I feel like this post and prior ones defining WS should have been the end of this thread
Thanks. It was until knicks_99 necro'd this thread from August, ignored the entire line of debate, and brought us all the way back to page 1.

IndyRealist
10-16-2013, 03:09 PM
I do think stats are necessary, but mostly for the simple fact you can't watch all the games. But they always should take a back seat to the eye test. Human perception lies all the time, but thats why you have DVR, game replays, youtube replays of games. If your looking for certain things to form a strong and well-founded opinion on, its not that hard to find yourself an eye test in the form of online game film. Nobody that watches James Harden thinks he is the 4th best player in the league. People that drool over his efficiency stats will though.
Now that the reasonable people are in the room....

But two perfectly reasonable people can come to vastly different conclusions watching the same game. Take this example (http://www.thenbageek.com/articles/who-will-benefit-from-sportvu), about the new SportVU that will be installed in all NBA arenas this
year:



Want to talk about rebounding? SportVU will tell you how many rebounding chances a player had, how many of his rebounds were contested or uncontested, and how much distance he travels for his rebounds. Reggie Evans led the league in rebounding percentage (the percentage of available rebounds that he grabbed while he was on the floor), but teammate Brook Lopez (in 18 games tracked by SportVU) actually converted a greater percentage of his rebound chances (63 percent vs. 62 percent) where he was in the vicinity of the ball. Furthermore, 54 percent of Lopez’s rebounds were contested, while only 31 percent of Evans’ were. And Lopez traveled 6.4 feet per rebound, while Evans traveled just 4.3 feet.
Here is a textbook example of looking at the data and coming to the exact wrong conclusion about what it means. Let me explain.

Virtually any coach who knows what he is talking about will tell his or her players that rebounding isn't really about who jumps higher, or stands taller. These things help, all things being equal. But they are rarely equal, and that is because rebounding is really about positioning.

Have you ever played racquetball or handball? Have you ever noticed that the really great players look like they aren't really moving very much, yet they always seem to be near the ball, in a perfect position to whack it in a direction that totally screws his or her opponent over? This is because the best racquetball players are not great because they are fantastic at reaching hard-to-get volleys. They are great because they have an uncanny knack for reading the angle, velocity, and spin of the ball, and are very good at getting to spots that are going to be near the ball. And so they appear as if they aren't trying as hard as their opponents.

Some of baseball's best outfielders aren't actually the ones making spectacular dive catches. That's because what they are really good at is getting a jump on a ball as soon as the hitter makes contact, rather than having to wait a few extra hundred milliseconds to recognize the ball's trajectory.

The world's best soccer goaltenders are not the best because they have arms like Mr. Fantastic to snatch goals away from Lionel Messi (although the occasional spectacular save will get you on ESPN). They are in fact the best because they are great at measuring the risk/rewards of leaving the box to interrupt dangerous crosses, headers, etc, because they come out at just the right moment to challenge strikers, and because they know exactly what position to grab to block a header/kick during those times when intercepting the cross isn't possible.

I could go on. The point I am trying to make here is that the data above are pretty clear demonstrators that Reggie Evans is much better than Brook Lopez at rebounding because he is great at positioning. Reggie Travels 4.3 feet to Lopez 6.4 feet? That's probably because Reggie's much faster and more accurate at geting to the most likely spot where the ball is going to come off the rim. More of Lopez's rebounds are contested? That probably means that Lopez isn't blocking out as well, or that since he isn't as close to the ball when it comes off the rim, he has to travel to a spot where some other player is also trying to get it (in other words, it's easier to grab uncontested rebounds when you are the first guy to get to the ball).

It's amazing that two people can look at the data and come to exact opposite conclusions. And this is why SportVU is not going to be a panacea for badly run basketball teams. In fact, it will probably make them worse.

And frankly, all people include stats in their evaluations, they just improperly weigh them. Most people overvalue raw points per game, and don't even bother to look at usage, FG%, or even MINUTES PLAYED. Even the most diehard stat-hater will quote PPG, RPG, APG. Everyone uses stats. They just use them wrong.

ghettosean
10-16-2013, 03:18 PM
And for all the guys who love to boast "Why you think all these teams have analytics departments?"

