PDA

View Full Version : Department of Defense (CP3 best perimeter defender in NBA?)



KnicksorBust
02-25-2015, 03:52 PM
http://grantland.com/features/department-of-defense/

Must read.

Goose17
02-25-2015, 03:59 PM
Read that already. Great read but best perimeter defender? Best at his position maybe but best out of all perimeter players? Nah.

But there's never been a proper defensive analytic yet and this one is clearly still in it's infancy. A lot of variables and it's not great but it's a step in the right direction.

KnicksorBust
02-25-2015, 05:48 PM
Read that already. Great read but best perimeter defender? Best at his position maybe but best out of all perimeter players? Nah.

But there's never been a proper defensive analytic yet and this one is clearly still in it's infancy. A lot of variables and it's not great but it's a step in the right direction.

Who is better? I would argue PG defense has never been more valuable.

ewing
02-25-2015, 05:56 PM
he is not, i really don't see how this is a must read

KnicksorBust
02-25-2015, 05:58 PM
he is not, i really don't see how this is a must read

Whose better ? And I can guarantee you said that bc you either didnt read it or skimmed it. Lol

ewing
02-25-2015, 05:59 PM
Whose better ? And I can guarantee you said that bc you either didnt read it or skimmed it. Lol

why would i need to read it. Paul is not the best perimeter defender in the league. i'll read it later but these people are hurting there own cause

KnicksorBust
02-25-2015, 06:03 PM
Whose better ? And I can guarantee you said that bc you either didnt read it or skimmed it. Lol

why would i need to read it.

Lol. Called it. :) it is a must read. Find the time.

Goose17
02-25-2015, 06:17 PM
Who is better? I would argue PG defense has never been more valuable.

You realise perimeter defenders don't just include PGs right?

KnicksorBust
02-25-2015, 06:19 PM
Who is better? I would argue PG defense has never been more valuable.

You realise perimeter defenders don't just include PGs right?

Did you even read the full article? Lol

kdspurman
02-25-2015, 06:30 PM
i thought it's been Kawhi, non homerism aside. (prior to reading the article) But it is an interesting concept with the metrics used. I don't pretend to understand all advanced stats, but this seems simple & effective enough. Looks like Paul has been great. I knew he was always was an elite defender

Sportfan
02-25-2015, 06:34 PM
I don't think Chris Paul is versatile enough to be called the best perimeter defender in the NBA. Best PG defender sure, but considering he can't guard half of perimeter players takes him out of the running for this argument.

Also having Deandre Jordan helps CP3 a lot, Doc's system has always allowed PG's to be effective defensively, from Rondo to Bradley and to now Paul. All 3 have been worse defenders when Doc wasn't the coach

Goose17
02-25-2015, 06:35 PM
Did you even read the full article? Lol

Yes. I Did. I read it before you did. Somebody shared this article a day or so ago.

I answered your question. Now please answer mines. I will only ask it once more and if you don't I will assume you are ducking it (which it looks like you are)

Are you aware that being the best permiter defender isn't necessarily the same as being the best defender at the point guard position?

And do you understand that this stat is in its infancy? All advanced analytics are but especially this one with it being so new.

KnicksorBust
02-25-2015, 06:35 PM
i thought it's been Kawhi, non homerism aside. (prior to reading the article) But it is an interesting concept with the metrics used. I don't pretend to understand all advanced stats, but this seems simple & effective enough. Looks like Paul has been great. I knew he was always was an elite defender

See and this is one of the pure brilliance moments of the article. We all know Kawhi Leonard is a great defender. But after reading this, we can now elaborate on specific things make him special. For a perimeter player to be able to limit his opponents 3pt attempts to such a staggering degree is such a huge accomplishment.

In regards to CP3, not only does his significantly limit shots but he is holds them to limited output.

Sportfan
02-25-2015, 06:38 PM
See and this is one of the pure brilliance moments of the article. We all know Kawhi Leonard is a great defender. But after reading this, we can now elaborate on specific things make him special. For a perimeter player to be able to limit his opponents 3pt attempts to such a staggering degree is such a huge accomplishment.

In regards to CP3, not only does his significantly limit shots but he is holds them to limited output.

Except the Clippers give up the 4th most 3 point attempts per game.

It's not all Paul shutting them down, it's also because LA's wings are terrible and teams prefer shooting 3's when matched up against Crawford/Rivers/Barnes

KnicksorBust
02-25-2015, 06:42 PM
See and this is one of the pure brilliance moments of the article. We all know Kawhi Leonard is a great defender. But after reading this, we can now elaborate on specific things make him special. For a perimeter player to be able to limit his opponents 3pt attempts to such a staggering degree is such a huge accomplishment.

In regards to CP3, not only does his significantly limit shots but he is holds them to limited output.

Except the Clippers give up the 4th most 3 point attempts per game.

It's not all Paul shutting them down, it's also because LA's wings are terrible and teams prefer shooting 3's when matched up against Crawford/Rivers/Barnes

Parker
Conley
Lillard
Westbrook
Curry
Etc.

He is not matching up with scrubs... PG is loaded.

valade16
02-25-2015, 06:44 PM
Interesting article that gives us another measurement that is detailed to measure on court defensive impact however the main crux of the article, Chris Paul's ability to both keep people from shooting and them shooting a lower %, is still subject to external factors that could influence the result.

Still, I think it's obvious Chris Paul is a very good defender.

Vee-Rex
02-25-2015, 06:49 PM
Definitely a fun article and gives us another way to look at how people are defending. Still a ways off before it's extremely conclusive, though.

CP3 is definitely a top perimeter defender in the league, but probably not the best.

KnicksorBust
02-25-2015, 06:50 PM
Interesting article that gives us another measurement that is detailed to measure on court defensive impact however the main crux of the article, Chris Paul's ability to both keep people from shooting and them shooting a lower %, is still subject to external factors that could influence the result.

Still, I think it's obvious Chris Paul is a very good defender.

Good post

KnicksorBust
02-25-2015, 06:50 PM
Definitely a fun article and gives us another way to look at how people are defending. Still a ways off before it's extremely conclusive, though.

CP3 is definitely a top perimeter defender in the league, but probably not the best.

What did you feel the article is missing?

Sportfan
02-25-2015, 06:59 PM
Parker
Conley
Lillard
Westbrook
Curry
Etc.

He is not matching up with scrubs... PG is loaded.

Again, how can you ignore the fact that teams shoot at will against the Clippers from deep? Basketball is not a 1 on 1 sport, CP3 might not be giving up open looks to PG's, but he's not making a monumental difference. As I said earlier, I prefer a guy like Tony Allen or Leonard who is tasked with guarding the best offensive perimeter player.

By the way this is based off of last year only. I'm willing to bet the numbers this year say different

Redrum187
02-25-2015, 07:00 PM
I really enjoyed reading this article. I've always felt he was the best defensive PG but this gives another POV to question my belief.

KnicksorBust
02-25-2015, 07:06 PM
Parker
Conley
Lillard
Westbrook
Curry
Etc.

He is not matching up with scrubs... PG is loaded.

Again, how can you ignore the fact that teams shoot at will against the Clippers from deep? Basketball is not a 1 on 1 sport, CP3 might not be giving up open looks to PG's, but he's not making a monumental difference. As I said earlier, I prefer a guy like Tony Allen or Leonard who is tasked with guarding the best offensive perimeter player.

By the way this is based off of last year only. I'm willing to bet the numbers this year say different

Speculative at best. Its easy to ignore when your response to CP3's defense is blaming inflation and you use Austin Rivers 17 games of 19mpg as proof. With flimsy arguments like that how can I take you seriously?

Vee-Rex
02-25-2015, 07:17 PM
What did you feel the article is missing?

Not much, honestly.

I'd just like to see it with a plethora of players and see what it says. Like, search for a player's name and see what it tells about them. So it has only shown players that I'd have a good idea of where they were defensively in the first place.

KnicksorBust
02-25-2015, 07:19 PM
What did you feel the article is missing?

Not much, honestly.

I'd just like to see it with a plethora of players and see what it says. Like, search for a player's name and see what it tells about them. So it has only shown players that I'd have a good idea of where they were defensively in the first place.

