yardbarker

OWS distribution reveals championship style pg play

Moderators: tsherkin, Doctor MJ, CellarDoor

Post#1 OWS distribution reveals championship style p
Tue Jul 24, 2012 7:56 am by trainwreckog

Offensive Win Share (OWS) stats of top pg's show that in the last 30 years of basketball history, it is not an optimal style of team play to have a pg with an OWS that is very high and/or materially higher than anyone else on the team. PG's that have very high OWS and/or an OWS that is distinctly the highest on their team do not make the NBA Finals, and almost never even make the conference finals.

Contrastingly, PG's on teams that HAVE made the NBA Finals do NOT have high OWS figures, and have OWS figures that are either not the highest on their team (Isiah, Stockton), or barely the highest on their team (Tony Parker).

Additionally, as would be expected, on teams that made the Finals, the distribution of OWS was not only weighted less towards the PG, but was more balanced with more teammates having meaningful/material OWS figures than teams with PG's that dominated the OWS distribution (and of course, didn't make the Finals).

This leads me to believe that the modern style of pg play where the pg increasingly makes most of the decisions on the offensive end and teammates wait to get "dimed" to rather than make a share of the important decisions themselves, is NOT an optimal style that gives the team a chance to play at a championship level. Players like Chris Paul, Steve Nash and Derrick Rose fit this profile, and unlike their aforementioned championship-level counterparts, have very high OWS figures and a less balanced distribution of OWS on their respective teams.

To summarize, the OWS statistics show that teams with a PG that does NOT dominate the OWS distribution are capable of making the NBA Finals, while teams that have a PG with a very high OWS simply do not make the NBA Finals. The only exception to this rule was Magic Johnson, however, he was much more than just a PG, who operated a high proportion of possessions from the post. Which leads to another point - teams make the NBA Finals all the time with a player that dominates the OWS distribution, IF he is a dominant post player, i.e. Magic/Kareem, MJ, Shaq, Hakeem, Pau Gasol.

Interestingly, Phil Jackson's teams that never had real point guards and where the primary differentiating aspect of the triangle offense was offense initiation via the pass (rather than dribbling), always had at least 9 players with material OWS figures (0.5 or higher).

Any comments on these musings? Feel free to post objections or point out aspects that are missing.
trainwreckog
Banned User
Posts: 752
And1: 103
Joined: Jul 13, 2012
Top

Post#2 Re: OWS distribution reveals championship sty
Tue Jul 24, 2012 11:01 am by trainwreckog

someone correct me if i'm wrong here, but ows is = (marginal offense/marginal points per win), which in layman's terms, Player X OWS = (offense for player X over and above what the average nba player does in the same amount of possessions that player X had)/(league average scoring per game, pace-adjusted for Player X's team).

so when i look at horace grant's ows figure and i see it is higher than scottie pippen's, i figure his number of possessions is in the same vicinity as scottie's, and he is producing points in those possessions at an absurd rate, a higher rate than scottie is. given that horace was not that offensively skilled or competent, one would have to assume the offense suited his game and/or he learned the offense/was able to operate within the offense very well. given that phil jackson's offense empowered players more than a pg-centric offense, this is not surprising.

then when i compare horace's ows figure to that of deandre jordan and see that of course it is much higher, one can only assume that horace's number of possessions was much higher than deandre's, yet he was still producing points in those possessions at the same (or likely higher) clip than deandre. as we know, deandre pretty much only gets dimed to, whereas horace either had the opportunity to make more meaningful offensive plays within the offense, or, within the offense was able to receive the ball that much more often than deandre. i would imagine that most of deandre's points came off of passes from paul, whereas horace's must have come from a wider, more balanced range of players. again, given the differences in phil jackson's offense and del negro's, this is not surprising.
trainwreckog
Banned User
Posts: 752
And1: 103
Joined: Jul 13, 2012
Top

Post#3 Re: OWS distribution reveals championship sty
Tue Jul 24, 2012 1:49 pm by ElGee

The problem with your conclusion is that you are working from a faulty premise. Win Shares does not equal player quality, nor does it accurately measure what you are trying to measure.

Win Shares underrates distributing point guards, because the classic box score that it is based on does not measure where they are contributing well. As a result, if you had a great distributing point guard on a bad team, he would come out as the WS leader. Great distributing PG's on solid teams will not come out as the OWS leader:

eg Nash behind Amare in 2005

I'm also wondering what teams you are looking at to make these conclusions. I see stuff like:

D. Barros 95 -- second leading OWS player is Willie Burton, at 1.9 OWS. They win 24 games as Philadelphia stinks AND Barros is not a distributing PG.

G. Payton 96 -- Only the leader by technicality as Schrempf accrues more per 48...plus Payton isn't a classic distributing PG. The team also wins 64-games and loses to the GOAT-level Bulls in 6. Hardly a damning occurrence.

