Heard Magic and Wilbon Ragging on Metrics.

[Diesel]

I think some of you all have been waiting for a moneyball arguement.

What WIlborn and Magic both alluded to was the lineups without Wade.

Mike Wilbon Said ‏@MikeWilbonSaid

"I don't hear all the metrics-heads. Where are they? 'I don't wanna have @DwyaneWade on the floor.' Really?" #Wilbon 6/20/13 @RealMikeWilbon

When you start to use stats to base your lineups then you do stuff like sitting Duncan at the end of games so that you don't have to worry about him getting fouled?

[Buzzard]

To summarize what other posters have said, the only people who rag on metrics are people who (1) do not understand the metrics themselves or (2) do not understand the utility of the metrics.One of my favorite lines from Moneyball: Anti-intellectual resentment is common in all of American life and it has many diverse expressions.

I am a stat head of sorts but rag on metrics a lot because most hard core NBA fans are also hard core fantasy players and NBA 2K players. You see it in fantasy stats and see it on your nintendo, so it must be true.

[Diesel]

The subquestion of this thread is what does it mean for us.

We supposedly have two disciples of Metrics as our GM and Coach. Do you supposed that we will be metric'd out?

[Buzzard]

The subquestion of this thread is what does it mean for us.

We supposedly have two disciples of Metrics as our GM and Coach. Do you supposed that we will be metric'd out?

From the 3pt line I do. Inside the paint only to a certain extent, simply because i think Danny is old school and believes in this saying "you can't teach size".

[KB21]

Diesel, the "problem" with metrics, is that if someone can look at one stat that proves or disproves their opinion about a player, they'll tend to roll with that to make a TOTAL evaluation about a player.

Whether it be PER, or eFG%, or TS% or Offensive/Defensive rating, or any other metric, if a stat can prove/disprove things about a player, "statheads" will roll with that.

Kyle Korver, by all metrics, is one of the best offensive players in the league. He had an off the chart TS% of .637 an eFG% of .618. Kobe Bryant has a TS% of .570 and an eFG% of .504

So which player would you want to see most of your offense going through? If you simply looked at the stats, there's no question the answer to that would be Kyle Korver. But if you actually watched them play, it would hands down be Kobe Bryant.

You can come to a lot of conclusions by looking at stats. But those stats also need to coincide with what you actually SEE that player do.

I think that's what Magic and Wilborn really allude to. They trust their eyes more than they trust analytics.

That's not what the advanced metrics say at all, nor is that what those of us who like to use advanced metrics state. That is a statement that is only made by those that do not understand the advanced metrics at all and simply turn a blind eye to any analytical method that is out there that goes against the conventional wisdom largely perpetuated by the media.

You left out one important part of your evaluation there.

Kyle Korver has a 14.5 USG%. Kobe Bryant had a 31.9 USG.

There is a clear difference in who is a very efficient spot up jump shooter and who is a high usage, relatively high efficiency player.

Where Kobe Bryant is overrated from an efficiency standpoint is when he is compared with other high usage players. Kevin Durant, LeBron James, and James Harden absolutely smoke Kobe Bryant when it comes to efficiency, and yes, I would take all three of them over Kobe Bryant.

As good of tools as TS% and USG% are though, they still don't tell much about the player's overall impact on the floor.

A couple of stats that do that fairly well though that I have found are ASPM (Advanced Statistical Plus Minus) and VORP (Value Over Replacement Player.

On the Hawks, Kyle Korver's ASPM was 1.325, which was second on the team to Al Horford's 2.241. His VORP was 2.317, which was 4th on the team behind Al, Josh Smith, and Jeff Teague. Statistically, those four players were by far the best players on the Hawks this year, and that clearly passes the eye ball test as well.

For good measure, Kobe Bryant's ASPM was 4.71, which was 6th in the NBA behind LeBron James, Kevin Durant, Russell Westbrook, Chris Paul, and James Harden. His VORP was 5.38, again 6th in the league behind the same 5 guys.

[KB21]

The subquestion of this thread is what does it mean for us.

We supposedly have two disciples of Metrics as our GM and Coach. Do you supposed that we will be metric'd out?

Not only do the Hawks have a GM and Head Coach that believe in metrics, they have also hired a consultant who is one of the gurus among the analytics community.

Neil Paine is a consultant for the Hawks.

[Buzzard]

For good measure, Kobe Bryant's ASPM was 4.71, which was 6th in the NBA behind LeBron James, Kevin Durant, Russell Westbrook, Chris Paul, and James Harden. His VORP was 5.38, again 6th in the league behind the same 5 guys.