Good point. One of my favorite parts of the article was where they differentiated between Duncan and Hibbert.

mike_noodles
02-25-2015, 09:57 PM
I was skeptical. It is pretty good. Fun to watch actually. There's lots of room for improvement. For example there's points where the blue lines are clearly connected to the wrong player when there's a lot of motion in an area, like a screen down around the paint. It may not be a big deal, but it could make a difference.

mike_noodles
02-25-2015, 10:03 PM
Again, how can you ignore the fact that teams shoot at will against the Clippers from deep? Basketball is not a 1 on 1 sport, CP3 might not be giving up open looks to PG's, but he's not making a monumental difference. As I said earlier, I prefer a guy like Tony Allen or Leonard who is tasked with guarding the best offensive perimeter player.

By the way this is based off of last year only. I'm willing to bet the numbers this year say different

This was another thing I questioned. There are times when there is no defender assigned to a player on the floor because two are assigned to one offensive player. I haven't seen enough of it to know if a defender is assigned immediately when a player touches the ball. And if so, are they making sure it's the right player. But the question is; are there times when a shooter is so wide open that no defender is assigned?

PowerHouse
02-25-2015, 10:31 PM
.

PowerHouse
02-25-2015, 10:32 PM
.

PowerHouse
02-25-2015, 10:33 PM
Read that already. Great read but best perimeter defender? Best at his position maybe but best out of all perimeter players? Nah.



why would i need to read it. Paul is not the best perimeter defender in the league. i'll read it later but these people are hurting there own cause


I don't think Chris Paul is versatile enough to be called the best perimeter defender in the NBA. Best PG defender sure, but considering he can't guard half of perimeter players takes him out of the running for this argument.

Also having Deandre Jordan helps CP3 a lot, Doc's system has always allowed PG's to be effective defensively, from Rondo to Bradley and to now Paul. All 3 have been worse defenders when Doc wasn't the coach




CP3 is definitely a top perimeter defender in the league, but probably not the best.

I find it pretty surprising that many are quick to baselessly dispute CP as the best perimeter defender without offering a candidate of their own as the best (with some statistical measures to back the argument would be nice too)

Blitzbolt
02-25-2015, 10:52 PM
Homer pick Tony Allen he is just very chaotic and all over the place so this measurments or whatever don't tell the whole story.

For example the Grizzlies 3rd string PG(Cathaletes)is third in that list it just makes no since to me.

ewing
02-25-2015, 11:09 PM
When you have guys like Butler and Leonard that can defend one through 3 at an elite level and 1 through 4 most nights. calling paul the best perimeter defender in the league is a joke. You realize crossmatches happen in the NBA right? When they happen to Jimmy Butler you just match up. When they happen to a Chris Paul your defensive has to collapse and scramble. These guys you can match on literally anyone allow you have someone good to match on anyone 1 through at any time and to hide ****** defenders. Paul doesn't do that. At the point guard spot the greatest value you can bring to your team defensively is not allowing the other team to start there offensive and honestly i don't think Chris Paul is very good at that either. He sticks with his guy well, he's got real strong hands, he double down well and forcing your guy to defend Chris Paul all night helps. Does anyone really think Paul makes the Clippers that much better defensively?

flea
02-25-2015, 11:13 PM
Yeah like others have said, a lot goes into this. Having your PG chuck 3s while CP3 is defending him is just poor basketball, that's part of why it's rarely done. Credit to him for being the best defender at his position, stats like this don't mean a lot when you compare guys who have different roles.

Being an elite perimeter defender in the zone defense era requires a lot, otherwise you're just not that impactful unless you're pulling a James Harden. The converse is that if you just have a decent IQ you probably won't kill your team on the perimeter (see: Tony Parker's laughably good defensive metrics). +/- helps parse this some, but it's not great. It's nice that work is being done but I'm sort of skeptical as to whether it's even possible to measure individual defense on the same level as offense.

ewing
02-25-2015, 11:24 PM
Yeah like others have said, a lot goes into this. Having your PG chuck 3s while CP3 is defending him is just poor basketball, that's part of why it's rarely done. Credit to him for being the best defender at his position, stats like this don't mean a lot when you compare guys who have different roles.

Being an elite perimeter defender in the zone defense era requires a lot, otherwise you're just not that impactful unless you're pulling a James Harden. The converse is that if you just have a decent IQ you probably won't kill your team on the perimeter (see: Tony Parker's laughably good defensive metrics). +/- helps parse this some, but it's not great. It's nice that work is being done but I'm sort of skeptical as to whether it's even possible to measure individual defense on the same level as offense.

If the goal of the work is to tell me who the best perimeter defender in the league is and you come up with CP3 you are doing **** work. That doesn't mean all the #s are useless but the Clips aren't much worse defensively if you replaced CP3 with Tony Parker and the Spurs would go down a couple notches if you replaced Leonard with Jeff Green.

ewing
02-25-2015, 11:29 PM
Btw, I am taking Wall and Beldsoe over CP3 as PGs at min.

flea
02-25-2015, 11:35 PM
I totally agree, the only people in the discussion for great perimeter defenders IMO are the ones who do it nightly. They study for it, it's part of the gameplan, and if it works you win the game because of it. Shawn Marion, Shane Battier, Leonard, Tony Allen, Danny Green (this season), etc. Those guys are impactful nightly because it's their job to be. Most players are impactful, good or bad, only incidentally and based on matchups. This is the reason I laugh at anyone praising Lebron as any sort of all-NBA defender. It's a slap in the face to guys whose employment hinges on what they do defensively - it's not just some feather in the cap people can talk about when comparing you with the GOAT.

But don't get me wrong, +/- has problems too. For instance, I don't care that the Kings are much better defensively with Cousins on the floor. All that means to me is that Cousins is probably the best defender on the 27th ranked defense. He's probably middle of the pack as far as centers go defensively, but that's not good enough to contend for a big man. You're hurting your team if you're a star big man and you're not a top 10 defender.

Shammyguy3
02-25-2015, 11:53 PM
Attention: any link KoB posts is sabotage. Don't read anything.


1. The article was terrible.


2. I can't read


3. I'm not wearing any underwear

two truths and 1 lie

Chronz
02-26-2015, 01:17 AM
When you have guys like Butler and Leonard that can defend one through 3 at an elite level and 1 through 4 most nights. calling paul the best perimeter defender in the league is a joke. You realize crossmatches happen in the NBA right? When they happen to Jimmy Butler you just match up. When they happen to a Chris Paul your defensive has to collapse and scramble.
I disagree, it really depends on the gameplan. Come playoffs, you'd be surprised how little help he receives, especially compared to how much he provides. The dude has mastered the art of rotate and recover, then on switches, he has a low center of gravity, his solid base isn't easily dislodged. Guys can shoot over him, but they wont get any deeper than he allows and he does enough to harass the likes of Durant.

Just did a quick youtube search and found an example. This was in R2 last year when Caron was abusing Jamal Crawford in the post and Doc sicced CP3 on him. JVG makes mention of the "better matchup"
http://youtu.be/ApsC0GTSrqw?t=50s

Its sad but he was one of the better defenders vs Durant as well. His defense against KD was what sparked that comeback in their G4 win. I dont know of any other midget PG who could put up that kind of fight against a 6"11 pogo stick/MVP.

When CP3 doesn't have to generate all the offense, his defense is of greater consequence, he hasn't had that opportunity much. Likely carries the greatest 2-way load IMO.


At the point guard spot the greatest value you can bring to your team defensively is not allowing the other team to start there offensive and honestly i don't think Chris Paul is very good at that either. He sticks with his guy well, he's got real strong hands, he double down well and forcing your guy to defend Chris Paul all night helps. Does anyone really think Paul makes the Clippers that much better defensively?
I truly do, I think you underrate him. I think the most important aspect a PG defender must have is fighting over picks and understanding the nuances of help and recover. When CP3 defends non-shooters (which he doesn't really have the luxury of) he plays the passing lanes as well as any PG. You dont have to be tall to stunt opposing teams PnR. Not to mention hes a good rebounder for his position/size.

ewing
02-26-2015, 09:43 AM
I disagree, it really depends on the gameplan. Come playoffs, you'd be surprised how little help he receives, especially compared to how much he provides. The dude has mastered the art of rotate and recover, then on switches, he has a low center of gravity, his solid base isn't easily dislodged. Guys can shoot over him, but they wont get any deeper than he allows and he does enough to harass the likes of Durant.