G. Payton 97 -- Payton leads again, the 7 SRS Sonics lose in 7 to Houston (no HCA)

S. Marbury 01 - 4.3 OWS ahead of second best Aaron Williams. Again, not a distributing PG. And again, a terrible team.

You get the idea...I won't go through them all in detail, but the list of PG's here is something like

K. Anderson
B. Armstrong
D. Augustin
D. Barros
C. Billups
M. Blaylock
T. Brandon
D. Brown
J. Calderon
S. Cassell
B. Davis
S. Francis
A. Hardaway
T. Hardaway
D. Harper
D. Harris
J. Humphries
M. Jackson
K. Johnson
M. Johnson
J. Kidd
T. Lawson
F. Lever
S. Marbury
A. Miller
E. Murdock
S. Nash
T. Parker
C. Paul
G. Payton
T. Porter
M. Price
D. Rose
S. Skiles
K. Smith
J. Stockton
D. Stoudamire
R. Strickland
I. Thomas
D. Williams
G. Williams


I'm sure there are plenty more in even more horrible situations, but that list seems close enough to comprehensive to make the point. Look at the different types of PG's leading their teams in WS:

Distributors-First:
D. Augustin
T. Brandon
J. Calderon
S. Cassell
B. Davis
T. Hardaway
D. Harper
M. Jackson
K. Johnson
M. Johnson
J. Kidd
A. Miller
S. Nash
C. Paul
M. Price
S. Skiles
J. Stockton
D. Williams

Scorer-Distributors:
K. Anderson
C. Billups
M. Blaylock
S. Francis
A. Hardaway
D. Harris
J. Humphries
T. Lawson
F. Lever
S. Marbury
T. Parker
G. Payton
T. Porter
D. Rose
D. Stoudamire
R. Strickland
I. Thomas
G. Williams

Shooters:
B. Armstrong
D. Barros
D. Brown
E. Murdock
K. Smith

You're never going to find a "shooting" PG -- as in, PG by name only -- leading a decent offensive team in win shares unless that player has like a 70% TS%. The scoring group is going to generate good win shares often with their own stats, versus their distribution. But let's jump inside that group:

-G. Williams was on a very good team. They fell apart without him in 1981, and of course, won a championship in 1979 in which he led the team in OWS.
-I. Thomas was on some not-so-hot teams in Detroit when he led in OWS. As he aged, and shot less, the Pistons went to 3 consecutive Finals and won two championships.
-Strickland, Stoudamire, Marbury, Humphries, Harris and Francis were just on terrible, terrible teams. They led due to their own scoring essentially.
-Rose, Porter, Payton, Penny, Blaylock and Billups have 4 Finals appearances in teams leading in WS and 1 title. That does not include all teams but Mookie's making the CF's at least one more time.

We haven't reached the distributor group and so far we have
-multiple championships
-multiple Finals appearances
-PG teams doing very well in years they didn't lead because another star came on board

For the distributors,
-Magic Johnson has 4 championships and 8 Finals appearances by himself.
-Jason Kidd made the Finals in 2003
-A bunch of CF's appearances. (Nash, Williams, Hardaway, Price per minute)

That means that since 1979, teams with a PG leading their OWS have made 13 Finals (19.1%) and won 6 championships (17.6%).

Many of the distributors leading their team in OWS are on absolutely dreadful teams, and once they pick up competent offensive players, they no longer lead the team because OWS does not track their creation. All told, I see very little basis for what you'e concluded here unless you can present some fresh evidence that I'm missing.
ElGee
Assistant Coach
Posts: 3,884
And1: 822
Joined: Mar 7, 2010
Top

Post#4 Re: OWS distribution reveals championship sty
Wed Jul 25, 2012 7:08 am by trainwreckog

.........
.........

you completely missed my point.

My point was that pg's, regardless of type, that have very high ows (over 7.0) do not make the nba finals. period. that's it.

the only exceptions are magic johnson who did it many times, and who was a dominant post/inside player. as i said in my original post - dominant inside players with very high ows, FREQUENTLY make the finals. the other exceptions are penny hardaway once in 1995, also a dominant post/inside player, and lebron james, another big pg that dominates inside - clearly a trend with big pgs that dominate inside. and of course, as i said in my original post, dominant inside players like duncan, shaq, malone, pau gasol, hakeem, kareem, etc., all have high ows figures and frequently made the nba finals.

so find me a pg other than the 3 big ones i mentioned that had a high ows figure that made the NBA Finals, and then you will prove me wrong - but you won't be able to. it is a fact that shorter, more perimeter-oriented pgs cannot make the nba finals with high ows figures - their teams MUST HAVE more a more balanced ows distribution to play at a championship level. the numbers shown later in this post prove this.

you brought up gary payton in 1996. this is an example that supports my point that i actually missed. payton's ows was not high (5.9) and not materially higher than most of his teammates - i.e. his ows did not dominate the distribution. this more balanced attack of offensive contribution to the team's wins, led to a finals berth, which is my point.