I'd be willing to bet they use them to evaluate within the context of their own team, and they understand the flaws in crossing them over to other teams. It makes a whole world of sense to hire analytics departments to find out your best lineup combos, your hot spots for each player, who thrives next to who, who gives the team a boost, who struggles next to who, what type of pace you should play etc...

They sure as hell don't use them to rank/compare/rate players from different teams thats for sure. These stats have a long way to go until you can do that without a million grains of salt.

If they put up a statue of you I will worship it!!! :worthy:

Great post!

Guppyfighter
10-16-2013, 05:20 PM
Arguing the value of stats with someone who doesn't even acknowledge the failings and limitations of human observation and memory is utterly pointless. It's like trying to argue that the earth orbits the Sun with someone who uses the Ptolemeic model of the solar system as the basis of assumption.

As for Win Shares, Morey himself said he uses Win Shares, however he uses his OWN formula for determining win shares and uses a *much* deeper pool of statistical data as a basis of his calculations. Fans attempting statistical analysis of teams are operating very much in the dark. The statistical data we typically receive is extremely limited. This is not the same position that GM's like Daryl Morey is in, who uses highly granular data sets provided by professional data gathering systems like the SportsVu product. The difference is akin to attempting to design an atom bomb, except Daryl Morey is using super computers and fans are using a slide ruler. Sure, you might reach the same conclusions as Morey, but he'll do it far faster and have a much higher level of accuracy.

As for the general value of Statistics, they aren't meant to be definitive answers to anything, they're meant to be used as a way of calculating probability and correlations, in other words the likelihood that a player will perform at a given level and the probability of that level increasing and decreasing when placed on the floor with or against certain opponents or teams under specified conditions.

Statistics aren't supposed to be 100% predictive of singular outcomes, and anyone that uses that as an argument against statistics doesn't understand the value of statistics.

For example, I can line up a shot in pool and assign a probability of you making that shot based upon historical observation, however the statistic itself does not predict whether you will or will not make the next attempt, just the probability that you may succeed or fail.

As for Win Shares themselves, they are just one statistical model attempting to predict a certain aspect of athletic performance. Take it for what it is, understand the assumptions behind the model, and consider the limitations of the model and you'll understand the value of the statistic.

Taking a single statistical model and using it as a "proof" or "disproof" of anything without understanding the basis of the model does a disservice to everyone.

Best way I have heard anyone put it.

Guppyfighter
10-16-2013, 05:24 PM
If they put up a statue of you I will worship it!!! :worthy:

Great post!


Why? D-Leethal is projecting how he thinks teams runs? He didn't actually say how they work. He said how they think they work. And he is dead ****ing wrong. This is another great example of people throwing **** at the fan hoping it sticks. You can't say something like this without any kind of evidence and try to pass it off as some kind of fact. That's what ESPN does.

The way D-Leethal argues is akin to how ESPN reports. It's god awful. Subjective. And without merit.

Keeping throwing **** at the wall, D. Maybe some of it will stick.

PatsSoxKnicks
10-17-2013, 04:17 AM
Why? D-Leethal is projecting how he thinks teams runs? He didn't actually say how they work. He said how they think they work. And he is dead ****ing wrong. This is another great example of people throwing **** at the fan hoping it sticks. You can't say something like this without any kind of evidence and try to pass it off as some kind of fact. That's what ESPN does.

The way D-Leethal argues is akin to how ESPN reports. It's god awful. Subjective. And without merit.

Keeping throwing **** at the wall, D. Maybe some of it will stick.

Has he interviewed with an NBA team or something? How would he know? Because one of the things a lot of the analytics positions are seeking right now is the ability to analyze large datasets- likely for the incoming SportsVu tech as well as for the teams that use Vantage stats. But the reasoning for analyzing data like that could be numerous and the comparison of players could be necessary if you are considering a trade. I'm not sure that one all encompassing stat is the way to go if you are trying to measure a players' ability to fit with your team but at the same time, if one player is flat out better, you want to have a way to gauge that. Still, we're at the point where the NBA has pretty much every measurable data point they'd want, even on defense. Right down to the ability to measure how often a team hedges vs. switches on screens for example. Or the ability to measure how often a player contests a shot by getting his hand up. Or measuring hockey assists/passes to missed open shots. All measured.