And this is where the stats lie as far as the eye and common sense test go. Where is the dbl dbl big men on your list? There is something f***ed UP when a stat line claims you would get better results taking Russell Westbrook, Chris Paul, James Harden, and a old Kobe Bryant over D12.

I will give you Durant and Bron, but there is no one else in the NBA I would rather have than D12. And certainly not a 34 year old Kobe Bryant.

Edited June 22, 2013 by Buzzard

[MaceCase]

And this is where the stats lie as far as the eye and common sense test go. Where is the dbl dbl big men on your list? There is something f***ed UP when a stat line claims you would get better results taking Russell Westbrook, Chris Paul, James Harden, and a old Kobe Bryant over D12.

I will give you Durant and Bron, but there is no one else in the NBA I would rather have than D12. And certainly not a 34 year old Kobe Bryant.

Jeff Van Gundy is an old school coach and he absolutely hates the importance that people place on "Double-Doubles". As he often says himself, who had a greater impact on a game, the guy who got 13 points and 11 rebounds for his precious double-double or the guy who got 32 and 9? The eye test would absolutely tell you that Dwight Howard had one of his worst seasons on par with probably his first couple years in the league so why does it blow "common sense" away that the stats confirm it and that he was indeed outperformed the previous season? Are you honestly saying that someone like Omer Asik should be on par with those players simply based on averaging a "double-double"?

This is one of the issues with people and their lack of understanding with the stats, they are simply telling you what happened, what was recorded but you think it has no merit because you are trying to prognosticate the future or place criteria that the stat itself never claimed to determine. YES, Kobe Bryant had one of his statistically best seasons in his entire illustrious career at that age! YES, he outperformed a Dwight Howard coming off of back surgery.

[Admin]

Keep in mind that Dwight played the entire season last year with an injured back and on a team that didn't feature him in the offense the way that he was used to. He started off very poorly but was playing much better by the end of the season.

I don't think there's anything someone could really argue against a healthy Dwight being the 3rd best player in the league behind Lebron and Durant and it's arguable vs Durant. But I doubt we'll ever see a fully healthy Dwight ever again.

[supermariowest]

I am a stat head of sorts but rag on metrics a lot because most hard core NBA fans are also hard core fantasy players and NBA 2K players. You see it in fantasy stats and see it on your nintendo, so it must be true.

I'm not entirely sure what point you are trying to make, but most fantasy leagues use older, established statistics instead of advanced metrics.

[KB21]

This past year, Dwight had one of the worst seasons of his career. Part of that was due to lack of usage, and part of that was due to him playing at less than 100%, which he will likely be playing at for the rest of his career with his bad back.

I'm not sure why you think the ASPM numbers of this year don't grasp Howard's impact on the floor?

In 2011-2012, Howard's ASPM was 4.53, which was 6th in the NBA. In 2010-2011, his ASPM was 5.89. In 2009-2010, it was 6.31.

[Admin]

This past year, Dwight had one of the worst seasons of his career. Part of that was due to lack of usage, and part of that was due to him playing at less than 100%, which he will likely be playing at for the rest of his career with his bad back.

I'm not sure why you think the ASPM numbers of this year don't grasp Howard's impact on the floor?

In 2011-2012, Howard's ASPM was 4.53, which was 6th in the NBA. In 2010-2011, his ASPM was 5.89. In 2009-2010, it was 6.31.

Those are pretty darn good numbers for Dwight. What was his number this year? I expect it would be closer to 2 but that expecting it to be around 4 or so for the next 5 years should be about right.

[KB21]

Those are pretty darn good numbers for Dwight. What was his number this year? I expect it would be closer to 2 but that expecting it to be around 4 or so for the next 5 years should be about right.

His ASPM was in 2012-2013 was 1.94, which was a big drop off from where it had been. Is it an aberration? Probably, but I do wonder if the expectation that he will be back at 4-5 for the next 5 years or so is something that will be realized. The thing that concerns me about Dwight and the contract he will be getting is that his game is based off his athleticism. He's already showing signs of wear and tear with his back, that will likely be a chronic problem for him going forward. What kind of impact is he going to have on the game when he can't run as fast as he typically does or jump as high as he typically does? Are we already seeing some decline in those athletic skills?

He's 27, so he's right at the age peak where you start to see some decline. Athletic player tend to decline faster, particularly when they have injury problems. A player like Dirk Nowitzki didn't have the typical decline you see in players who are 30+, but his game was never based on athleticism.