so add payton to the list of top-rated, finals-making pgs that did not have high ows and whose ows did not dominate the distribution of ows on their teams. thank you for pointing this out.

as i mentioned originally, john stockton, isiah thomas and tony parker are other examples of finals-making, top-rated pg's that did not have high ows and whose ows did not dominate the distribution of ows on their teams. i include "top-rated" in describing the pgs, because obviously guys like kenny smith never had high ows ratings.

contrastingly, guys like chris paul, steve nash, and derrick rose, have very high ows, usually above 8.0, that is materially higher than most of the players on their team. this concentration causes a lack of balance in ows across the rest of the players on the team, which results in them almost never even making the conference finals, let alone the Finals.


and the the numbers basically prove what is intuitively obvious - the game should be played from the inside out. if a pg is NOT that rare breed of big pg that can dominate on the post, they CANNOT have a high, dominant ows figure and still make the nba finals. other teammates that are more effective inside need to be empowered to make actual decisions on the offensive end (rather than just wait to get "dimed" to), which would increase their ows and allow for a more balanced ows distribution across the team, the type of ows distribution that allows a team with a top-rated, perimeter-oriented pg to play at a championship level.


maybe if i flesh out the numbers it will make my point easier to understand...

excluding magic johnson, penny and lebron, here are the distributions of ows on teams that made the nba finals with top-level pgs (since non-top-level pgs like kenny smith obviously aren't going to have high ows, and so they aren't included here).

so this list of finals-making pgs includes isiah thomas, tony parker, john stockton, and as you pointed out, gary payton.. notice the BALANCE of ows on these teams - this is what allows teams that don't have a pg that can dominate inside ala magic, penny, or lebron to still play at a high enough level to make the nba finals. and contrast these balanced championship-type ows distributions with chris paul's teams listed below them:


1988-1989 Pistons OWS (isiah's ows is not high and far from dominating the distribution - 10 guys over 1.0, six guys over 3.0)

1 Joe Dumars...............4.8
2 Dennis Rodman..........4.1
3 Bill Laimbeer.............4.0
4 Adrian Dantley...........4.0
5 Vinnie Johnson..........3.4
6 Isiah Thomas..............3.1
7 Mark Aguirre............. 1.7
8 Rick Mahorn..............1.5
9 James Edwards...........1.3
10 John Salley...............1.1


1997-1998 Jazz OWS (stockton ows is not high and does not dominate the distribution - five guys over 3.8, 10 guys over 0.3, seven guys over 1.5)

1 Karl Malone..............12.1
2 Jeff Hornacek............7.4
3 John Stockton...........6.4
4 Adam Keefe...............4.8
5 Bryon Russell.............3.8
6 Shandon Anderson......2.9
7 Howard Eisley.............1.7
8 Antoine Carr...............0.8
9 Greg Foster................0.3
10 Chris Morris..............0.3


1996 seattle supersonics OWS distribution: (payton ows is not high and does not dominate the distribution)

1 gary payton...............5.9
2 hersey hawkins..........5.3
3 shawn kemp..............5.3
4 detlef schrempf.........5.2
5 sam perkins...............3.4
6 vincent askew........... 2.2
7 nate mcmillan.............1.0
8 ervin johnson.............0.5


2011-2012 Spurs OWS (this is not a Finals year for them, but look at any spurs team that made the finals and you will see the same balanced distribution - i only put this data up because i already had it previously and i was lazy to go get the other years - anyway, notice how tony parker's ows is not high and does not dominate the distribution.. this distribution for the spurs is similar to any other year)

1 Tony Parker...............5.5
2 Manu Ginobili.............3.4
3 Kawhi Leonard............3.1
4 Tim Duncan................3.0
5 Tiago Splitter..............3.0
6 Matt Bonner...............2.7
7 Danny Green...............2.5
8 DeJuan Blair................2.2
9 Gary Neal...................1.6
10 Richard Jefferson......1.4
11 Patrick Mills..............0.6
12 Boris Diaw.................0.5



NOW COMPARE THESE AFOREMENTIONED BALANCED OWS DISTRIBUTIONS, TO TEAMS CHRIS PAUL WAS ON:


Chris Paul's Team 2007-2008 OWS (notice how high Paul's OWS is and how it dominates the distribution - only five other guys are even over 1.0)

1 Chris Paul...................13.2
2 Tyson Chandler.............5.9
3 Peja Stojakovic.............5.8
4 David West....................4.3
5 Morris Peterson............2.1
6 Bobby Jackson..............0.9
7 Julian Wright................0.7
8 Ryan Bowen.................0.6
9 Rasual Butler................0.3


2008-2009 CP3 Team's OWS (this year was really unbalanced.. Paul dominating the distribution again with a very high ows, only seven guys total over 0.3, only three guys over 2.0)

1 Chris Paul..................13.3
2 David West..................4.2
3 Rasual Butler...............2.2
4 Peja Stojakovic............1.9
5 James Posey................1.7
6 Tyson Chandler............1.7
7 Antonio Daniels............0.6


