All NBA Team = Josh Smith

[thatsmydawg]

To answer your question, No, I dont think they should have the same FG %. However, Josh does take so many outside shots (and more total shots) that it becomes more relevant for this particular situation. Everyone can agree Josh should chill with the outside J's.Everyone agrees that Joe is way overpaid. Some people say forget The Contract and focus on what Joe is giving the team. If you are going to do that then you need to focus on what Josh is giving the team. Well, you might say, we had to overpay to get Joe to stay.Well, I would reply with, Everyone has been hurt during the season except Josh so it makes sense that a). he would take more shots and b). that he has remained in rythm all year where as the guys coming and going from various injuries are not in good rythm and thus are deferring to Josh to shoot since he has emerged as the anchor and Iron man for this team. Just thinking out loud. These are only opinions.

[thatsmydawg]

The Hawks have lost 6 games this year but 1 possession or less. So yeah sometimes that missed shot can be important.Edit: Also 6 wins would put us in the 3rd seed and within one game of Miami for the division.

And how many of those games would have been that close without all the other things Josh provided during those games? Ill go ahead and answer, NONE!

[01d0rd3r]

To answer your question, No, I dont think they should have the same FG %. However, Josh does take so many outside shots (and more total shots) that it becomes more relevant for this particular situation.Everyone can agree Josh should chill with the outside J's.Everyone agrees that Joe is way overpaid.Some people say forget The Contract and focus on what Joe is giving the team. If you are going to do that then you need to focus on what Josh is giving the team.Well, you might say, we had to overpay to get Joe to stay.Well, I would reply with, Everyone has been hurt during the season except Josh so it makes sense that a). he would take more shots and b). that he has remained in rythm all year where as the guys coming and going from various injuries are not in good rythm and thus are deferring to Josh to shoot since he has emerged as the anchor and Iron man for this team.Just thinking out loud. These are only opinions.

I agree, Josh brings a lot to this team and his PER should probably higher than it is because it does not take into account some defensive statistics that Josh preforms well at. Josh should absolutely be 3rd team All NBA because he edges Aldridge in defensive stats and rebounds. Joe and Josh are our two most important players and there are complaints that I have against Joe as well. I just see Josh's deficiencies as more detrimental to the Hawks as a whole.

[01d0rd3r]

And how many of those games would have been that close without all the other things Josh provided during those games? Ill go ahead and answer, NONE!

That logic swings both ways. Since Josh is such a large part of the team and he does so many things he can be attributed to keeping us in game and also making mistakes that cost us games. Of course Joe and Al can also be accused of this at different times.

[thatsmydawg]

That logic swings both ways. Since Josh is such a large part of the team and he does so many things he can be attributed to keeping us in game and also making mistakes that cost us games. Of course Joe and Al can also be accused of this at different times.

It can be a double edged sword at times, yes.

[KB21]

Why should Josh's PER be higher? This statement does not make any sense, I don't think you fully understand what PER is telling you. It does not tell you some all encompassing "how good the overall player is" metric. Anyone who thinks that is brainwashed by Hollinger.PER doesn't lie, its just not that effective of a measure. When you realize how ad hoc it is in all of its weights as well as how highly correlated it is with simple NBA Efficiency measure, its just not that effective outside of telling you players who put up big numbers across the board (downplaying any negative numbers like turnovers and missed shots). If you want a measure that looks at how much an individual contributes to wins then look at Win Shares. If you care about the defensive side of things, look at Defensive Win Shares. But even then, Win Shares isn't a perfect measure so you shouldn't think of it as being an ultimatum statistic.tangent/A statistic should not tell you how good/bad a player is. A specific question needs to be formulated, and then once that is formulated you analyze statistics that help look at the specific question. Statistics supplement an answer, they are not the answer./tanget

I agree with this. There is not end all/be all statistic that tells us everything. I do think that statistics such as PER, USG, TS%, eFG%, WS, and WS/48 tell us a lot more about the player than basic box score statistics do. I agree that they are tools to answer questions. Such a question is whether a player who scores 30 points on 25 shot attempts contributes more to winning than a player who scores 15 points on 8 shots but also grabs 8 rebounds, blocks a shot, gets 2 steals, and holds the players he is guarding to 1 for 8 shooting?