I'm not an expert on projecting these stats the way guys like DMosk and Neil Paine are, but based on what I've read, you might see Dwight have one more 4+ ASPM season but likely have a player who is in the 2-3 range over the course of the next three years before you see a drastic decline.

From my standpoint, Chris Paul is the only player I'd give a max contract to this off-season. The good thing about the new CBA is that the most years you can sign a player is 4, so you can somewhat protect yourself from yourself. Teams shouldn't do what the Hawks did in 2010 when they signed Joe Johnson to a 6 year, $119 million deal that basically killed their ability to add talent around him effectively.

[Buzzard]

Jeff Van Gundy is an old school coach and he absolutely hates the importance that people place on "Double-Doubles". As he often says himself, who had a greater impact on a game, the guy who got 13 points and 11 rebounds for his precious double-double or the guy who got 32 and 9? The eye test would absolutely tell you that Dwight Howard had one of his worst seasons on par with probably his first couple years in the league so why does it blow "common sense" away that the stats confirm it and that he was indeed outperformed the previous season? Are you honestly saying that someone like Omer Asik should be on par with those players simply based on averaging a "double-double"?

This is one of the issues with people and their lack of understanding with the stats, they are simply telling you what happened, what was recorded but you think it has no merit because you are trying to prognosticate the future or place criteria that the stat itself never claimed to determine. YES, Kobe Bryant had one of his statistically best seasons in his entire illustrious career at that age! YES, he outperformed a Dwight Howard coming off of back surgery.

I am talking about what I know, what I saw, and the elite players in the league. Howard was the leading rebounder in the NBA last season but KBs stats say he is not elite as in top five. Ok, I can live with that because there is a lot more to basketball than just rebounding.

Lets try this one, Howard was the only player in the NBA last season to avg, 15 or more points a game, 10 plus rebounds per game, and 2 or more blocks a game. Only is the keyword here. No other player came close to Howard's overall big man game. That is elite.

Still not enough, how about this Marc Gasol. The writers choice for Defensive Player of the Year. Avg. 14.1, 7.8, 1.0 and 1.7 vs D12's 17.1, 12.4, 1.1, and 2,5. Is it any wonder Howard laughed when he was questioned by reporters about this seasons award?

You harp on his injury, Dwight only missed 6 games last season. You think I am skewed against stats. That is not further from the truth. D12's worse season in years is better than any other big man had last season.

With everything I just posted, how in the world can anyone trust a stat that says Harden played better overall ball than Dwight? That is what I question. Its not whether Dwight was better than 4 or KB's top 6. I know he was. Its the stat he uses that shows me something different is what I question.

D12, injury or not, was the best big man in the game last season; but that is not good enough to beat out statistically any one of the three best guards in the league. Gimme a break.

Oh and he was only 2nd in FG%, Damn he really sucked last season.

Edited June 22, 2013 by Buzzard

[Buzzard]

This past year, Dwight had one of the worst seasons of his career. Part of that was due to lack of usage, and part of that was due to him playing at less than 100%, which he will likely be playing at for the rest of his career with his bad back.

I'm not sure why you think the ASPM numbers of this year don't grasp Howard's impact on the floor?

In 2011-2012, Howard's ASPM was 4.53, which was 6th in the NBA. In 2010-2011, his ASPM was 5.89. In 2009-2010, it was 6.31.

His usage is the key. Howard took 813 shots with the Lakers in 76 games. The season before he took 726 in only 54 games. 813 is the least amount of shots in a full season of games he has taken since his rookie year. I don't think its just coincidence that his scoring average dropped significantly. But hell, I ain't no rocket scientists so maybe it is and Howard really is no longer elite.

Do you think we can get him by just beating out a MLE?

Edited June 22, 2013 by Buzzard

[MaceCase]

I am talking about what I know, what I saw, and the elite players in the league. Howard was the leading rebounder in the NBA last season but KBs stats say he is not elite as in top five. Ok, I can live with that because there is a lot more to basketball than just rebounding.

Lets try this one, Howard was the only player in the NBA last season to avg, 15 or more points a game, 10 plus rebounds per game, and 2 or more blocks a game. Only is the keyword here. No other player came close to Howard's overall big man game. That is elite.

Still not enough, how about this Marc Gasol. The writers choice for Defensive Player of the Year. Avg. 14.1, 7.8, 1.0 and 1.7 vs D12's 17.1, 12.4, 1.1, and 2,5. Is it any wonder Howard laughed when he was questioned by reporters about this seasons award?

You harp on his injury, Dwight only missed 6 games last season. You think I am skewed against stats. That is not further from the truth. D12's worse season in years is better than any other big man had last season.