2011-2012 Clippers OWS (Paul dominating the distribution once again, only seven guys total over 0.3, only three guys over 3.0)

1 Chris Paul....................10.4
2 Blake Griffin..................6.6
3 DeAndre Jordan.............3.2
4 Randy Foye....................2.4
5 Mo Williams....................2.2
6 Chauncey Billups............1.5
7 Caron Butler..................1.2

So again, my overriding point is that pg's with very high ows like chris paul, steve nash, and derrick rose, that do not also dominate from the post, CANNOT make the nba finals. for these types of short pgs to have a championship-level team, their ows needs to go down, thus making the ows distribution on their team more balanced, just like the other short pgs that have made the nba finals have done

there has never been a pg in the last 30+ years other than magic johnson, penny hardaway, and lebron james (who were all big pgs that doubled as a dominant post players) that has made the nba finals with a very high ows figure. never.

Clearly, with a short pg that cannot dominate inside/from the post, a more balanced OWS distribution is a more optimal style of basketball, where the decision-making is not concentrated so heavily with the pg, allowing other players to contribute more to wins on the offensive end.
trainwreckog
Banned User
Posts: 752
And1: 103
Joined: Jul 13, 2012
Top

Post#5 Re: OWS distribution reveals championship sty
Wed Jul 25, 2012 3:07 pm by ElGee

(1) Magic Johnson was not a "dominant post/inside player." That's just wrong.
(2) Penny was not one either, although he would at least post up small guards more frequently than Magic.
(3) LeBron James didn't move into the post until quite recently.

trainwreck wrote:you completely missed my point.

My point was that pg's, regardless of type, that have very high ows (over 7.0) do not make the nba finals. period. that's it.


I thought you said this:

trainwreck wrote:This leads me to believe that the modern style of pg play where the pg increasingly makes most of the decisions on the offensive end and teammates wait to get "dimed" to rather than make a share of the important decisions themselves, is NOT an optimal style that gives the team a chance to play at a championship level.


???

so find me a pg other than the 3 big ones i mentioned that had a high ows figure that made the NBA Finals, and then you will prove me wrong - but you won't be able to. it is a fact that shorter, more perimeter-oriented pgs cannot make the nba finals with high ows figures - their teams MUST HAVE more a more balanced ows distribution to play at a championship level. the numbers shown later in this post prove this.


Terry Porter 1992
Chauncey Billups 2004
Chauncey Billups 2005
John Stockton 1997 (and 98 per minute)

Did you even read my post? :D

contrastingly, guys like chris paul, steve nash, and derrick rose, have very high ows, usually above 8.0, that is materially higher than most of the players on their team. this concentration causes a lack of balance in ows across the rest of the players on the team, which results in them almost never even making the conference finals, let alone the Finals.


Have you considered that defense has something to do with the success of a team??

and the the numbers basically prove what is intuitively obvious - the game should be played from the inside out. if a pg is NOT that rare breed of big pg that can dominate on the post, they CANNOT have a high, dominant ows figure and still make the nba finals.


Actually that was just proved to be patently false. Magic Johnson alone makes the argument false.

Do you have any idea what the offensive performance of Steve Nash's teams have been like in the playoffs?

So again, my overriding point is that pg's with very high ows like chris paul, steve nash, and derrick rose, that do not also dominate from the post, CANNOT make the nba finals.


But the point has been proven false, and all you've done is cherry-picked a few examples in which they had bad teammates. That's it. 3 "pro" and 3 "con" does not make a proof. :/
ElGee
Assistant Coach
Posts: 3,884
And1: 822
Joined: Mar 7, 2010
Top

Post#6 Re: OWS distribution reveals championship sty
Wed Jul 25, 2012 3:38 pm by trainwreckog

ElGee wrote:(1) Magic Johnson was not a "dominant post/inside player." That's just wrong. yes, he was. you just don't remember - i doubt you remember much of magic saying something like this. MOST of the lakers possessions in the half court went through magic on the post. see youtube clip below if you can't recall

(2) Penny was not one either, although he would at least post up small guards more frequently than Magic. both of what you said here is not true - penny WAS dominant on the post, but he DIDN'T post up more often than magic. do you even remember magic johnson and how he played? most of the lakers possessions he was operating as point foward from the post.. http://www.youtube.com/watch?v=V7LY1zSfeRY

(3) LeBron James didn't move into the post until quite recently. he still posted up via bully-ball a fair bit, and being dominant inside is not just posting up - it's rebounding from the guard spot and scoring inside in other ways (other than posting) more efficiently than the small gaurds.

trainwreck wrote:you completely missed my point.

My point was that pg's, regardless of type, that have very high ows (over 7.0) do not make the nba finals. period. that's it.
this is a typo - meant to say 8.0 because that is the lowest ows of chris paul, nash and rose using their best seasons, as i mention later in my post below.

