Retro Player of the Year 2014-15

[lessthanjake]

I was just reminded of this:

Read on Twitter

This uses synergy shot quality data and finds that Steph is alone at the top in terms of increasing the shot quality of his teammates. It’s not the right timeframe (2018-2022, rather than including 2015), but the results are consistent with the shot-quality analysis I did using PBPstats shot quality data stretching back to earlier years that include this 2015 season (at the following link and also spoilered below: https://forums.realgm.com/boards/viewtopic.php?p=107624238#p107624238)

Spoiler:

I want to also just provide some information that I think goes to the playmaking effect of Steph Curry’s gravity.

www.pbpstats.com has data on “shot quality” that a player’s teammates have had with and without that player on the floor. This is a bit of a back-of-the-napkin method, but I looked at regular season + playoffs and then averaged the shot quality increase with certain star players on the floor, for teammates that played at least 1,000 minutes with the given player. I chose a set of great playmakers and some other great players, and tried to use timeframes correlating with their best years and/or their stints with specific teams (note: the data starts at 2000-2001, so that’s why no timeframe includes years before that). The idea was to measure to what extent a player causes his teammates to get good looks. Here’s what I found:

Average Shot Quality Increase for Teammates with Player on the Floor (only teammates with 1,000+ minutes with the player)

1. Steph Curry (2014-2015 to 2018-2019): +5.1%

2. Steph Curry (2014-2015 to 2022-2023): +3.8%

3. Nikola Jokic (2020-2021 to 2022-2023): +3.7%

4. LeBron James (2014-2015 to 2017-2018): +2.9%

5. Steve Nash (2004-2005 to 2011-2012): +2.6%

6. LeBron James (2005-2006 to 2009-2010): +2.3%

7. James Harden (2012-2013 to 2019-2020): +2.2%

8. Draymond Green (2014-2015 to 2022-2023): +2.2%

9. Shaquille O’Neal (2000-2001 to 2003-2004): +1.9%

10. Kobe Bryant (2000-2001 to 2009-2010):: +1.9%

11. Tim Duncan (2000-2001 to 2006-2007): +1.7%

12. Chris Paul (2011-2012 to 2016-2017): +1.5%

13. Giannis Antetokounmpo (2019-2020 to 2022-2023): +1.5%

14. Luka Doncic (2019-2020 to 2022-2023): +1.3%

15. LeBron James (2010-2011 to 2013-2014): +1.0%

Indeed, in my analysis, the inclusion of later years brought Steph down a good bit, so the above tweet probably *understates* Steph’s lead in the relevant timeframe. So yeah, this adds to the evidence that Steph’s “creation” outdoes other players, including LeBron. Which shouldn’t be surprising to those who see how his presence warps defenses and uniquely pulls defenders *away* from the basket, which is better than pulling defenders towards the basket (since opening up space near the basket is better than opening up space away from the basket).

[Top10alltime]

Indeed, in my analysis, the inclusion of later years brought Steph down a good bit, so the above tweet probably *understates* Steph’s lead in the relevant timeframe. So yeah, this adds to the evidence that Steph’s “creation” outdoes other players, including LeBron. Which shouldn’t be surprising to those who see how his presence warps defenses and uniquely pulls defenders *away* from the basket, which is better than pulling defenders towards the basket (since opening up space near the basket is better than opening up space away from the basket).

That's not what creation is. Creation is made up of many things

Delivering the ball to your teammates (delivery quality)
Taking out defenders
The open-ness of shots

And... anyone who knows basketball, knows on-ball playmaking and scoring is far more impactful than off-ball playmaking. That's just how basketball works!

Take a look at any ATG off-ball payer like Dirk, Bird, Reggie. Then compare to on-ball legends like Magic, Nash, Luka, Bron.

Which one stands out more as PMs? The on-ball ones, and it shouldn't be surprising. Not that we shouldn't value off-ball PM, but not close to extent as on-ball PM, and I have yet to see evidence saying otherwise.

[lessthanjake]

Indeed, in my analysis, the inclusion of later years brought Steph down a good bit, so the above tweet probably *understates* Steph’s lead in the relevant timeframe. So yeah, this adds to the evidence that Steph’s “creation” outdoes other players, including LeBron. Which shouldn’t be surprising to those who see how his presence warps defenses and uniquely pulls defenders *away* from the basket, which is better than pulling defenders towards the basket (since opening up space near the basket is better than opening up space away from the basket).

That's not what creation is. Creation is made up of many things

Delivering the ball to your teammates (delivery quality)
Taking out defenders
The open-ness of shots

And... anyone who knows basketball, knows on-ball playmaking and scoring is far more impactful than off-ball playmaking. That's just how basketball works!

Take a look at any ATG off-ball payer like Dirk, Bird, Reggie. Then compare to on-ball legends like Magic, Nash, Luka, Bron.