[01d0rd3r]

Why should Josh's PER be higher? This statement does not make any sense, I don't think you fully understand what PER is telling you. It does not tell you some all encompassing "how good the overall player is" metric. Anyone who thinks that is brainwashed by Hollinger.PER doesn't lie, its just not that effective of a measure. When you realize how ad hoc it is in all of its weights as well as how highly correlated it is with simple NBA Efficiency measure, its just not that effective outside of telling you players who put up big numbers across the board (downplaying any negative numbers like turnovers and missed shots). If you want a measure that looks at how much an individual contributes to wins then look at Win Shares. If you care about the defensive side of things, look at Defensive Win Shares. But even then, Win Shares isn't a perfect measure so you shouldn't think of it as being an ultimatum statistic.tangent/A statistic should not tell you how good/bad a player is. A specific question needs to be formulated, and then once that is formulated you analyze statistics that help look at the specific question. Statistics supplement an answer, they are not the answer./tanget

You are right of course. The Stat measures what it is supposed to measure. What I mean to say is that Any kind of overall stat, even if it doesnt exist, would in my opinion have Josh as the 3rd best PF in the league behind Love and close to Griffen.

[KB21]

You are right of course. The Stat measures what it is supposed to measure. What I mean to say is that Any kind of overall stat, even if it doesnt exist, would in my opinion have Josh as the 3rd best PF in the league behind Love and close to Griffen.

The question then becomes how do you measure defense with a statistic? Some may say you look at blocked shots and steals. I think that gives you an idea, but it really doesn't tell you how good someone is as a defender. For example, Al Horford doesn't block many shots, but he is still a very good defender. I don't really understand exactly what all goes into defensive win shares, but I think it might be the closest gauge. Based on DWS, the two best defenders in the NBA this season are Dwight Howard and Josh Smith.

[Supporter]

I rarely ever rely on stats to measure a player because statistics are not always objective. For example, Player A could miss a shot but maybe it's a good shot to take, player B misses a shot as well, but it was a bad shot to take. As far as I know, in terms of stats both players have missed one shot, but the eye test shows which player made the better play. In the end of the day that's why people focus on accolades like All Star games, MVP's, Championship rings, etc.. because when it's all said and done, those are the guys who played the best basketball.Same goes with every stat, how can you really evaluate the a person's defense in terms of stats??? By blocks and steals??? Javale Mcgee is pretty good at blocking shots but by no means is he an excellent defender.

Edited April 10, 2012 by Supporter

[KB21]

I rarely ever rely on stats to measure a player because statistics are not always objective. For example, Player A could miss a shot but maybe it's a good shot to take, player B misses a shot as well, but it was a bad shot to take. As far as I know, in terms of stats both players have missed one shot, but the eye test shows which player made the better play. In the end of the day that's why people focus on accolades like All Star games, MVP's, Championship rings, etc.. because when it's all said and done, those are the guys who played the best basketball.Same goes with every stat, how can you really evaluate the a person's defense in terms of stats??? By blocks and steals??? Javale Mcgee is pretty good at blocking shots but by no means is he an excellent defender.

And that bears out with defensive win shares. McGee is 11th in block percentage, but he is 66th in defensive win shares. Serge Ibaka is 1st in block percentage among qualified players, but he is 14th in DWS. Dwight Howard is 40th in block percentage while Josh Smith is 51st, but they are 1 and 2 respectively in DWS.

[01d0rd3r]

The question then becomes how do you measure defense with a statistic? Some may say you look at blocked shots and steals. I think that gives you an idea, but it really doesn't tell you how good someone is as a defender. For example, Al Horford doesn't block many shots, but he is still a very good defender.I don't really understand exactly what all goes into defensive win shares, but I think it might be the closest gauge. Based on DWS, the two best defenders in the NBA this season are Dwight Howard and Josh Smith.

That has been a large problem with fans measuring a players defense for awhile. You have to rely heavily on the eyeball test to know if someone is a good defender particularly for perimeter defenders.

[thatsmydawg]

I think the problem might be the amount of different metrics that can be used. There is a metric out there that you think will prove your point but then someone else will find the holes in it. Maybe something more elementary and simple is the way to go.The hawks are 5th right now in opponents ppg and 5th on the road in the same category. Well that tells me that joe and josh are anchoring a pretty elite defense this year! The question then becomes why? Or better, how? Well that's where the many different stats can be debated over. you can look at it from an individual perspective, but I think when it comes to defense it is more prudent to look at from a team perspective. Maybe because of the Intangibles associated with defense, such as how many shots a guy changes because of his reputation for blocking shots or how well a guy prevents his man from catching the ball due to his off the ball defense. Can you really find an accurate stat for stuff like that?I think the hawks low turnover rate is a key factor in the ppg category I mentioned. Preventing easy baskets for your opponent. We aren't good in rebounding really so it's gotta be something else. Maybe it's just easier for me to look at it this way!