I thought you said this:

trainwreck wrote:This leads me to believe that the modern style of pg play where the pg increasingly makes most of the decisions on the offensive end and teammates wait to get "dimed" to rather than make a share of the important decisions themselves, is NOT an optimal style that gives the team a chance to play at a championship level.
this is very true, but unless you played you won't understand it so we can agree to disagree on it.
???

so find me a pg other than the 3 big ones i mentioned that had a high ows figure that made the NBA Finals, and then you will prove me wrong - but you won't be able to. it is a fact that shorter, more perimeter-oriented pgs cannot make the nba finals with high ows figures - their teams MUST HAVE more a more balanced ows distribution to play at a championship level. the numbers shown later in this post prove this.


Terry Porter 1992
Chauncey Billups 2004
Chauncey Billups 2005
John Stockton 1997 (and 98 per minute) - it appears you found one exception, and it happens to be the guy that does the least ball-pounding of any guard mentioned in the thread

Did you even read my post? :D

contrastingly, guys like chris paul, steve nash, and derrick rose, have very high ows, usually above 8.0, that is materially higher than most of the players on their team. this concentration causes a lack of balance in ows across the rest of the players on the team, which results in them almost never even making the conference finals, let alone the Finals.


Have you considered that defense has something to do with the success of a team?? my point was that (other than 1 season by john stockton that you found) pg's with high ows rates (over 8.0) that dominate their team's ows distribution cannot make the nba finals, unless they are the rare big guard that can be abnormally effective inside - this point is still very valid, other than the one exception you found of john stockton, you have not disproved it; and not coincidentally, the one exception you found is the pg that plays the MOST differently from the modern small, pg that wastes more dribbles and shot clock than ever before (the defense has a lot less time to make a mistake the more the pg is pounding the ball away on the perimeter, and it is harder to defend a team where there is a greater balance of players with the ball in their hands making more of the decisions than just the pg pounding the ball away calling all the shots; empirical evidence proves this).

and the the numbers basically prove what is intuitively obvious - the game should be played from the inside out. if a pg is NOT that rare breed of big pg that can dominate on the post, they CANNOT have a high, dominant ows figure and still make the nba finals.


Actually that was just proved to be patently false. Magic Johnson alone makes the argument false. this makes zero sense whatsoever. not sure what logic you are using here - magic would make the argument absolutely true. go back and watch some magic tape. he was on the post on most of the lakers half-court possessions.

Do you have any idea what the offensive performance of Steve Nash's teams have been like in the playoffs? but they get shut down in the clutch on the regular. that's why they can't break through. against the top defensive teams, they easily adjust to a pg that dominates all the offensive action and that is making all the material offensive decisions - nash's and paul's playoff failures are the very basis for my point about the the need for a slighly more balanced ows distribution for their teams to be optimally effective and harder to defend/less predictable, especially down the stretch of games. it is much harder to defend a team that has a wider range and better balance of players making decisions on the offensive end - and all the champions have this characteristic - this is the biggest reason nash's teams always come up short in the playoffs. they always come up short, and he's had a long career and a ton of chances with teams that are talented.

So again, my overriding point is that pg's with very high ows like chris paul, steve nash, and derrick rose, that do not also dominate from the post, CANNOT make the nba finals.


But the point has been proven false, and all you've done is cherry-picked a few examples in which they had bad teammates. That's it. 3 "pro" and 3 "con" does not make a proof. :/
not true, you found 1 example in 30 years that disproves my point, and it happens to be the most old-school, non-dribble-happy pg in the history of the game. the one exception you found is the pg that plays the MOST differently from the modern pg that has the bad habit of wasting dribbles and shot clock and not empowering their teammates to make any decisions, which makes it easier on the defense when the defense feels like defending hard, like down the stretch of a game in the playoffs - THE OWS DATA I POSTED MAKES IT OBVIOUS - THE CHAMPIONSHIP-LEVEL FINALS TEAMS HAVE BALANCE, WHILE PAUL AND NASH'S TEAMS DONT.
trainwreckog
Banned User
Posts: 752
And1: 103
Joined: Jul 13, 2012
Top

Post#7 Re: OWS distribution reveals championship sty
Wed Jul 25, 2012 8:10 pm by kabstah

contrastingly, guys like chris paul, steve nash, and derrick rose, have very high ows, usually above 8.0, that is materially higher than most of the players on their team. this concentration causes a lack of balance in ows across the rest of the players on the team, which results in them almost never even making the conference finals, let alone the Finals.

Chauncey made the Finals in 2005 with an OWS > 8, as well as multiple conference finals. Actually Nash and Rose have both made conference finals as well. Speaking of Nash and conference finals, the last time the Suns were there they put up over 109 points per game with an ORTG of like 120. This was against a Lakers team that was 4th in the league in defense with a DRTG under 104. To suggest that the Suns lost because their offense somehow wasn't good enough is, to put it succinctly, dumb.
kabstah
Rookie
Posts: 1,022
And1: 42
Joined: Feb 11, 2009
Top

Post#8 Re: OWS distribution reveals championship sty
Thu Jul 26, 2012 12:03 am by ElGee

OK first of all stop yelling. Stop typing in bold. Try and have a discussion about basketball without feeling the need to be so loud.