Guppyfighter
10-17-2013, 04:20 AM
Has he interviewed with an NBA team or something? How would he know? Because one of the things a lot of the analytics positions are seeking right now is the ability to analyze large datasets- likely for the incoming SportsVu tech as well as for the teams that use Vantage stats. But the reasoning for analyzing data like that could be numerous and the comparison of players could be necessary if you are considering a trade. I'm not sure that one all encompassing stat is the way to go if you are trying to measure a players' ability to fit with your team but at the same time, if one player is flat out better, you want to have a way to gauge that. Still, we're at the point where the NBA has pretty much every measurable data point they'd want, even on defense. Right down to the ability to measure how often a team hedges vs. switches on screens for example. Or the ability to measure how often a player contests a shot by getting his hand up. Or measuring hockey assists/passes to missed open shots. All measured.

I wish we had access to all these stats. Maybe someday. But you are definitely right.

IndyRealist
10-17-2013, 09:26 AM
More data does not mean better data, or properly interpreted data. The post I quoted above concerning SportVU is a good read.

torocan
10-17-2013, 10:32 AM
More data does not mean better data, or properly interpreted data. The post I quoted above concerning SportVU is a good read.

It really isn't quite as simple as that. With the advent of SportsVu, the quality of the raw data actually borders on stunning.

Here is an interesting quote from Kirk Goldsberry (Harvard professor) taken from a Slate article linked below...


“We look at that data and we say this isn’t just good data, this is the best space-time data,” Goldsberry says. “It’s just an incredible amount of information, regardless of whether it’s about NBA or anything else … There’s very few people who have ever seen any data like this.”

The problem of course is the quantity of raw data, and the fact that every team is approaching the data in independent parallel tracks.

As many as 1/4 of the NBA teams don't even have an Analytics department, and the size of the Analytics departments (and presumably the quality) varies tremendously. Nobody actually knows how large the Rockets' analytics department is, but it is *rumored* to have a budget as large as 10x larger as the smaller analytics departments. For example, they had 5 accredited representatives at the Sloan conference, and at any given time you may spot anywhere from 1-3 job openings being posted online (interns and full time analytics employees).

One of the central problems in terms of the advancement of Analytics in general is the proprietary nature of Analytics analysis. While the SportsVu raw data is accessible by every team, the methods for approaching and analyzing that data is unique to every team. Think of it as 30 different companies running independent R&D departments without any sharing of methods or conclusions.

Some teams are probably on the right track, other teams are potentially pursuing lines of research that will yield no results at all, while a handful of teams are research and analysis juggernauts operating the equivalent of Xerox Parc (google it if you don't know what that is).

Here's an excellent article from Slate discussing this exact question...

http://www.slate.com/articles/sports/sports_nut/2013/02/nba_stats_gurus_can_t_work_together_anymore_that_s _a_problem.html

So while acceptance of Analytics is rapidly gaining hold in the NBA, expect the results from Analytics to be highly varied depending on the team and organizations commitment to analytics.

PatsSoxKnicks
10-17-2013, 12:08 PM
I wish we had access to all these stats. Maybe someday. But you are definitely right.

More stats like points per chance off a eurostep or FG% after 2 dribbles etc.

And by someday, I think you can expect access to these types of stats in the next few months. Though you will have to pay for it. Already a preview here (compiled midseason though):
http://blog.cacvantage.com/

IndyRealist
10-17-2013, 04:02 PM
It really isn't quite as simple as that. With the advent of SportsVu, the quality of the raw data actually borders on stunning.

Here is an interesting quote from Kirk Goldsberry (Harvard professor) taken from a Slate article linked below...



The problem of course is the quantity of raw data, and the fact that every team is approaching the data in independent parallel tracks.

As many as 1/4 of the NBA teams don't even have an Analytics department, and the size of the Analytics departments (and presumably the quality) varies tremendously. Nobody actually knows how large the Rockets' analytics department is, but it is *rumored* to have a budget as large as 10x larger as the smaller analytics departments. For example, they had 5 accredited representatives at the Sloan conference, and at any given time you may spot anywhere from 1-3 job openings being posted online (interns and full time analytics employees).

One of the central problems in terms of the advancement of Analytics in general is the proprietary nature of Analytics analysis. While the SportsVu raw data is accessible by every team, the methods for approaching and analyzing that data is unique to every team. Think of it as 30 different companies running independent R&D departments without any sharing of methods or conclusions.