Just did a quick youtube search and found an example. This was in R2 last year when Caron was abusing Jamal Crawford in the post and Doc sicced CP3 on him. JVG makes mention of the "better matchup"
http://youtu.be/ApsC0GTSrqw?t=50s

Its sad but he was one of the better defenders vs Durant as well. His defense against KD was what sparked that comeback in their G4 win. I dont know of any other midget PG who could put up that kind of fight against a 6"11 pogo stick/MVP.

When CP3 doesn't have to generate all the offense, his defense is of greater consequence, he hasn't had that opportunity much. Likely carries the greatest 2-way load IMO.


I truly do, I think you underrate him. I think the most important aspect a PG defender must have is fighting over picks and understanding the nuances of help and recover. When CP3 defends non-shooters (which he doesn't really have the luxury of) he plays the passing lanes as well as any PG. You dont have to be tall to stunt opposing teams PnR. Not to mention hes a good rebounder for his position/size.

I don't disagree, with much of what you presented. I think still think the notion of Paul being the best perimeter defender in the league is laughable. i do think the most important think a PG can do on D is apply pressure- that's why i love guys like Avery Bradley. As far a screen roll goes, he fights and is good, it also helps to have Tyson or Jordan as the player the guard is being pushed into when you go over. If it was David Lee, i think you would see him go under more and it would be the right thing to do

KnicksorBust
02-26-2015, 11:57 AM
I totally agree, the only people in the discussion for great perimeter defenders IMO are the ones who do it nightly. They study for it, it's part of the gameplan, and if it works you win the game because of it. Shawn Marion, Shane Battier, Leonard, Tony Allen, Danny Green (this season), etc. r.

"How Analytics Made Me a Better Basketball Player, with Shane Battier"
https://www.youtube.com/watch?v=nxcbuV0_WEM
originally posted: Feb 25, 2015

Transcript: Sports are about gaining the competitive advantage and for so long the traditional ways of gaining the advantage is through harder training. You know basketball which I played it was honing your jump shot. It was working harder in the weight room to become faster, jump a little higher. I was lucky to sort of grow up in sort of the golden age of analytics as it pertains to basketball in the age of really the emergence of big data and sports. The way I look at it is itís just another way like honing your jump shot or honing your jump hook or getting faster or stronger to gain a competitive advantage on the basketball court, this time using numbers and data.

Before I really learned analytics I tried to guard a guy, Kobe Bryant, who in my estimation was the toughest competitor that I ever played against. And all I had to rely on was the old eyeball test scouting report. Kobeís got a really good right hand. You have to keep him out of the painted area. Heís a great finisher. So yeah, any Joe Schmo fan could tell you those things. But after studying and going through the school of analytics I knew exactly to a tee who Kobe Bryant was. And I knew as a defender trying to stop him Kobeís worst case scenario and my best case scenario was to make him shoot a pull up jumper going to his left hand, all right. The average possession of the Los Angeles Lakers in 2008 was generated .98 points per possession, .98. So you took the average possession of the Lakers. They were going to score .98 points every time they had a possession. And so Kobe Bryant only shot the left handed pull up jumper at a 44 percent clip. So every time that he went left and shot that pull up jumper he was generating .88 points per possession.

Well thatís a tenth of a point less than the average Laker possession. And so if I could make him do that time and time again which is a lot tougher to do than to say, Iím shaving off a tenth of a point every single time. Iím actually making him detrimental to his team. And the way you have to look at it is all these tenths of points, all right, add up. They add up here, they add up here, they add up there. And all of a sudden those tenths of a point become points. And in the NBA as we all know the margin between wins and loses is very, very thin. So those tenths of points matter. And thatís all it really is. Itís no different than playing the stock market. Youíre trying to shave percentage points off your risk. And if you can accumulate enough, guess what? Youíre going to do very well. And so guarding a guy like Kobe Bryant, understanding exactly who he is, what his weakness is, made me a much better defender and allowed me to stick around the NBA for 13 years.

KnicksorBust
02-26-2015, 11:58 AM
I disagree, it really depends on the gameplan. Come playoffs, you'd be surprised how little help he receives, especially compared to how much he provides. The dude has mastered the art of rotate and recover, then on switches, he has a low center of gravity, his solid base isn't easily dislodged. Guys can shoot over him, but they wont get any deeper than he allows and he does enough to harass the likes of Durant.

Just did a quick youtube search and found an example. This was in R2 last year when Caron was abusing Jamal Crawford in the post and Doc sicced CP3 on him. JVG makes mention of the "better matchup"
http://youtu.be/ApsC0GTSrqw?t=50s

Its sad but he was one of the better defenders vs Durant as well. His defense against KD was what sparked that comeback in their G4 win. I dont know of any other midget PG who could put up that kind of fight against a 6"11 pogo stick/MVP.

When CP3 doesn't have to generate all the offense, his defense is of greater consequence, he hasn't had that opportunity much. Likely carries the greatest 2-way load IMO.


I truly do, I think you underrate him. I think the most important aspect a PG defender must have is fighting over picks and understanding the nuances of help and recover. When CP3 defends non-shooters (which he doesn't really have the luxury of) he plays the passing lanes as well as any PG. You dont have to be tall to stunt opposing teams PnR. Not to mention hes a good rebounder for his position/size.

This is someone who probably watched 98% of Chris Paul's minutes last year. #analyticspluseyetest

KnicksorBust
02-26-2015, 12:02 PM
I don't disagree, with much of what you presented. I think still think the notion of Paul being the best perimeter defender in the league is laughable. i do think the most important think a PG can do on D is apply pressure- that's why i love guys like Avery Bradley. As far a screen roll goes, he fights and is good, it also helps to have Tyson or Jordan as the player the guard is being pushed into when you go over. If it was David Lee, i think you would see him go under more and it would be the right thing to do

Chronz goes on a rant about how extremely valuable Chris Paul's defense is for the Clippers and then you say you agree but still think it's laughable. That makes no sense. This is the golden era of PGs and pick and rolls. Defense from that position has never been more valuable and he is significantly better than his contemporaries. I think it's laughable that you downplay the value of the guy guarding Curry/Westbrook/Lillard.

ewing
02-26-2015, 12:13 PM
Chronz goes on a rant about how extremely valuable Chris Paul's defense is for the Clippers and then you say you agree but still think it's laughable. That makes no sense. This is the golden era of PGs and pick and rolls. Defense from that position has never been more valuable and he is significantly better than his contemporaries. I think it's laughable that you downplay the value of the guy guarding Curry/Westbrook/Lillard.

Chronz went on a rant about what Chris Paul does well defensively. I agree he does many things well and I have never said he wasn't a good defender.

Minimal
02-26-2015, 12:28 PM
I'll just say 1 thing. LeBron can guard CP3, CP3 can't guard LeBron.

kdspurman
02-26-2015, 12:41 PM
I'll just say 1 thing. LeBron can guard CP3, CP3 can't guard LeBron.

Not too relevent for this particular topic in all honesty. Lebron has shown he can be really good defensively, but more so in spurts.

Minimal
02-26-2015, 01:02 PM
Not too relevent for this particular topic in all honesty. Lebron has shown he can be really good defensively, but more so in spurts.
Yeah I understand that, he is not the best defender during the regular season, but when the time comes, he is really good.
What I ment by that phrase is that there are almost none PGs that can stop a good SG or SF, which are basically perimeter players. The only defense on the perimeter that they can do are against PGs. When switches happen they are on the losing side. Thats what I meant so best perimeter defenders in my opinion are Kawhi, Draymond Green, Tony Allen, LeBron type of players, not an undersized PG. Chris Paul is not even the best PG defender, Westbrook is. Westbrook leads OKC Thunder in Defensive +/-. And you don't even need stats to prove he is the best defender, you can clearly see it.

RiLoc
02-26-2015, 01:21 PM
My takeaway was that it's very interesting and I think it'd be invaluable for team building and game planning, but I don't think it's a fair measurement in terms of boiling defensive ability down into a number and ranking players by it. I think the closest we have is Defensive RAPM (http://stats-for-the-nba.appspot.com/) or Defensive RPM (http://espn.go.com/nba/statistics/rpm/_/sort/DRPM) (RPM explanation (http://espn.go.com/nba/story/_/id/10740818/introducing-real-plus-minus)).