Second, you don't know what you are talking about here. Trying to claim Magic Johnson and LeBron James were post players for the hearts of their career is whacky. Ignoring defense is whacky. Believing that Steve Nash's teams struggled in the clutch is just 100% wrong. Believing that they struggled on offense in the PS is 100% wrong.

There's nothing wrong with not knowing this, but this is a forum to discuss statistical analysis (and often, theoretical ideas) and constructive discourse is welcome -- you making a bunch of false claims isn't going to help anyone with anything. You ignoring someone's teammates and talking about "distribution" is absolutely whacky. You literally bolding and caps-locking the same thing doesn't change anything. I've told you what's wrong with your conclusion...you can either try to understand, or continue to plug your ears and scream.

Cheers.
ElGee
Assistant Coach
Posts: 3,884
And1: 822
Joined: Mar 7, 2010
Top

Post#9 Re: OWS distribution reveals championship sty
Thu Jul 26, 2012 7:42 am by trainwreckog

ElGee wrote:OK first of all stop yelling. Stop typing in bold. Try and have a discussion about basketball without feeling the need to be so loud.

Second, you don't know what you are talking about here. Trying to claim Magic Johnson and LeBron James were post players for the hearts of their career is whacky. Ignoring defense is whacky. Believing that Steve Nash's teams struggled in the clutch is just 100% wrong. Believing that they struggled on offense in the PS is 100% wrong.

There's nothing wrong with not knowing this, but this is a forum to discuss statistical analysis (and often, theoretical ideas) and constructive discourse is welcome -- you making a bunch of false claims isn't going to help anyone with anything. You ignoring someone's teammates and talking about "distribution" is absolutely whacky. You literally bolding and caps-locking the same thing doesn't change anything. I've told you what's wrong with your conclusion...you can either try to understand, or continue to plug your ears and scream.

Cheers.



so in 30 years, that is 3 seasons where a regular-sized pg was able to dominate the ows distribution for his team and make the finals - otherwise, all the other finals-making, regular-sized pgs were on teams where the ows distribution was balanced - that would be 27 of 30 years where the teams of regular-sized pgs made the finals and had a balanced distribution of ows. and yes, i suppose that would make it 3 out of 30 years where a regular-sized pg was able to dominate the ows distribution and make the nba finals. hmmm... wonder who is right on this?

what say you?

so clearly, my primary point - that for regular-sized pg's, a pg style of play which allows for a balanced distribution of ows across the team is much more optimal than a pg style where the pg dominates the ows distribution - is indeed quite valid and backed up by the empirical evidence we have been discussing and even more supported by the paltry 3 anomalies in 30 years that you were able to dig up.

also, nobody is trying to scream. my bolding was mainly to delineate my responses from yours. i also bolded certain points because i felt they were more nuanced and not as quantitatively-based, so they would stand out and be easier to read. the game of basketball is nuanced and cannot ever be solved by stats. so rudimentary arguments such as "well, nash's teams have the highest offensive efficiency rating, blah, blah, blah...," far from explain why he has underachieved in the playoffs throughout his long career, despite all the talent he has played with.

dirk flourished much more without him.

as the data suggests about regular-sized pgs, your boy nash needs to adjust his style of play to make the Finals. it's too late now, because he won't be the primary player on a team since he's past his prime, so he won't get the same credit for a ring that he would have gotten if he had adjusted his pg style earlier in his career to a style more conducive to a balanced ows distribution, which as we can see, gives a regular-sized pg a much better chance of making the Finals...

in my opinion (and i am basing this on watching the other regular-sized pgs that HAVE made the Finals), a pg style that is more conducive for a balanced ows distribution would mean a more dribble-efficient style where nash had the ball in his hands slightly less, thus empowering his teammates on the offensive end to make a greater share of the decisions rather than stand and wait to get "dimed" to. the "stand-and-wait-to-get-dimed-to" pg style that nash and chris paul employ is such a suboptimal pg style that the top teams ultimately adjust to and defend effectively down the stretch of tight playoff games - this is not shown in most advanced stats. and while this suboptimal style is occasionally good enough to make the conference finals, as we see by the evidence, almost never the finals...... and i would argue that stockton and chauncey are two pg's that don't actually use the dribble-happy style that paul and nash use and their domination of the ows distribution was due to other factors.... but i've digressed....

cheers,
trainwreckog
Banned User
Posts: 752
And1: 103
Joined: Jul 13, 2012
Top

Post#10 Re: OWS distribution reveals championship sty
Tue Jul 31, 2012 9:26 pm by trainwreckog