Some teams are probably on the right track, other teams are potentially pursuing lines of research that will yield no results at all, while a handful of teams are research and analysis juggernauts operating the equivalent of Xerox Parc (google it if you don't know what that is).

Here's an excellent article from Slate discussing this exact question...

http://www.slate.com/articles/sports/sports_nut/2013/02/nba_stats_gurus_can_t_work_together_anymore_that_s _a_problem.html

So while acceptance of Analytics is rapidly gaining hold in the NBA, expect the results from Analytics to be highly varied depending on the team and organizations commitment to analytics.

I don't disagree that it's amazing data. Where I do disagree is that i think much of the data will not be useful, and that many teams will have great difficulty telling the difference between useful data and "interesting but doesn't change anything" data.

sunsfan88
10-17-2013, 04:59 PM
An extremely stupid stat that only geeks use.

Literally the only good "advanced stat" is true shooting %. Everything else is completely whack. My favorite is when idiots try to use per 48 and per 36 stats. As if players play 48 mins so often lmao and nobody accounts for fatigue in per 36 stats either.

I know I'm gonna get a lot of hate for this post by all the advanced stat admirers but idgaf.

torocan
10-17-2013, 05:04 PM
I don't disagree that it's amazing data. Where I do disagree is that i think much of the data will not be useful, and that many teams will have great difficulty telling the difference between useful data and "interesting but doesn't change anything" data.

I don't disagree in terms of ability to interpret the data. And really I'm not concerned about "many" teams. As long as there are teams that are finding ways to use it meaningfully that knowledge will eventually permeate throughout the league over time as staff members, interns and academic research are disseminated from team to team and league to league.

As for the quality of the data, I'm not in a position to determine the actual value of the data in terms of individual parts, just that the quality of the data is the highest it's ever been in terms of granularity, detail and precision. Whether parts of that data will turn out to be useful or useless is something that will be impossible to know at this early stage of Basketball Analytics.

There were tons of companies that looked at the large amounts of consumer data as containing large amounts of useless information, until the process of data mining and heuristic algorithms reached the point that it was actually useful. Just ask Facebook or Google about whether any of their data is "useless".

I suspect the same will apply to the basketball data being accumulated... ie, the value won't be apparent until someone actually figures out a way to make it valuable either individually or as part of a wider statistical model.

IndyRealist
10-17-2013, 05:10 PM
Sunsfan88, So...you don't get that the point of per minute stats is because two players you are comparing are unlikely to play the same number of minutes? Per game stats are next to useless without accounting for minutes played, at the very least.

Chronz
10-18-2013, 10:45 AM
An extremely stupid stat that only geeks use.

Literally the only good "advanced stat" is true shooting %. Everything else is completely whack. My favorite is when idiots try to use per 48 and per 36 stats. As if players play 48 mins so often lmao and nobody accounts for fatigue in per 36 stats either.

I know I'm gonna get a lot of hate for this post by all the advanced stat admirers but idgaf.

Hate?
Per36, per48 are the same thing.

Per minute stats are simply to put players on more of an equal playing field, you can account for whatever factors you feel exist but just know your "fatigue" theory isn't always true. Often times players play better with an uptick in minutes. It depends on the player obviously but per minute stats are just suppose to reflect level of play while actually on the court. You misunderstanding a statistic doesn't make it stupid, passing off theories as if they are facts however, do.

Heatcheck
10-18-2013, 03:20 PM
Hate?
Per36, per48 are the same thing.

Per minute stats are simply to put players on more of an equal playing field, you can account for whatever factors you feel exist but just know your "fatigue" theory isn't always true. Often times players play better with an uptick in minutes. It depends on the player obviously but per minute stats are just suppose to reflect level of play while actually on the court. You misunderstanding a statistic doesn't make it stupid, passing off theories as if they are facts however, do.

hold on, your doing the same thing, advanced stats are a theory, you cant prove them as fact, but you damn sure pass them off as fact. Its all opinions.

remember that year Terrel Brandon and Donyell Marshall were better Steve Nash? me neither, until PER told me so.

Guppyfighter
10-18-2013, 05:00 PM
hold on, your doing the same thing, advanced stats are a theory, you cant prove them as fact, but you damn sure pass them off as fact. Its all opinions.

remember that year Terrel Brandon and Donyell Marshall were better Steve Nash? me neither, until PER told me so.