IndyRealist
02-26-2015, 01:27 PM
See this is why people are so against stats. They measured a very specific thing: attempts vs. average and fg% vs. average, for opponents who the player is primarily responsible for guarding.

But people are here talking about versatility and guarding multiple positions and ball pressure. They did not take the time to understand what the stats are saying. They did not note that the authors are using SportVU data to compile their results. They just resorted to cliches.

You should at least understand how something works and what it is saying before you dismiss it.

KnicksorBust
02-26-2015, 01:28 PM
Yes. I Did. I read it before you did. Somebody shared this article a day or so ago.

I answered your question. Now please answer mines. I will only ask it once more and if you don't I will assume you are ducking it (which it looks like you are)

Are you aware that being the best permiter defender isn't necessarily the same as being the best defender at the point guard position?

And do you understand that this stat is in its infancy? All advanced analytics are but especially this one with it being so new.

Yes I understand that perimeter defense isn't the same as point guard defense.Did I really need to explicitly state that? I didn't think your question was actually serious. :)

In regards to your other question, are you insinuating that new = flawed?

KnicksorBust
02-26-2015, 01:31 PM
Chronz went on a rant about what Chris Paul does well defensively. I agree he does many things well and I have never said he wasn't a good defender.

And yet it's "laughable" to discuss him as the top perimeter defender... despite the fact that he's made multiple all-defensive teams, despite the fact that statistical evidence illustrates that he significantly limits opponents shots AND efficiency, and despite the fact that you concede he's a good defender.

ewing
02-26-2015, 01:34 PM
See this is why people are so against stats. They measured a very specific thing: attempts vs. average and fg% vs. average, for opponents who the player is primarily responsible for guarding.

But people are here talking about versatility and guarding multiple positions and ball pressure. They did not take the time to understand what the stats are saying. They did not note that the authors are using SportVU data to compile their results. They just resorted to cliches.

You should at least understand how something works and what it is saying before you dismiss it.

The what is being debated is the use of the stats to assert that CP3 is the best perimeter defender in the league not the validity of the of the stats themselves. Versatility and ball pressure are 2 things that make someone valuable that are not measured here

KnicksorBust
02-26-2015, 01:36 PM
Yeah I understand that, he is not the best defender during the regular season, but when the time comes, he is really good.
What I ment by that phrase is that there are almost none PGs that can stop a good SG or SF, which are basically perimeter players. The only defense on the perimeter that they can do are against PGs. When switches happen they are on the losing side. Thats what I meant so best perimeter defenders in my opinion are Kawhi, Draymond Green, Tony Allen, LeBron type of players, not an undersized PG. Chris Paul is not even the best PG defender, Westbrook is. Westbrook leads OKC Thunder in Defensive +/-. And you don't even need stats to prove he is the best defender, you can clearly see it.

Who is a better defender Chandler Parsons or Chris Paul?

ewing
02-26-2015, 01:39 PM
and yet it's "laughable" to discuss him as the top perimeter defender... Despite the fact that he's made multiple all-defensive teams, despite the fact that statistical evidence illustrates that he significantly limits opponents shots and efficiency, and despite the fact that you concede he's a good defender.

yes, how many times do you want me to say it. You can keep saying he is. Doesn't make it any less stupid

KnicksorBust
02-26-2015, 01:42 PM
My takeaway was that it's very interesting and I think it'd be invaluable for team building and game planning, but I don't think it's a fair measurement in terms of boiling defensive ability down into a number and ranking players by it. I think the closest we have is Defensive RAPM (http://stats-for-the-nba.appspot.com/) or Defensive RPM (http://espn.go.com/nba/statistics/rpm/_/sort/DRPM) (RPM explanation (http://espn.go.com/nba/story/_/id/10740818/introducing-real-plus-minus)).

What do you find unfair about tracking shots allowed by a defender?

KnicksorBust
02-26-2015, 01:43 PM
yes, how many times do you want me to say it. You can keep saying he is. Doesn't make it any less stupid

Have I ever actually said CP3 is the best perimeter defender in the NBA? :laugh:

KnicksorBust
02-26-2015, 01:45 PM
See this is why people are so against stats. They measured a very specific thing: attempts vs. average and fg% vs. average, for opponents who the player is primarily responsible for guarding.

But people are here talking about versatility and guarding multiple positions and ball pressure. They did not take the time to understand what the stats are saying. They did not note that the authors are using SportVU data to compile their results. They just resorted to cliches.

You should at least understand how something works and what it is saying before you dismiss it.

Yeah I thought it was so interesting. Things that we already know "offensive players attack Harden more," "Kawhi Leonard is a great defender" "Hibbert protects the rim" ... etc. They are all confirmed. Now we get to look at them even more objectively with evidence to support our claims.

lol, please
02-26-2015, 02:09 PM
Did Klay Thompson retire? :confused:

Chronz
02-26-2015, 02:33 PM
Im partial to Wall here as well. Wonder what he looks like by this measure.

Bruno
02-26-2015, 03:49 PM
this stuff is pretty brilliant.

i think that if you're 6'4 or under Chris Paul is probably one of the top defenders perimeter defender, arguably the best. Obviously nobodys going to chose Chris Paul as the perimeter defender of choice if the offensive perimiter player is Kevin Durant. top one or two at his position, the position where you have to bring it almost every night, totally.

RiLoc
02-26-2015, 04:58 PM
My takeaway was that it's very interesting and I think it'd be invaluable for team building and game planning, but I don't think it's a fair measurement in terms of boiling defensive ability down into a number and ranking players by it. I think the closest we have is Defensive RAPM (http://stats-for-the-nba.appspot.com/) or Defensive RPM (http://espn.go.com/nba/statistics/rpm/_/sort/DRPM) (RPM explanation (http://espn.go.com/nba/story/_/id/10740818/introducing-real-plus-minus)).
What do you find unfair about tracking shots allowed by a defender?
I don't find anything unfair about it in general, I don't think it's fair to boil down a player's defensive ability down singularly by the metrics mentioned in the article.

There is more to defense than how often and how much the opposing player you're guarding shoots. The article is awesome and there is real tangible value to knowing that information, I just don't think you can take that data and use it as the singular measurement of a defenders ability.

KnicksorBust
02-26-2015, 08:18 PM
Did Klay Thompson retire? :confused:

Be real. I don't have time for this in my life. :laugh:

KnicksorBust
02-26-2015, 08:19 PM
this stuff is pretty brilliant.

i think that if you're 6'4 or under Chris Paul is probably one of the top defenders perimeter defender, arguably the best. Obviously nobodys going to chose Chris Paul as the perimeter defender of choice if the offensive perimiter player is Kevin Durant. top one or two at his position, the position where you have to bring it almost every night, totally.

Bruno makes my life easier. Although if Ewing read this he would probably say "Yeah this is completely correct. Bruno mad respect.. CP3 is the man. Great defender. Locks down a tough position... OH BUT BY THE WAY Paul is laughable as the top defender! laughable!!!" :) jk

KnicksorBust
02-26-2015, 08:22 PM
I don't find anything unfair about it in general, I don't think it's fair to boil down a player's defensive ability down singularly by the metrics mentioned in the article.

There is more to defense than how often and how much the opposing player you're guarding shoots. The article is awesome and there is real tangible value to knowing that information, I just don't think you can take that data and use it as the singular measurement of a defenders ability.

I have respect for your post but honestly who is saying to do that? I'm not saying it. The article isn't saying it. The greatest myth about analytic people is that they strut around in their George Mikan goggled frames spouting "PER and PER only define greatness!" "BOW TO TRUE SHOOTING PERCENTAGE%" "YO MAMA LOVE MY ADJUSTED RAPM!"

ewing
02-26-2015, 11:27 PM
damian lillard can put the ball in the hole. Saying he is the best scorer in the NBA is laughable

valade16
02-27-2015, 12:03 PM
Have we considered the fact that because Chris Paul is guarding PGs, who generally have better court awareness and are better passers, that the reason they don't shoot very much vs. Chris Paul is because they are passing to better shooting opportunities for their teammates?

The guy Tony Allen or Kawhi Leonard are guarding are probably way less likely to give up the rock so that inflates their shooting numbers against.