With everything I just posted, how in the world can anyone trust a stat that says Harden played better overall ball than Dwight? That is what I question. Its not whether Dwight was better than 4 or KB's top 6. I know he was. Its the stat he uses that shows me something different is what I question.

D12, injury or not, was the best big man in the game last season; but that is not good enough to beat out statistically any one of the three best guards in the league. Gimme a break.

Oh and he was only 2nd in FG%, Damn he really sucked last season.

This seems to be a problem that you are having lately where you think that being the best or second best at your position elevates you above all members at other positions. These stats are not created to point out the best 5 man rotation, they are pointing out the best 5 or so players for that singular year, period.

Lets just do this a simple none advanced or perfected away of just taking basic box scores and comparing Harden and Dwight.

25.9, 4.9, 5.8, 1.8, .5

17.1, 12.4, 1.4, 1.1, 2.4

+8.8pts

-7.5rebounds

+4.4assists

+.7steals

-1.9blocks

+4.5 total

No need to compare PER, or Win Shares, or TS% or eFG. I just added and subtracted their basic boxscore stats and Harden came out +4.5 ahead. What, because he's the second SG or 3rd G or 4th wing he should be placed behind a guy who is the first C for some reason? What, we shouldn't trust any stat now that disagrees with what you saw? Sorry, stats just don't work like that otherwise Josh would somehow be elevated to top 10 in the league for being the best tweener.

Edited June 22, 2013 by MaceCase

[NBASupes]

And this is where the stats lie as far as the eye and common sense test go. Where is the dbl dbl big men on your list? There is something f***ed UP when a stat line claims you would get better results taking Russell Westbrook, Chris Paul, James Harden, and a old Kobe Bryant over D12.

I will give you Durant and Bron, but there is no one else in the NBA I would rather have than D12. And certainly not a 34 year old Kobe Bryant.

The best overall stat is on 82games.com. The rest are good to look at but it doesn't tell you someone's overall impact on the game with is most important if you are trying to win titles. The rest is good for seeing potential. ability to do certain things, and what not. A lot of stats can be flawed depending on how the person presenting them uses it.

[Buzzard]

The best overall stat is on 82games.com. The rest are good to look at but it doesn't tell you someone's overall impact on the game with is most important if you are trying to win titles. The rest is good for seeing potential. ability to do certain things, and what not. A lot of stats can be flawed depending on how the person presenting them uses it.

The things that stats don't show is Howard in the lane changing shots, while Mike Conley puts up 17.3 and 8.3 in 6 games against Paul in the playoffs, Conley only averaged 14.6 and 6.1 in the regular season. Gasol played worse against Howard than he did against Duncan when they lost. A players stats don't show that shit unless you watch the games and look at what the opposing teams box score is like.

[Admin]

The best overall stat is on 82games.com. The rest are good to look at but it doesn't tell you someone's overall impact on the game with is most important if you are trying to win titles. The rest is good for seeing potential. ability to do certain things, and what not. A lot of stats can be flawed depending on how the person presenting them uses it.

Which stat are you referring to, their simple rating? If so I haven't seen that stat referenced in any article that I can ever remember. If you are referring to +/- then that stat is listed all over the place and it's really too 'simple' to be used as a main stat to judge a player or to compare players.

[Buzzard]

This seems to be a problem that you are having lately where you think that being the best or second best at your position elevates you above all members at other positions. These stats are not created to point out the best 5 man rotation, they are pointing out the best 5 or so players for that singular year, period.

Lets just do this a simple none advanced or perfected away of just taking basic box scores and comparing Harden and Dwight.

25.9, 4.9, 5.8, 1.8, .5

17.1, 12.4, 1.4, 1.1, 2.4

+8.8pts

-7.5rebounds

+4.4assists

+.7steals

-1.9blocks

+4.5 total

No need to compare PER, or Win Shares, or TS% or eFG. I just added and subtracted their basic boxscore stats and Harden came out +4.5 ahead. What, because he's the second SG or 3rd G or 4th wing he should be placed behind a guy who is the first C for some reason? What, we shouldn't trust any stat now that disagrees with what you saw? Sorry, stats just don't work like that otherwise Josh would somehow be elevated to top 10 in the league for being the best tweener.

No they don't work like that and most stats are weighted to the offense. The reason is obvious, you can't measure a player missing a shot when all D12 did was almost block it. But you can measure a assist even though it takes the 2nd player making the damn shot..

Its pretty obvious that scoring is the small mans stat, where most formulas miss out is giving 100% credit for a pass. and 100% credit to the player making the shot. The scorer is the one doing the hardest part at least 75% of the time.