Post of the year.

I am going to inform a noble peace prize committee you have just proved the fallibility of math by calling it a theory. In the mean time, other things that are just theories, Evolution, Gravity, and Mantle Convection.

There are facts and opinions. Math consists of facts. They are not suggestions. They are measured results of what occurred on the court.

IndyRealist
10-18-2013, 05:21 PM
hold on, your doing the same thing, advanced stats are a theory, you cant prove them as fact, but you damn sure pass them off as fact. Its all opinions.

remember that year Terrel Brandon and Donyell Marshall were better Steve Nash? me neither, until PER told me so.

Except if you understood PER, you'd know it's not really "advanced".

Guppyfighter
10-18-2013, 05:26 PM
Yeah, no one uses PER.

bagwell368
10-19-2013, 11:06 PM
Your the guy who said Melo was the 6th best offensive player on the Knicks. You are the problem, not the solution. You are the only taking these stats to a level they are not even close to being ready to be at yet.

You really can't get anything right. I said based on advanced stats that Melo was the 6th most efficient offensive player on the Knicks. I said nothing about best. I don't like him because he's a volume shooter, doesn't play D. Advanced stats are useful to back up the fact that he's not efficient and a crappy defensive player (for a guy that Knick fans claim is a top 5 player for sure).

At least try and understand what I write...

Guppyfighter
10-19-2013, 11:12 PM
Uh... Melo is pretty efficient. 570 TS percentage.

tredigs
10-19-2013, 11:20 PM
Uh... Melo is pretty efficient. 570 TS percentage.

His career high isn't even 57% TS.

Guppyfighter
10-19-2013, 11:23 PM
Typo

I meant 560.

Swashcuff
10-20-2013, 03:26 PM
I just came back to PSD and thought I'd chime in a bit on this.

I am an Allen Iverson fanboy so naturally I should HATE WS and most other advanced statistics since they don't score him very well, but I go against that ideal and appreciate advanced stats for what they ARE. Stats.

I realize that many of those who hate advanced stats either A. don't understand them or B. want them to represent something that they weren't made to represent. I think that right there is what causes 14 pages of useless debates.

The advanced stats naysayers are the ones who apparently want WS to tell them how good a shot blocker a player is or eFG% to tell them how good a FT shooter a player is and if it doesn't then the stat is flawed and shouldn't be used, by that notion then every stat is flawed because no stat can tell you everything about a player, PPG doesn't tell you how good a rebounder something is and SPG doesn't tell you how good they are from the arc, so then why hold every single advanced stat to such a pedestal.

I can't speak for everyone who uses advanced stats (and there are some guys who use them wildly not knowing JACK SH** about what they represent or how to use them or not to use them in context) but I know guys like Chronz, Hawkeye15, tredigs, PSK, MHC, KoB, IndyRealist etc watches as much basketball as any of us and they understand and appreciate the correlation between stats (advanced or otherwise) and what they see on the floor. Guys who don't even try to understand the stats in which they bash don't so how on earth can you expect your opinion on the stats to be accurate?

As for the "eye test" being used by itself all I want to guys to answer me is a single question, who's the hottest between Scarlet Johansson, Eva Mendes, Kate Upton and Jessica Biel?

See how many different answers we get and then you'd see what's wrong with using just your eyes.

bagwell368
10-21-2013, 06:44 AM
Uh... Melo is pretty efficient. 570 TS percentage.

It was .560 last year, and that's 102nd out of 469. Above average, but nowhere near where he'd have to be to make "top 3" as his fans claim given all his other shortcomings.

For eFG% it's 164/469.

General statements for all:

My beef with Win Shares goes like this:

1. the DWS component is woefully short of data for an accurate number. earlier (70's and before) it's pathetic.
2. a similar player playing similar minutes on a 60 win team will get more WS than the same guy on a 40 win team. Except in cases of GOAT level players, that sort of difference has more to do with the team and coaches than it does for the player. Providing the percentage of team WS would be helpful to tease this sort of thing out better.

PER is not an advanced stat FYI.

Advanced stats blow away traditional stats, and the advanced stats used by teams are very much deeper than available for free on BR. Just like in baseball, advanced stats are a tool, not meant to replace, but instead enhance the input to the eye, and to help avoid letting subjective factors (like fans of Carmelo or AI) sweeping away objective common sense and judgement.