That being said, I think you could draw a pretty even parallel of defensive impact by this measurement when comparing PGs, because they are playing against players of similar ball handling and passing abilities, but that is just one inherent flaw with the data when comparing Chris Paul to a more traditional wing defender.

valade16
02-27-2015, 12:05 PM
damian lillard can put the ball in the hole. Saying he is the best scorer in the NBA is laughable

As a Blazer fan I can say Lillard is starting to get very overrated. Not only is he not the best scorer in the NBA, he is not even the best scoring PG. Westbrook and Curry are easily, easily better. And there are number of guys at PG after that you could argue are every bit as good if not better.

ewing
02-27-2015, 12:07 PM
As a Blazer fan I can say Lillard is starting to get very overrated. Not only is he not the best scorer in the NBA, he is not even the best scoring PG. Westbrook and Curry are easily, easily better. And there are number of guys at PG after that you could argue are every bit as good if not better.

it'd be like saying chris paul is the best perimeter defender in the NBA btw, i love lillard and enjoy watching your team

D-Leethal
02-27-2015, 12:10 PM
I have respect for your post but honestly who is saying to do that? I'm not saying it. The article isn't saying it. The greatest myth about analytic people is that they strut around in their George Mikan goggled frames spouting "PER and PER only define greatness!" "BOW TO TRUE SHOOTING PERCENTAGE%" "YO MAMA LOVE MY ADJUSTED RAPM!"

To be honest, there are plenty of posters who might not outright say that but certainly imply it.

This is the same forum where Mason Plumlee is a better offensive player than Nik Vucevic because TS% and ORTG say so.

KnicksorBust
02-27-2015, 12:13 PM
damian lillard can put the ball in the hole. Saying he is the best scorer in the NBA is laughable

This is the second time you've done this. Your comparison is not logical. Paul has thousands of plays of data to support that he limits opponents shot attempts and overall scoring to elite levels. Lillard is not at the top (or even top 3) of any major statistical category. He's 11th in PPG and has a respected 56%TS% but that's not even close to the top 20.

Making a bad comparison doesn't make your arguments any stronger. In fact it makes them weaker.

ewing
02-27-2015, 12:16 PM
This is the second time you've done this. Your comparison is not logical. Paul has thousands of plays of data to support that he limits opponents shot attempts and overall scoring to elite levels. Lillard is not at the top (or even top 3) of any major statistical category. He's 11th in PPG and has a respected 56%TS% but that's not even close to the top 20.

Making a bad comparison doesn't make your arguments any stronger. In fact it makes them weaker.

thanks

KnicksorBust
02-27-2015, 12:20 PM
To be honest, there are plenty of posters who might not outright say that but certainly imply it.

This is the same forum where Mason Plumlee is a better offensive player than Nik Vucevic because TS% and ORTG say so.

I'm not familiar with the Plumlee > Vuc discussion but just because there are some that oversimplify comparisons doesn't make the statistics any less relevant. If advanced statistics support an argument then I see no problem in the validity of using them. You just need to keep adding context to the argument until there is a full picture. Subjective AND objective proof.

D-Leethal
02-27-2015, 12:42 PM
I'm not familiar with the Plumlee > Vuc discussion but just because there are some that oversimplify comparisons doesn't make the statistics any less relevant. If advanced statistics support an argument then I see no problem in the validity of using them. You just need to keep adding context to the argument until there is a full picture. Subjective AND objective proof.

You said it was a myth that people make those arguments - it is not a myth around here actually.

KnicksorBust
02-27-2015, 05:08 PM
I'm not familiar with the Plumlee > Vuc discussion but just because there are some that oversimplify comparisons doesn't make the statistics any less relevant. If advanced statistics support an argument then I see no problem in the validity of using them. You just need to keep adding context to the argument until there is a full picture. Subjective AND objective proof.

You said it was a myth that people make those arguments - it is not a myth around here actually.

Smh.

The myth is that analytic supporters believe any singular stat can define the scope of a player's overall impact. Feel free to quote all the posts you can find where people make that argument. What you are referencing are debates where the posters use advanced statistics to prove a point.

valade16
02-27-2015, 05:30 PM
If I were to make a list of the Top defensive PGs (among top quality PGs) it'd be something like (in no order):

John Wall
Chris Paul
Kyle Lowry
Mike Conley

I think those 4 are really good defenders who are also among the Top PGs in the league.

DODGERS&LAKERS
02-27-2015, 09:06 PM
Have we considered the fact that because Chris Paul is guarding PGs, who generally have better court awareness and are better passers, that the reason they don't shoot very much vs. Chris Paul is because they are passing to better shooting opportunities for their teammates?

The guy Tony Allen or Kawhi Leonard are guarding are probably way less likely to give up the rock so that inflates their shooting numbers against.

That being said, I think you could draw a pretty even parallel of defensive impact by this measurement when comparing PGs, because they are playing against players of similar ball handling and passing abilities, but that is just one inherent flaw with the data when comparing Chris Paul to a more traditional wing defender.

But they are not comparing point guard shooting vs shooting guard shooting. They are comparing one point guards average shot attempts to when they face a particular player. So if Lillard took on average a shot 30 times for every 100 possessions, this data says if Paul was guarding him he might shoot only 25/100 and do so less efficiently. Meaning a 5% drop off.
If Harden shoots 40/100 and Kawaii holds him to 36/100, Paul was a better defender according to this.

DODGERS&LAKERS
02-27-2015, 09:11 PM
I am glad they are attempting to look at different metrics of defensive ability. I hate when somebody tries to throw at me the defensive rating of a player to say that he is better than another player. If that was the case, Carlos Boozer is a way better defender than Chris Paul. With a defensive rating of 102 to Pauls 104 for their career.

Defensive rating is totally reliant upon a player's teammates and the coaching staffs scheme and enforcement of it. This stat seems to hold individual players more accountable although teammates and coaching will still have a heavy influence on it

Chronz
02-27-2015, 09:21 PM
Have we considered the fact that because Chris Paul is guarding PGs, who generally have better court awareness and are better passers, that the reason they don't shoot very much vs. Chris Paul is because they are passing to better shooting opportunities for their teammates?
Well it depends on the looks those PG's generate, Im guessing the next step for this stat is to account for assists (if its not doing that already, I dunno), but this is based on old data, Im presuming when the Clippers were a better defensive team than they've been this year. Its not all on CP3 but the mere fact that the position that dominates the rest of the league the most, is dropping off to this degree against CP3 and forcing inferior teammates to beat his inferior teammates is still indicative of his defensive stature. The guy truly is relentless in forcing turnovers and fighting for position, I almost think he enjoys bodying up bigger guys and forcing them to miss.


The guy Tony Allen or Kawhi Leonard are guarding are probably way less likely to give up the rock so that inflates their shooting numbers against.

But if they do their jobs, they should lower the expected efficiency to a greater degree. That CP3 does both is the ideal outcome. That CP3 is a legit starter going up against prime time players makes him unique. Guys like Tony Allen and Kawhi dont carry his offensive responsibility, hell, TA doesn't even play starters minutes. So even if TA is a better defender, hes not out there, impacting the game all day. Hes out there half the time and only focuses on 1 end. Give CP3 that luxury and hes a better defender too.

I doubt CP3 is the best defender on the perimeter, but what I do know, is that his defensive prowess goes largely ignored when discussing the best PG's. Most people focus on the offensive end when it comes to PG's, which is understandable, but it doesn't mean we should ignore defense entirely. I really wish CP3 had big swings that could take the pressure off him defensively. We would appreciate his defense alot more if the team were elite on that end.


That being said, I think you could draw a pretty even parallel of defensive impact by this measurement when comparing PGs, because they are playing against players of similar ball handling and passing abilities, but that is just one inherent flaw with the data when comparing Chris Paul to a more traditional wing defender.
Without looking at the actual data, I cant comment on any sort of inherent flaw. All we know is that the preliminary findings back what most analysts within the game can attest to. Hopefully we get this out soon.