Edited April 10, 2012 by thatsmydawg

[PureGreatness]

I still till this day don't know what the hell PER is. I heard it involves math...which I'm not good in math.

[benhillboy]

Has here ever been an All-NBA player who didn't go to the All-Star Game? I would doubt it. And as far as the eyeball test to gauge defense, just compare perimeter defender A to Iman Shumpert and frontcourt defender A to Silly Dwight. Iman's ball pressure, lateral movement, and hand speed are second to none, and Dwight challenges as many shots at the rim as humanly possible. I would think team defensive principles like guarding the PNR or rotations on help would be much harder to quantify and attribute to any single player or players. The praise I guess would go to the coaching staff, and the quality of the results would lie in how far the team went after Game 1 of the Conference Finals since the Final 4 teams in the NBA most times represent the most formidable team defenses. It is common knowledge that defense and rebounding wins O'Brian trophies, so that seems like a good indicator to me, although the Defensive Win Share Theory is interesting. In 5 years there probably will be a nice rubric to quantify defense much better, developed by someone who is much smarter with more (paid) time to analyze advanced basketball statistics on their hands than any of us. I would love to sit down with Popovich and pick his brain as to what kind of statistical analysis he likes and dislikes. Getting a candid, one word, yes or no answer to whether his team is better with or without Rose based on defensive ratings from Thibideau would be cool too.

Edited April 10, 2012 by benhillboy

[Admin]

I still till this day don't know what the hell PER is. I heard it involves math...which I'm not good in math.

http://www.basketball-reference.com/about/per.html

Calculating PER

The Player Efficiency Rating (PER) is a per-minute rating developed by ESPN.com columnist John Hollinger. In John's words, "The PER sums up all a player's positive accomplishments, subtracts the negative accomplishments, and returns a per-minute rating of a player's performance." It appears from his books that John's database only goes back to the 1988-89 season. I decided to expand on John's work and calculate PER for all players since minutes played were first recorded (1951-52).

All calculations begin with what I am calling unadjusted PER (uPER). The formula is:

uPER = (1 / MP) *

[ 3P

+ (2/3) * AST

+ (2 - factor * (team_AST / team_FG)) * FG

+ (FT *0.5 * (1 + (1 - (team_AST / team_FG)) + (2/3) * (team_AST / team_FG)))

- VOP * TOV

- VOP * DRB% * (FGA - FG)

- VOP * 0.44 * (0.44 + (0.56 * DRB%)) * (FTA - FT)

+ VOP * (1 - DRB%) * (TRB - ORB)

+ VOP * DRB% * ORB

+ VOP * STL

+ VOP * DRB% * BLK

- PF * ((lg_FT / lg_PF) - 0.44 * (lg_FTA / lg_PF) * VOP) ]

Most of the terms in the formula above should be clear, but let me define the less obvious ones:

factor = (2 / 3) - (0.5 * (lg_AST / lg_FG)) / (2 * (lg_FG / lg_FT))

VOP = lg_PTS / (lg_FGA - lg_ORB + lg_TOV + 0.44 * lg_FTA)

DRB% = (lg_TRB - lg_ORB) / lg_TRB

I am not going to go into details about what each component of the PER is measuring; that's why John writes and sells books.

Problems arise for seasons prior to 1979-80:

1979-80 — debut of 3-point shot in NBA

1977-78 — player turnovers first recorded in NBA

1973-74 — player offensive rebounds, steals, and blocked shots first recorded in NBA

The calcuation of uPER obviously depends on these statistics, so here are my solutions for years when the data are missing:

Zero out three-point field goals, turnovers, blocked shots, and steals.

Set the league value of possession (VOP) equal to 1.

Set the defensive rebound percentage (DRB%) equal to 0.7.

Set player offensive rebounds (ORB) equal to 0.3 * TRB.