IMO, many posters on the main forum are so lacking in the ability to view and judge the game at a deep level, it's a shame that they don't use advanced stats which well exceed their conclusions in almost all cases - but instead argue endlessly what they don't know. It's like watching someone that's blindfolded looking for the donkey, and predictably failing over and over.

bagwell368
10-21-2013, 06:46 AM
I just came back to PSD and thought I'd chime in a bit on this.

I am an Allen Iverson fanboy so naturally I should HATE WS and most other advanced statistics since they don't score him very well, but I go against that ideal and appreciate advanced stats for what they ARE. Stats.

I realize that many of those who hate advanced stats either A. don't understand them or B. want them to represent something that they weren't made to represent. I think that right there is what causes 14 pages of useless debates.

The advanced stats naysayers are the ones who apparently want WS to tell them how good a shot blocker a player is or eFG% to tell them how good a FT shooter a player is and if it doesn't then the stat is flawed and shouldn't be used, by that notion then every stat is flawed because no stat can tell you everything about a player, PPG doesn't tell you how good a rebounder something is and SPG doesn't tell you how good they are from the arc, so then why hold every single advanced stat to such a pedestal.

I can't speak for everyone who uses advanced stats (and there are some guys who use them wildly not knowing JACK SH** about what they represent or how to use them or not to use them in context) but I know guys like Chronz, Hawkeye15, tredigs, PSK, MHC, KoB, IndyRealist etc watches as much basketball as any of us and they understand and appreciate the correlation between stats (advanced or otherwise) and what they see on the floor. Guys who don't even try to understand the stats in which they bash don't so how on earth can you expect your opinion on the stats to be accurate?

As for the "eye test" being used by itself all I want to guys to answer me is a single question, who's the hottest between Scarlet Johansson, Eva Mendes, Kate Upton and Jessica Biel?

See how many different answers we get and then you'd see what's wrong with using just your eyes.

Good post.

Heatcheck
10-21-2013, 10:08 AM
Post of the year.

I am going to inform a noble peace prize committee you have just proved the fallibility of math by calling it a theory. In the mean time, other things that are just theories, Evolution, Gravity, and Mantle Convection.

There are facts and opinions. Math consists of facts. They are not suggestions. They are measured results of what occurred on the court.

no **** Einstein, that doesn't mean your application of the math is correct. but yeah go ahead and put advanced stats in the same category as mantle convection or gravity, which is a law, not a theory btw.

Chronz
10-21-2013, 12:44 PM
hold on, your doing the same thing, advanced stats are a theory, you cant prove them as fact, but you damn sure pass them off as fact. Its all opinions.
Not seeing what you think is the theory.


remember that year Terrel Brandon and Donyell Marshall were better Steve Nash? me neither, until PER told me so.
The creator of PER would tell you, no single statistic is the end all be all of any argument, so no try again. You misusing statistics isn't an argument.

Heatcheck
10-21-2013, 12:50 PM
Not seeing what you think is the theory.


The creator of PER would tell you, no single statistic is the end all be all of any argument, so no try again. You misusing statistics isn't an argument.

you damn sure act like it though

Swashcuff
10-21-2013, 03:42 PM
you damn sure act like it though

Then you damn sure don't read his posts, he actually criticizes PER on more occasions than not and says that it isn't an end all and that you absolutely have to use other stats and context with it. No one acts as if PER is the end all except the man who created it, of course since it's his creation he had to stand by it.

Heatcheck
10-21-2013, 03:50 PM
Then you damn sure don't read his posts, he actually criticizes PER on more occasions than not and says that it isn't an end all and that you absolutely have to use other stats and context with it. No one acts as if PER is the end all except the man who created it, of course since it's his creation he had to stand by it.

yeah we've had recent back and forths about PER and he definitely did not leave me with that impression. but thanks for opinion.

Clippersfan86
10-21-2013, 03:56 PM
Chronz has called me out at least 5 times for using PER as an efficiency stat. So I have to vouch for the fact that he definitely is NOT somebody who uses PER a ton or defends it blindly as fact, as Swash already said.