DODGERS&LAKERS
02-27-2015, 09:32 PM
Well it depends on the looks those PG's generate, Im guessing the next step for this stat is to account for assists (if its not doing that already, I dunno), but this is based on old data, Im presuming when the Clippers were a better defensive team than they've been this year. Its not all on CP3 but the mere fact that the position that dominates the rest of the league the most, is dropping off to this degree against CP3 and forcing inferior teammates to beat his inferior teammates is still indicative of his defensive stature. The guy truly is relentless in forcing turnovers and fighting for position, I almost think he enjoys bodying up bigger guys and forcing them to miss.


But if they do their jobs, they should lower the expected efficiency to a greater degree. That CP3 does both is the ideal outcome. That CP3 is a legit starter going up against prime time players makes him unique. Guys like Tony Allen and Kawhi dont carry his offensive responsibility, hell, TA doesn't even play starters minutes. So even if TA is a better defender, hes not out there, impacting the game all day. Hes out there half the time and only focuses on 1 end. Give CP3 that luxury and hes a better defender too.

I doubt CP3 is the best defender on the perimeter, but what I do know, is that his defensive prowess goes largely ignored when discussing the best PG's. Most people focus on the offensive end when it comes to PG's, which is understandable, but it doesn't mean we should ignore defense entirely. I really wish CP3 had big swings that could take the pressure off him defensively. We would appreciate his defense alot more if the team were elite on that end.


Without looking at the actual data, I cant comment on any sort of inherent flaw. All we know is that the preliminary findings back what most analysts within the game can attest to. Hopefully we get this out soon.

I haven't watched enough Clipper games to know, but is Doc still employing his old defensive schemes he used in Boston with Thibs as his defensive coordinator? They used to load up on the strong side where the ball was to force the offensive team to swing it around. Are you guys still doing that?

If so a smart point guard would move the ball and not shoot as frequently. He would not be able to get to the rim because they are basically playing the zone with the Deandre back there ready to contest. I am also interested to see if they are taking into account how many assist the other player gets as well as turnovers forced

DODGERS&LAKERS
02-27-2015, 09:34 PM
I wonder if they are able to go back in history to see how other players fared. It would be interesting to see how guys stacked up and whether their defensive reputation held any water.

KnicksorBust
02-28-2015, 12:57 PM
I am glad they are attempting to look at different metrics of defensive ability.

The new technology developed will produce some of the most comprehensive data we have ever seen. This information only heightens the level of debate.

As a HS basketball coach I love sharing little pieces of this stuff with my players. I showed some players the Kawhi Leonard and James Harden tables and had a great discussion.

Goose17
02-28-2015, 02:43 PM
As long as people don't put too much value into these. I mean when stats like this are in their infancy there's usually more flaws than people first realise.

Chronz
02-28-2015, 03:40 PM
I haven't watched enough Clipper games to know, but is Doc still employing his old defensive schemes he used in Boston with Thibs as his defensive coordinator? They used to load up on the strong side where the ball was to force the offensive team to swing it around. Are you guys still doing that?
Its funny, I was watching last nights game with your question in mind. Its abit fuzzy now but there was a switch with DJ guarding (IIRC)Conley. It was an isolation clear out but CP3 never left the strong side box. He was active and jumping around to ensure EVERYONE knew he was there (including the opposing team+ref) and they forced either a violation or a missed shot because both DJ and CP3 knew what the other was thinking. So short answer yes, we are overloading the strong side box but thats alot more commonplace in 2015 than it was in 2008. As a result, teams have evolved away from that type of offense (except for the Lakers), and the DAntoni style has reigned supreme.
The real differentiation among PG's in most systems will always be his partners in the PnR, but its stats like this that will help us identify the best at hedging, the players who require less time to get back into position. These are the early findings but they only support most of what the public believes anyways, a good start.


If so a smart point guard would move the ball and not shoot as frequently. He would not be able to get to the rim because they are basically playing the zone with the Deandre back there ready to contest. I am also interested to see if they are taking into account how many assist the other player gets as well as turnovers forced
Agreed, one thing about the Clippers is that they play the regular season the way they would play PG's like Curry, TP, Westbrook, Lillard, Harden in the playoffs. They practice that **** all the time. Its why they've contained those players to a degree but there is a discernible difference when its CP3 out there. So even if our defense isn't ranked as highly, I feel like the fact the we never have to abandon our defensive principles is a strength that most playoff teams arent ready for. They can turn it on to a level few teams can.

Sadds The Gr8
02-28-2015, 03:48 PM
If I were to make a list of the Top defensive PGs (among top quality PGs) it'd be something like (in no order):

John Wall
Chris Paul
Kyle Lowry
Mike Conley

I think those 4 are really good defenders who are also among the Top PGs in the league.
I dont know what the analytics say but from what I've watched, Lowry doesn't deserve to be on that list.

IndyRealist
02-28-2015, 10:38 PM
As long as people don't put too much value into these. I mean when stats like this are in their infancy there's usually more flaws than people first realise.

The stats aren't usually flawed when they're done by professionals (as opposed to dabbling amateurs), they tell you exactly what they're supposed to. The flaw is in people not understanding the stats and misinterpreting what they say.

IndyRealist
02-28-2015, 10:41 PM
I'm not familiar with the Plumlee > Vuc discussion but just because there are some that oversimplify comparisons doesn't make the statistics any less relevant. If advanced statistics support an argument then I see no problem in the validity of using them. You just need to keep adding context to the argument until there is a full picture. Subjective AND objective proof.

It was actually something like a 4 paragraph analysis, but D focused on one line and refused to look at the full picture. But that's the way when people have their head stuck in the sand.

http://espn.go.com/espn/feature/story/_/id/12331388/the-great-analytics-rankings#nba

Trwood12
02-28-2015, 10:53 PM
why would i need to read it. Paul is not the best perimeter defender in the league. i'll read it later but these people are hurting there own cause

You keep saying that he's not the best, but you seem reluctant to say who you think is better. So once again, who do you think is better?

ewing
02-28-2015, 11:33 PM
You keep saying that he's not the best, but you seem reluctant to say who you think is better. So once again, who do you think is better?


I did. Multiple times, i referenced Jimmy Bulter and kawhi leonard. I'm pretty sure i also said i did not think he was the best at his position and said i would take at min Wall before him on that end of the floor.

Goose17
03-01-2015, 08:30 AM
The stats aren't usually flawed when they're done by professionals (as opposed to dabbling amateurs), they tell you exactly what they're supposed to. The flaw is in people not understanding the stats and misinterpreting what they say.

The vast majority of advanced stats are flawed, I mean look at PER for example. The stats we're accustomed to are still in their infancy and have a long way to go before they can truly measure the game with pin point accuracy. A brand new one is even more in it's infancy. Just logical that it has way more flaws than people think.

The real flaw is putting too much value into one stat.

KnicksorBust
03-01-2015, 01:11 PM
The vast majority of advanced stats are flawed, I mean look at PER for example. The stats we're accustomed to are still in their infancy and have a long way to go before they can truly measure the game with pin point accuracy. A brand new one is even more in it's infancy. Just logical that it has way more flaws than people think.

The real flaw is putting too much value into one stat.

PER is ****ing genius.

HouRealCoach
03-01-2015, 01:13 PM
Conley, Iguodala, Tony Allen, Iman Shumpert, Russell Westbrook, Kawhi Leonard are all better than him

Goose17
03-01-2015, 01:19 PM
PER is ****ing genius.

I know. And it's also heavily flawed. So much so that it's own creator has abandoned it.

IndyRealist
03-01-2015, 06:38 PM
The vast majority of advanced stats are flawed, I mean look at PER for example. The stats we're accustomed to are still in their infancy and have a long way to go before they can truly measure the game with pin point accuracy. A brand new one is even more in it's infancy. Just logical that it has way more flaws than people think.

The real flaw is putting too much value into one stat.

I think PER says exactly what it's supposed to say: measuring the popular opinion of a player's ability based on his boxscore. Where people go wrong is in thinking it actually measures how productive a player is, which it does not. PER's coefficients were chosen to confirm to preconceived notions, it's just that years of ESPN brainwashing has convinced the masses otherwise.

People think TS% is nonsense because it doesn't take into account the player's role, his teammates around him, or the volume he shoots. What they don't realize is that it's not MEANT to account for those things. TS% is a tool, meant to be used in conjunction with other metrics, watching the games, and video analysis. That's misunderstanding on the part of the observer, not a flaw in the stat itself. TS% tells you exactly what it's supposed to.