..
Cliffs of this thread:

in the last 30 years, there have been 3 times where a regular-sized pg had an offensive win share (ows) above 7.5 and still made the nba finals.

exceptions to this trend are pgs that played in forward's bodies - lebron, penny, and magic. this is not surprising since it is typical for finals teams to have non-pgs that dominate their team's ows distribution with ows figures well above 7.5.

however, for regular-sized pg's that make the finals, an ows above 7.5 is quite an anomaly.
..
trainwreckog
Banned User
Posts: 752
And1: 103
Joined: Jul 13, 2012
Top

Post#11 Re: OWS distribution reveals championship sty
Wed Aug 1, 2012 11:04 am by trainwreckog

its interesting that now that nash is on board, kobe quickly lobbied (successfully of course) for the princeton offense, which in his words, allows for "more ball movement, changing sides on the floor, everybody being involved – those are championship principles. That's championship DNA."

this as opposed to nash's ball-dominant style, where ball movement via the dribble is opted for over the pass, and where he makes most of the decisions on the offensive end, while teammates wait to get "dimed" to rather than make decisions themselves.

kobe's figured it out, but he always had this figured out about the game of basketball - thanks in no small part to phil jackson's triangle offense, which also uses these principles.
trainwreckog
Banned User
Posts: 752
And1: 103
Joined: Jul 13, 2012
Top

Post#12 Re: OWS distribution reveals championship sty
Fri Aug 3, 2012 11:13 am by Nivek

trainwreck: Your point is easy to understand, but what's the...umm...point? Teams should avoid getting guys like Chris Paul if they want to reach the Finals?

Why have you chosen reaching the Finals as the criteria for success? Have you accounted for the quality of a player's teammates? Have you performed the same analysis for every other position? Are there meaningful distinctions between the results of high-OWS PGs and high-OWS players at other positions.

From what you've posted here, it looks like you've started with the conclusion and gone looking for data to support that conclusion.

Also, you're really kinda undercutting whatever it is you're trying to prove with claims that Kobe is an advocate of ball movement and "everyone being involved." He may be saying that now, but that's not how he's played the game throughout his career.

One last thing: someone mentioned defense in this thread, which is something you've ignored so far. Saying a Nash-led team failed to reach the Finals because there's something wrong with its offense is inane.
Nivek





Head Coach
User avatar
Posts: 6,861
And1: 686
Joined: Sep 29, 2010
Top

Post#13 Re: OWS distribution reveals championship sty
Fri Aug 3, 2012 7:36 pm by Doctor MJ

As Nivek said, I get what you're saying trainwreck, but you're jumping to a conclusion unsupported by your data.

I can do an analysis that shows that teams with point guards with certain jersey numbers haven't won titles. Does that mean I can assume causality? Of course not.

I like OWS as a stat, but it's not it's not precise enough to do anything like what you're trying to do with it. There are a million ways to play the game. There are good ways to rack up the stats that help your OWS, and there are not-so-good ways. You can't do an analysis like this except with a stat fine grained enough that the good and not-so-good remain distinct.

Generally speaking though, what you're doing here is just a version of the non-statistical arguments along these lines that have existed for year. The question posed is, If point guards are so important, why aren't the good ones winning titles?

The answer really isn't that complicated. Basketball is a team game, with a lot of factors, and some luck. that goes into winning a title. More concretely, star point guards are starring on offense not defense. So the real question to ask is whether star point guards have led great offenses? And of course, they have, like crazy. Much more than any other position really.

So then the only reason you wouldn't want a star point guard is if having one meant your defense had to suck. But that's crazy, why would anyone think a star point guard had to suck more on defense than an average point guard? There isn't any, so yes, it's good to have a star point guard. Depending on the situation it might not be the thing you want most for your team, but you'd never try to avoid it.
In 2007
The rule is the rule. It's not a matter of fairness. It's a matter of correctness.

In 2014
People just misunderstand...You’ve got to use some common sense, too.

I,for one, welcome our new insect overlords.
Doctor MJ



Senior Mod
Senior Mod
Posts: 32,128
And1: 2,309
Joined: Mar 10, 2005
Location: Treys are for Kicks!
Top

Post#14 Re: OWS distribution reveals championship sty
Sat Aug 4, 2012 6:13 am by mysticbb

Nivek wrote:Why have you chosen reaching the Finals as the criteria for success?


That is a logical choice given his agenda. He makes a typical mistake non-scientists (and sometimes scientists) are making, instead of formulating a hypothesis, testing the hypothesis and drawing a conclusion upon the results of the test, trainwreck tries to confirm his belief. A typical confirmation bias. That obviously makes the whole effort useless, but I doubt that he will understand that.
mysticbb


Banned User
Posts: 8,205
And1: 697
Joined: May 28, 2007
Top

Post#15 Re: OWS distribution reveals championship sty
Sun Aug 12, 2012 5:23 am by trainwreckog

..
the data speaks for itself.

clearly, you have a better chance of making the nba finals with either a balanced ows distribution, or, if someone is going to dominate the distribution, it should be your forward or center (or mj).

this is simply historical fact.


sure talent plays a big role in who makes the finals, but it would be a miraculous coincidence if out of the many many teams that had pgs with ows > 7.5 since 1980, only 2 were devoid of sufficient talent.
trainwreckog
Banned User
Posts: 752
And1: 103
Joined: Jul 13, 2012
Top

Post#16 Re: OWS distribution reveals championship sty
Sun Aug 12, 2012 2:08 pm by thizznation

In crunch time when the game on the line would you rather have a 6'6'' wingman that can get his?