Some of these solutions may not be elegant, but I think they are reasonable. After uPER is calculated, an adjustment must be made for the team's pace. The pace adjustment is:

pace adjustment = lg_Pace / team_Pace

League and team pace factors cannot be computed for seasons prior to 1973-74, so I estimate the above using:

estimated pace adjustment = 2 * lg_PPG / (team_PPG + opp_PPG)

To give you an idea of the accuracy of these estimates, here are the actual pace adjustments and the estimated pace adjustments for teams from the Eastern Conference in 2002-03:

Tm Act Est

ATL 1.00 0.99

BOS 1.00 1.02

CHI 0.97 0.98

CLE 0.97 0.99

DET 1.05 1.06

IND 0.99 1.00

MIA 1.04 1.08

MIL 1.01 0.96

NJN 0.99 1.03

NOH 1.01 1.02

NYK 1.00 0.98

ORL 0.98 0.97

PHI 1.00 0.99

TOR 1.01 1.01

WAS 1.03 1.03

For all seasons where actual pace adjustments can be computed, the root mean square error of the estimates is 0.01967.

Now the pace adjustment is made to uPER (I will call this aPER):

aPER = (pace adjustment) * uPER

The final step is to standardize aPER. First, calculate league average aPER (lg_aPER) using player minutes played as the weights. Then, do the following:

PER = aPER * (15 / lg_aPER)

The step above sets the league average to 15 for all seasons.

Those are the gory details. If you have any comments or questions, please send me some feedback.

[Supporter]

That just reeks of someone who doesn't understand statistics, or at least the central limit theorem or law of large numbers. You are focused on one individual instance, that a missed shot counts the same. OK fine. But think: aren't shots repeated during a game and throughout the course of a year? So while you are concerned that the shots are counted the same, isn't it more likely that when you take a repeated sampling by an individual player of "good shots" and "bad shots" that we will see the true value over time? You don't use anecdotes to argue against statistics, that is pure foolishness. That is like saying "well Portland won the lottery in 2007 but weren't at all likely to win it so statistics are biased".

I challenge you and your eyeball test. I want you to watch a game where you cover up the score. Sit there with your eyeball test and tell me who won the game. Err...clearly that won't work because player's know the score and at the end of the game some will clearly be happy while the others won't. So scratch the end of the game, just tell me who was winning at the end of the 3rd quarter. Go ahead, do it. If statistics clearly are biased and are useless then you should have no problem whatsoever to be able to match who is winning. And you cannot keep score because that is a statistics.

I understand that over the long run, stats bear some value in evaluating a player. But it still doesn't tell the whole truth. Maybe I used the wrong wording when I said that stats aren't objective. All I am saying is that the stats of different players can't be evaluated in the same way because of variables such as the system they are in and the players around them. All variables aren't going to be the same from player to player, so when you have stats like per, or win shares or whatever it is it doesn't tell the truth because guys play on different teams, different positions, and next to different players. For example player A might average 10 rebounds a season compared to Player B that averages 8 rebounds, but you can't really evaluate them solely on stats alone because player A might play on a team that has no decent rebounders while player B does.

I don't know why you are asking me to not look at the score, I am only concerned with an individual's stats, not the final score to a game.

Edited April 10, 2012 by Supporter

[PureGreatness]

http://www.basketbal.../about/per.html

Calculating PER

The Player Efficiency Rating (PER) is a per-minute rating developed by ESPN.com columnist John Hollinger. In John's words, "The PER sums up all a player's positive accomplishments, subtracts the negative accomplishments, and returns a per-minute rating of a player's performance." It appears from his books that John's database only goes back to the 1988-89 season. I decided to expand on John's work and calculate PER for all players since minutes played were first recorded (1951-52).

All calculations begin with what I am calling unadjusted PER (uPER). The formula is:

uPER = (1 / MP) *

[ 3P

+ (2/3) * AST

+ (2 - factor * (team_AST / team_FG)) * FG

+ (FT *0.5 * (1 + (1 - (team_AST / team_FG)) + (2/3) * (team_AST / team_FG)))

- VOP * TOV

- VOP * DRB% * (FGA - FG)

- VOP * 0.44 * (0.44 + (0.56 * DRB%)) * (FTA - FT)

+ VOP * (1 - DRB%) * (TRB - ORB)

+ VOP * DRB% * ORB

+ VOP * STL

+ VOP * DRB% * BLK

- PF * ((lg_FT / lg_PF) - 0.44 * (lg_FTA / lg_PF) * VOP) ]

Most of the terms in the formula above should be clear, but let me define the less obvious ones:

factor = (2 / 3) - (0.5 * (lg_AST / lg_FG)) / (2 * (lg_FG / lg_FT))

VOP = lg_PTS / (lg_FGA - lg_ORB + lg_TOV + 0.44 * lg_FTA)

DRB% = (lg_TRB - lg_ORB) / lg_TRB

I am not going to go into details about what each component of the PER is measuring; that's why John writes and sells books.