KnicksorBust
03-02-2015, 02:09 PM
I think PER says exactly what it's supposed to say: measuring the popular opinion of a player's ability based on his boxscore. Where people go wrong is in thinking it actually measures how productive a player is, which it does not. PER's coefficients were chosen to confirm to preconceived notions, it's just that years of ESPN brainwashing has convinced the masses otherwise.

People think TS% is nonsense because it doesn't take into account the player's role, his teammates around him, or the volume he shoots. What they don't realize is that it's not MEANT to account for those things. TS% is a tool, meant to be used in conjunction with other metrics, watching the games, and video analysis. That's misunderstanding on the part of the observer, not a flaw in the stat itself. TS% tells you exactly what it's supposed to.

You don't think that in general PER can accurately show the effectiveness of a player? I'm not saying it's perfect. I'm not saying it's right every time. But overall, you don't think it shows the value of a player?

Chronz
03-02-2015, 03:16 PM
You don't think that in general PER can accurately show the effectiveness of a player? I'm not saying it's perfect. I'm not saying it's right every time. But overall, you don't think it shows the value of a player?
There's a thread on this you should read. Mobile can't search but I'll show you later

Goose17
03-02-2015, 04:15 PM
I think PER says exactly what it's supposed to say: measuring the popular opinion of a player's ability based on his boxscore. Where people go wrong is in thinking it actually measures how productive a player is, which it does not. PER's coefficients were chosen to confirm to preconceived notions, it's just that years of ESPN brainwashing has convinced the masses otherwise.

People think TS% is nonsense because it doesn't take into account the player's role, his teammates around him, or the volume he shoots. What they don't realize is that it's not MEANT to account for those things. TS% is a tool, meant to be used in conjunction with other metrics, watching the games, and video analysis. That's misunderstanding on the part of the observer, not a flaw in the stat itself. TS% tells you exactly what it's supposed to.

I see your point and I agree. But a lot of these stats are in their infancy and I think better versions will appear.

KnicksorBust
03-02-2015, 06:18 PM
There's a thread on this you should read. Mobile can't search but I'll show you later

I will check back. I have to scout tonight so I'll be stuck on my phone for a lot of the night.

IndyRealist
03-02-2015, 08:02 PM
You don't think that in general PER can accurately show the effectiveness of a player? I'm not saying it's perfect. I'm not saying it's right every time. But overall, you don't think it shows the value of a player?

From Wayne Winston, graduate of both MIT and Yale, professor at University of Houston C. T. Bauer College of Business, consultant for the Dallas Mavericks, and author of Mathletics (http://www.amazon.com/Mathletics-Gamblers-Enthusiasts-Mathematics-Basketball/dp/0691154589):


I have had a great time teaching a sports and math class to Kelley School of Business students at Indiana University. One of my ace students Paul Aynilian did a study trying to estimate John Hollingerís famous ESPN PER Ratings based on box score statistics. We found that

45.75*(Points/Minute)+22.55*(Rebounds/Minute)+32.8*(Assists/Minute)+58.2*(Steals/Minute)-48.65*(Turnovers/Minute) -39.73*(Missed FGís per minute) -20.6*(Missed FT per minute)+38.37*(Blocked Shots Per Miute)-18.68*(Personal Fouls Per Minute)

explains over 99% of the variation in this seasonís PER rankings and is off by an average of .37 in estimating the PER of the top 200 NBA players whose stats are on Yahoo.com. So basically our simple formula virtually duplicates the PER rating without a lot of mumbo jumbo.

The disturbing thing about these weights is that if an NBA player shot 33 .33% (1-3) then the more shots they take the higher their PER because shooting 1 for 3 gives you a net contribution of 2(45.75)-2(39.73)>0!! Clearly this is bad because a 33% shooter is not a good shooter and with these weights the more shots a bad shooter takes, the higher his PER rating.
http://waynewinston.com/wordpress/?p=1326

KnicksorBust
03-02-2015, 11:39 PM
null

I knew this already. This doesnt answer my question

IndyRealist
03-03-2015, 12:15 PM
I knew this already. This doesnt answer my question

Amost every player shoots over 33% on 2pt FG. So almost every player raises his PER just by shooting more. Not by shooting well, just taking bad shots. So how can you think that it accurately shows effectiveness?

All metrics agree Lebron, Paul, Curry, etc are great players. Where you need metrics is for players like Jamal Crawford pre-Clippers, where there are legitimate questions about people's observations of him. Jamal Crawford is the exact kind of player public perception overrates, and the exact kind of player PER overrates. Why? Because he takes a crap ton of bad shots and does nothing else (and refuses to play PG, but that's another thread). But because he has the ball a lot, occasionally puts up 40pt games, and gets on highlight reels.

Conversely PER underrates players like Tony Allen, who rarely takes shots but has a HUGE impact on the game. And no, it's a myth that all metrics underrate defensive players, it's just that people think PER is "all metrics". PER overvalues shot taking (not making) and thus undervalues everything else, like efficiency, rebounding, turnovers, etc. And it overvalues shot taking by such a wide margin that it makes detrimental players look decent.

RiLoc
03-03-2015, 01:42 PM
I dislike PER so much. Every time I hear an ESPN analyst use PER it makes me want to facedesk. Poorly masking the reality that to most ESPN analysts PER is just a magic number that furthers their reputation as a well rounded basketball analyst that is knowledgeable about advanced statistics, when they don't know what goes into PER, the weaknesses of PER and what the difference is between PER and other all-in-one ratings.

KnicksorBust
03-03-2015, 01:50 PM
Amost every player shoots over 33% on 2pt FG. So almost every player raises his PER just by shooting more. Not by shooting well, just taking bad shots. So how can you think that it accurately shows effectiveness?


Because if I want to see if a player is shooting efficiently I will check his TS% not his PER.


All metrics agree Lebron, Paul, Curry, etc are great players. Where you need metrics is for players like Jamal Crawford pre-Clippers, where there are legitimate questions about people's observations of him. Jamal Crawford is the exact kind of player public perception overrates, and the exact kind of player PER overrates. Why? Because he takes a crap ton of bad shots and does nothing else (and refuses to play PG, but that's another thread). But because he has the ball a lot, occasionally puts up 40pt games, and gets on highlight reels.

Jamal Crawford pre-Clippers had a PER consistently between 14-16. That's basically league average. How is he being overrated?


Conversely PER underrates players like Tony Allen, who rarely takes shots but has a HUGE impact on the game. And no, it's a myth that all metrics underrate defensive players, it's just that people think PER is "all metrics". PER overvalues shot taking (not making) and thus undervalues everything else, like efficiency, rebounding, turnovers, etc. And it overvalues shot taking by such a wide margin that it makes detrimental players look decent.

Tony Allen's career PER is 14.4. First of all, in your own chosen example, PER rates them as about equal players. Unless you think Tony Allen, who basically has played half of games his entire career, is significantly more valuable than Jamal Crawford this proves very little. Second of all, I've never seen anyone use PER to measure defense. Clearly in any comparison of the two players you would have to describe the advantage Tony Allen brings on that end of the court. This is where a new metric for defense would be valuable. Hence, why this article that I posted on page 1 is so interesting. If we could compare the opponent's FGs and defensive efficiency charts that would be a huge step forward in illustrating who has the better defensive impact. Then we can use things like raw stats (points, rebounds, assists, etc.) + advanced statistics for efficiency (like PER and TS%) and now their defensive charts to make a full objective argument while adding in your own subjective context from what we see watching them play. That's how you raise the level of debate.

KnicksorBust
03-03-2015, 01:53 PM
I dislike PER so much. Every time I hear an ESPN analyst use PER it makes me want to facedesk. Poorly masking the reality that to most ESPN analysts PER is just a magic number that furthers their reputation as a well rounded basketball analyst that is knowledgeable about advanced statistics, when they don't know what goes into PER, the weaknesses of PER and what the difference is between PER and other all-in-one ratings.

So what can they talk about? What stat isn't flawed? What opinion isn't subjective?

SPURSFAN1
03-03-2015, 01:54 PM
https://www.youtube.com/watch?v=F22t7CxmKUM
kawhi wins this hands down.