Or a 6'0'' master facilitator that can always get a dime to someone else to get his?


Personally I choose the 6'6'' wingman
thizznation
Senior
Posts: 529
And1: 36
Joined: Aug 10, 2012
Top

Post#17 Re: OWS distribution reveals championship sty
Sun Aug 12, 2012 5:47 pm by Doctor MJ

trainwreckog wrote:..
the data speaks for itself.


Apparently it doesn't speak to you in a language you understand though. Data analysis has been going on for a long time and they've become quite sophisticated quite literally to keep people from making the exact mistakes you're making when you assume causation out of correlation.

Clearly you like the idea of looking at data and using it to make objective statements on events. All of us here applaud that, but you've got to go learn the fundamentals before your statements are meaningful.
In 2007
The rule is the rule. It's not a matter of fairness. It's a matter of correctness.

In 2014
People just misunderstand...You’ve got to use some common sense, too.

I,for one, welcome our new insect overlords.
Doctor MJ



Senior Mod
Senior Mod
Posts: 32,128
And1: 2,309
Joined: Mar 10, 2005
Location: Treys are for Kicks!
Top

Post#18 Re: OWS distribution reveals championship sty
Sun Aug 12, 2012 5:51 pm by Doctor MJ

thizznation wrote:In crunch time when the game on the line would you rather have a 6'6'' wingman that can get his?

Or a 6'0'' master facilitator that can always get a dime to someone else to get his?


Personally I choose the 6'6'' wingman


When you say "can get his", you talk like someone who is scoring at will. This is not what happens. It's true that certain wingmen are known for always being able to "create a shot", but they do this by completely changing their shooting form so that they don't get the ball deflected. When they do this, their efficiency gets very bad.

Simply put: In the clutch, the only scenario where you want to just hand it to your star and let him make find his shot is when you have so little faith in your ability to actually make passes that you'd rather just accept what would normally be considered a bad possession to avoid the possibility of an even worse one.
In 2007
The rule is the rule. It's not a matter of fairness. It's a matter of correctness.

In 2014
People just misunderstand...You’ve got to use some common sense, too.

I,for one, welcome our new insect overlords.
Doctor MJ



Senior Mod
Senior Mod
Posts: 32,128
And1: 2,309
Joined: Mar 10, 2005
Location: Treys are for Kicks!
Top

Post#19 Re: OWS distribution reveals championship sty
Mon Aug 13, 2012 1:23 am by Rapcity_11

Novel idea here:

Why don't we evaluate the offensive performance of the teams in question, rather than the overall performance? If the goal is to determine the better offensive style...?

The way guys approach analysis just blows my mind sometimes.

An indictment of ball dominant PG's needs to come from the basis of offensive performance, not overall.
Rapcity_11



General Manager
User avatar
Posts: 8,875
And1: 1,038
Joined: Jul 26, 2006
Top

Post#20 Re: OWS distribution reveals championship sty
Mon Aug 13, 2012 10:07 am by Nivek

trainwreckog wrote:..
the data speaks for itself.


No it doesn't because you haven't looked at it in a meaningful way. You haven't performed the same analysis with other positions. You haven't considered context. You've basically looked for stuff that supported your opinion going in and decided that this data supports your pre-held opinion.

clearly, you have a better chance of making the nba finals with either a balanced ows distribution, or, if someone is going to dominate the distribution, it should be your forward or center (or mj).

this is simply historical fact.


sure talent plays a big role in who makes the finals, but it would be a miraculous coincidence if out of the many many teams that had pgs with ows > 7.5 since 1980, only 2 were devoid of sufficient talent.


You shouldn't be talking about something being a "miraculous coincidence if..." at the end of a solid analysis. You should know. But you don't, because your research is incomplete. Which means that your conclusion isn't supported by the data you've presented.

And let me just say -- your conclusion may ultimately be correct. It would be an interesting finding if it turns out to be. But right now, the data you've presented doesn't support the conclusion you're claiming.
"A lot of what we call talent is the desire to practice."
-- Malcolm Gladwell

Check out my blog about the Wizards, movies, writing, music, TV, sports, and whatever else comes to mind.
Nivek





Head Coach
User avatar
Posts: 6,861
And1: 686
Joined: Sep 29, 2010
Top

Next

Return to Statistical Analysis


Who is online

Users browsing this forum: No registered users