Problems arise for seasons prior to 1979-80:

1979-80 — debut of 3-point shot in NBA

1977-78 — player turnovers first recorded in NBA

1973-74 — player offensive rebounds, steals, and blocked shots first recorded in NBA

The calcuation of uPER obviously depends on these statistics, so here are my solutions for years when the data are missing:

Zero out three-point field goals, turnovers, blocked shots, and steals.

Set the league value of possession (VOP) equal to 1.

Set the defensive rebound percentage (DRB%) equal to 0.7.

Set player offensive rebounds (ORB) equal to 0.3 * TRB.

Some of these solutions may not be elegant, but I think they are reasonable. After uPER is calculated, an adjustment must be made for the team's pace. The pace adjustment is:

pace adjustment = lg_Pace / team_Pace

League and team pace factors cannot be computed for seasons prior to 1973-74, so I estimate the above using:

estimated pace adjustment = 2 * lg_PPG / (team_PPG + opp_PPG)

To give you an idea of the accuracy of these estimates, here are the actual pace adjustments and the estimated pace adjustments for teams from the Eastern Conference in 2002-03:

Tm Act Est

ATL 1.00 0.99

BOS 1.00 1.02

CHI 0.97 0.98

CLE 0.97 0.99

DET 1.05 1.06

IND 0.99 1.00

MIA 1.04 1.08

MIL 1.01 0.96

NJN 0.99 1.03

NOH 1.01 1.02

NYK 1.00 0.98

ORL 0.98 0.97

PHI 1.00 0.99

TOR 1.01 1.01

WAS 1.03 1.03

For all seasons where actual pace adjustments can be computed, the root mean square error of the estimates is 0.01967.

Now the pace adjustment is made to uPER (I will call this aPER):

aPER = (pace adjustment) * uPER

The final step is to standardize aPER. First, calculate league average aPER (lg_aPER) using player minutes played as the weights. Then, do the following:

PER = aPER * (15 / lg_aPER)

The step above sets the league average to 15 for all seasons.

Those are the gory details. If you have any comments or questions, please send me some feedback.

-________________________________________________________-

[thatsmydawg]

I think you are looking at this the right way. Start simple with things, make sure your statistics address a sound question, and then start building from there. My strong belief is that as soon as things become more complicated, you lose a lot of people. This can be either because your audience doesn't understand the statistic you bring up (and then start to make excuses of "correlation doesn't imply causation" or calling "lies, damned lies, and statistics!") or you do not understand the statistic you are bringing up. A big rule of thumb, if you don't understand it then don't use it.The changing of shots is something that is hard to measure but my understanding from watching some of the Sloan Sports Conference over the years is that the data have been lacking but proprietary data is being compiled to gather data on this. So once the data become available, then you can go into churning the data into meaningful statistics. So from the individuals perspective, development of those statistics are a short time away (I think within 3 to 5 years).However, have you ever looked into adjusted plus-minus? It seems you are more concerned about the impact of a player on the team as a whole and that is the idea behind adjusted plus-minus. You can break down adjusted plus-minus into overall, offensive, and defensive to look at how efficient the team is based upon one player. It needs a lot of observations to be reliable, but the concept of plus-minus is rather straight-forward. Look at how a team performs when an individual is on versus off the court. Over time, the only thing that is changing is your unit of observation (the individual) and so with enough observations this approximates how the individual contributes to the team. The adjusted part of the plus-minus situation is what makes it a complicated statistic. All the adjusted part is trying to do is correct for biases in substitution patterns. A bias might be observed if there are correlations between players (positive would imply that your interested player goes out of the game typically at the same time another player does; negative would imply that your interested player goes out of the game typically when another player enters the game). If you are interested in this stuff, I would check it out.

Thanks for the info. Ill definitely look at that. I wonder what that stat will look like when they finish gathering the data?

[Plainview1981]

And this is precisely why you have to take his shot selection into account. With some players, you have to take the bad with the good. Josh's shot selection is bad, but he is clearly above average with everything else. Do you realize just how much he has to be killing it inside to overcome his shot selection and poor free throw shooting? The guy has posted a 21+ PER this season despite of his shot selection.

Josh has been the best Power Forward in the NBA since the all star break.

[Buzzard]

And that bears out with defensive win shares. Serge Ibaka is 1st in block percentage among qualified players, but he is 14th in DWS.

I don't see how defensive win shares bears anything out if it makes someone like Serge look like a top 20 all-star. Don't get me wrong, I wish we had him but he is not a top 20 all-star...