Heatcheck
10-21-2013, 04:10 PM
Chronz has called me out at least 5 times for using PER as an efficiency stat. So I have to vouch for the fact that he definitely is NOT somebody who uses PER a ton or defends it blindly as fact, as Swash already said.

its not so much PER as it is advanced stats in general, I threw the PER out there in the other post because its the one I remember off the top of my head. what I got from the convo is he loves advanced stats, everything he wrote was advanced stats and the way I understood it was as they go he goes. but its cool, like I said in that thread, its the type of argument I prefer to have in person instead of in posts.

Swashcuff
10-21-2013, 04:49 PM
i hate advanced stats espeically in bball. Too many people think they know it all by just looking up advanced stats and never seeing a player play or understand the situation that player is in.


Every players production (or points produced in this case) is a product of their team component. The team component is everywhere in basketball - its a freakin' team sport, and not a team sport of individual matchups like baseball. All 5 guys feed off each other every second they are out there. You can't isolate players in a vaccum like you want to and I know it hurts you stat heads on the inside because you know its true.


Advanced stats in basketball are just silly. You can't isolate the action enough to quantify everything, like you can and should in baseball.


Win Share to me is the most flawed of the advanced stats because of the team aspect of basketball. TS%, USG%, PER, eFG% and TOV% are all good stats though.

Advanced stats work much better in baseball.


This reply is a COMPLETE fail, as advanced statistics in baseball have been able to isolate a pitcher's performance without regard for the defense (ERA is a very weak stat, and no one really pays it much regard anymore), and the same can be said for offensive players, as NO ONE with any credibility uses RBI has a legit statistic. Advanced metrics in baseball have done an amazing job isolating individual control in regards to performance in a way the NBA could never do, due to the nature of the game.

Try again.

The owner of the Dallas Mavericks Mark Cuban admitted on NBA Game Time on NBATV last season that he built his 2011 Mavs based SOLELY on the use of Advanced Statistics. Two years earlier of course he hired a new Director of Analytics Roland Beech (founder of 82games.com).

So this idea that advanced stats have no place in basketball and only works in baseball has been proven wrong two years ago when a team that was built solely on the use of advanced metrics won the NBA championship.

Try again.

Clippersfan86
10-21-2013, 04:52 PM
Swash we also can't forget the Grizzlies last year. Almost everybody including me criticized the hell out of how Hollinger was running the team off of a statistical POV.. but look what happened. He was able to get more out of the team by subtraction of pure talent in Gay, by trading it for a more team oriented, defensive minded set of guys that had much smaller names. Coach Hollins blasted the front office multiple times for this.

Chronz
10-21-2013, 05:17 PM
you damn sure act like it though
Why would I misuse a statistic in a way that the inventor of it says not to? You dont buy a coupe and then complain that it doesn't fit your entire family.

I do consider PER, its a very useful snapshot of a players statistical success while accounting for league averages. I dont always want to look up translated rebound rates between eras and whatnot, so PER gives you that single number to compare it to. It has its flaws, some that I feel are overstated but its still useful.

Chronz
10-21-2013, 05:22 PM
its not so much PER as it is advanced stats in general, I threw the PER out there in the other post because its the one I remember off the top of my head. what I got from the convo is he loves advanced stats, everything he wrote was advanced stats and the way I understood it was as they go he goes. but its cool, like I said in that thread, its the type of argument I prefer to have in person instead of in posts.

Heres the thing, unless you are someone who NEVER uses stats, of ANY kind, you cant complain about advanced stats.

IndyRealist
10-21-2013, 06:48 PM
There is absolutely no one who watches basketball and uses zero stats. EVERYONE looks at least at the basic box score. The whole reason that advanced metrics exist is not because people can't watch a game and know what's going on, but because they look at the box score and say "holy crap 35 points!" while ignoring the 37 shot attempts and 8 turnovers. And when someone doesn't watch a game, which is 90% of games played, all they go by is the box score, which only compounds the problem.

bagwell368
10-23-2013, 04:01 PM
There is absolutely no one who watches basketball and uses zero stats. EVERYONE looks at least at the basic box score. The whole reason that advanced metrics exist is not because people can't watch a game and know what's going on, but because they look at the box score and say "holy crap 35 points!" while ignoring the 37 shot attempts and 8 turnovers. And when someone doesn't watch a game, which is 90% of games played, all they go by is the box score, which only compounds the problem.

Right