Chronz
03-03-2015, 02:01 PM
Send him our last debate on this indy, would save some time and then kob could argue my points for me with much more vigor. Im too novice at this

IndyRealist
03-03-2015, 04:24 PM
.......
Yes, Jamal Crawford is a substantially below average player. The ONE thing he did on the court, scoring, he did at a below average effectiveness. But he did a lot of it, so his PER said he was average.

I chose Tony Allen because his effectiveness is an order of magnitude large than PER indicates, compared to Crawford who is substantially worse than his "average" PER.

As I stated, there are existing metrics that rate players like Tony Allen highly despite his raw box score numbers. But these metrics do not conform to popular opinion, so they are dismissed, while PER is completely inaccurate describing player effectiveness but is widely accepted because it confirms what people already think.

I'm not sure what you're arguing, because you seem to try to use PER ratings to justify the use of PER, which is circular logic. At no point did you address that PER rewards crappy shooters, which by definition would make it a poor stat to judge offense ( since you acknowledge that PER doesn't measure defense). So what good is it? It measures what the layperson thinks of a player's stat line, because that's how the coefficients were designed: to mimic uninformed opinion.

PatsSoxKnicks
03-03-2015, 06:01 PM
From Wayne Winston, graduate of both MIT and Yale, professor at University of Houston C. T. Bauer College of Business, consultant for the Dallas Mavericks, and author of Mathletics (http://www.amazon.com/Mathletics-Gamblers-Enthusiasts-Mathematics-Basketball/dp/0691154589):

http://waynewinston.com/wordpress/?p=1326

He doesn't consult with the Mavs anymore...

Anyways, regarding the article, I still have to read it but at a quick glance, it seems like Vantage already had this.

PatsSoxKnicks
03-03-2015, 06:03 PM
Yes, Jamal Crawford is a substantially below average player. The ONE thing he did on the court, scoring, he did at a below average effectiveness. But he did a lot of it, so his PER said he was average.

I chose Tony Allen because his effectiveness is an order of magnitude large than PER indicates, compared to Crawford who is substantially worse than his "average" PER.

As I stated, there are existing metrics that rate players like Tony Allen highly despite his raw box score numbers. But these metrics do not conform to popular opinion, so they are dismissed, while PER is completely inaccurate describing player effectiveness but is widely accepted because it confirms what people already think.

I'm not sure what you're arguing, because you seem to try to use PER ratings to justify the use of PER, which is circular logic. At no point did you address that PER rewards crappy shooters, which by definition would make it a poor stat to judge offense ( since you acknowledge that PER doesn't measure defense). So what good is it? It measures what the layperson thinks of a player's stat line, because that's how the coefficients were designed: to mimic uninformed opinion.

So are you basically trying to argue Usage doesn't matter? Good luck with that. Hope you aren't one of those WoW guys.

PatsSoxKnicks
03-03-2015, 06:06 PM
I think PER says exactly what it's supposed to say: measuring the popular opinion of a player's ability based on his boxscore. Where people go wrong is in thinking it actually measures how productive a player is, which it does not. PER's coefficients were chosen to confirm to preconceived notions, it's just that years of ESPN brainwashing has convinced the masses otherwise.

People think TS% is nonsense because it doesn't take into account the player's role, his teammates around him, or the volume he shoots. What they don't realize is that it's not MEANT to account for those things. TS% is a tool, meant to be used in conjunction with other metrics, watching the games, and video analysis. That's misunderstanding on the part of the observer, not a flaw in the stat itself. TS% tells you exactly what it's supposed to.

Also, this is blatantly false and shows you have no clue about how a stat is developed. The way to calculate PER is out there for everyone to see: http://www.basketball-reference.com/about/per.html

Read Hollinger's Basketball Prospectus book on how he derives the calculation- everything has a reason. You may disagree with the reason but there's a reason behind it.

As for the accuracy of PER, obviously there are better metrics out there and I'm sure Hollinger would even tell you that. Regardless, no one is using any one number metrics for player evaluation in the NBA.

IndyRealist
03-03-2015, 07:43 PM
Also, this is blatantly false and shows you have no clue about how a stat is developed. The way to calculate PER is out there for everyone to see: http://www.basketball-reference.com/about/per.html

Read Hollinger's Basketball Prospectus book on how he derives the calculation- everything has a reason. You may disagree with the reason but there's a reason behind it.

As for the accuracy of PER, obviously there are better metrics out there and I'm sure Hollinger would even tell you that. Regardless, no one is using any one number metrics for player evaluation in the NBA.

The weights are arbitrary. There is no statistical backing for his choices, no analysis of relative weights justifying his reasoning. He picked them out of the air, based on what he felt was correct (at the time). That's not how you derive a statistical model. Regardless of how you sugar coat it, the only logic for PER is "this agrees with my preconceived notions".

IndyRealist
03-03-2015, 07:50 PM
So are you basically trying to argue Usage doesn't matter? Good luck with that. Hope you aren't one of those WoW guys.

I never said usage doesn't matter, I said PER rewards crappy shooters for taking more crappy shots. Usage is neither a positive nor a negative, any more than minutes played is. It's context. I'm not sure how you could get to "usage doesn't matter" from "crappy shooters shouldn't be rewarded for taking crappy shots".

PatsSoxKnicks
03-04-2015, 01:17 AM
I never said usage doesn't matter, I said PER rewards crappy shooters for taking more crappy shots. Usage is neither a positive nor a negative, any more than minutes played is. It's context. I'm not sure how you could get to "usage doesn't matter" from "crappy shooters shouldn't be rewarded for taking crappy shots".

You didn't really answer my question though- are you one of those WoW guys? Because those guys have no clue what they're talking about and there's a reason none of them have any experience with any NBA analytics/front office staffs.

The reason I ask is because you seem to be leaning in their direction based on what you're saying (which I'm not exactly sure what it is you're saying with Usage is neither positive or negative- you must be aware of the debate)- Usage clearly matters in the whole Usage-Efficiency debate. This has been proven over and over and over again and even the WoW guys essentially know this to be true but they continue to go against it because it's how they keep themselves relevant.

PatsSoxKnicks
03-04-2015, 01:27 AM
The weights are arbitrary. There is no statistical backing for his choices, no analysis of relative weights justifying his reasoning. He picked them out of the air, based on what he felt was correct (at the time). That's not how you derive a statistical model. Regardless of how you sugar coat it, the only logic for PER is "this agrees with my preconceived notions".

Have you read Hollinger's Basketball Prospectus books? Seems like you haven't. It's not much different from how Pelton derived his WARP formula or Oliver derived his ORtg formulas etc. The weights might not be correct- and they probably are out of date for sure- but he outlined his logic for the formula in his books, which you clearly haven't read.

I'm not even really arguing it's accuracy because it's not the greatest in terms of predictions or even close- most know that in the analytic community but what you're arguing is just ridiculous and nonsensical. There was a clear logic behind his weights at the time and he makes that clear in how he outlines it in his book- which you haven't read I'm sure. Sure stats like VORP and RPM are clearly better but you don't need to bash the author of said stat who has contributed so much to the basketball analytic community. He outlined his logic clearly in his book. Disagree with it if you want but don't say he came up with it out of thin air- that's what statheads hate the most.

KnicksorBust
03-04-2015, 01:49 PM
The weights are arbitrary. There is no statistical backing for his choices, no analysis of relative weights justifying his reasoning. He picked them out of the air, based on what he felt was correct (at the time).

:rolleyes: Not worth continuing this debate.

KnicksorBust
03-04-2015, 01:49 PM
Have you read Hollinger's Basketball Prospectus books? Seems like you haven't. It's not much different from how Pelton derived his WARP formula or Oliver derived his ORtg formulas etc. The weights might not be correct- and they probably are out of date for sure- but he outlined his logic for the formula in his books, which you clearly haven't read.

I'm not even really arguing it's accuracy because it's not the greatest in terms of predictions or even close- most know that in the analytic community but what you're arguing is just ridiculous and nonsensical. There was a clear logic behind his weights at the time and he makes that clear in how he outlines it in his book- which you haven't read I'm sure. Sure stats like VORP and RPM are clearly better but you don't need to bash the author of said stat who has contributed so much to the basketball analytic community. He outlined his logic clearly in his book. Disagree with it if you want but don't say he came up with it out of thin air- that's what statheads hate the most.

Hey Psk! :) How's it going? You a big shot yet or what?