Which one stands out more as PMs? The on-ball ones, and it shouldn't be surprising. Not that we shouldn't value off-ball PM, but not close to extent as on-ball PM, and I have yet to see evidence saying otherwise.

My post actually didn’t say anything about off-ball or on-ball playmaking. You read that in yourself and then argued with it. Everyone creates to at least some degree on-ball and off-ball, and shot quality data tells us that Steph’s combination of the two is better than anyone else’s at getting his teammates good looks. That’s what matters.

But, since you raise the issue, I do want to address this idea that on-ball playmaking is better than off-ball playmaking. This is something I’ve seen discussed a fair bit elsewhere recently, and I think there’s some important concepts to note. A player will almost always draw more attention on the ball than off the ball, because he actually is a more immediate threat when he has the ball. Whether it’s Steph Curry or anyone else, they will almost certainly draw more attention with the ball. That’s the intuition you’re running with here. But you’re not reasoning through the whole picture here. In particular, you’re ignoring that *someone* is on the ball. If Steph Curry is off the ball, he might individually be drawing a little less attention than he draws when he’s on the ball, but someone on his team is on the ball and is drawing more attention than they normally would. The net result for the team might warp the defense more than if Steph were on the ball, particularly if Steph draws almost as much attention off the ball as on it and his teammate draws far more attention on the ball than off it. It’s basically the concept of comparative advantage. And since very few players warp defenses off the ball anything like Steph does, it is easy to imagine it being the case that Steph being off the ball allows for a unique maximization of the overall pressure on defenses, beyond what on-ball playmakers can have their teams achieve (and perhaps beyond what other off-ball playmakers could). The other thing to remember here is that even the most ball-dominant guys don’t actually *always* have the ball. Even if you thought the pressure on the defense is always maximized when the superstar has the ball (which, again, isn’t necessarily right), the overall picture will be maximized throughout a possession if the superstar is able to exert lots of pressure on the defense during the inevitable time when he doesn’t have the ball.

In any event, the above is all fairly abstract, but when the shot quality data tells us that Steph’s presence on the court improves the shot quality of his teammates more than any other player’s presence does (including LeBron), then the conclusion here is obvious about who is creating the most for their teammates, regardless of how you think that creation works.

[metta-tonne]

I was just reminded of this:

Read on Twitter

This uses synergy shot quality data and finds that Steph is alone at the top in terms of increasing the shot quality of his teammates. It’s not the right timeframe (2018-2022, rather than including 2015), but the results are consistent with the shot-quality analysis I did using PBPstats shot quality data stretching back to earlier years that include this 2015 season (at the following link and also spoilered below: https://forums.realgm.com/boards/viewtopic.php?p=107624238#p107624238)

Spoiler:

I want to also just provide some information that I think goes to the playmaking effect of Steph Curry’s gravity.

www.pbpstats.com has data on “shot quality” that a player’s teammates have had with and without that player on the floor. This is a bit of a back-of-the-napkin method, but I looked at regular season + playoffs and then averaged the shot quality increase with certain star players on the floor, for teammates that played at least 1,000 minutes with the given player. I chose a set of great playmakers and some other great players, and tried to use timeframes correlating with their best years and/or their stints with specific teams (note: the data starts at 2000-2001, so that’s why no timeframe includes years before that). The idea was to measure to what extent a player causes his teammates to get good looks. Here’s what I found:

Average Shot Quality Increase for Teammates with Player on the Floor (only teammates with 1,000+ minutes with the player)

1. Steph Curry (2014-2015 to 2018-2019): +5.1%

2. Steph Curry (2014-2015 to 2022-2023): +3.8%

3. Nikola Jokic (2020-2021 to 2022-2023): +3.7%

4. LeBron James (2014-2015 to 2017-2018): +2.9%

5. Steve Nash (2004-2005 to 2011-2012): +2.6%

6. LeBron James (2005-2006 to 2009-2010): +2.3%

7. James Harden (2012-2013 to 2019-2020): +2.2%

8. Draymond Green (2014-2015 to 2022-2023): +2.2%

9. Shaquille O’Neal (2000-2001 to 2003-2004): +1.9%

10. Kobe Bryant (2000-2001 to 2009-2010):: +1.9%

11. Tim Duncan (2000-2001 to 2006-2007): +1.7%

12. Chris Paul (2011-2012 to 2016-2017): +1.5%

13. Giannis Antetokounmpo (2019-2020 to 2022-2023): +1.5%

14. Luka Doncic (2019-2020 to 2022-2023): +1.3%

15. LeBron James (2010-2011 to 2013-2014): +1.0%

Indeed, in my analysis, the inclusion of later years brought Steph down a good bit, so the above tweet probably *understates* Steph’s lead in the relevant timeframe. So yeah, this adds to the evidence that Steph’s “creation” outdoes other players, including LeBron. Which shouldn’t be surprising to those who see how his presence warps defenses and uniquely pulls defenders *away* from the basket, which is better than pulling defenders towards the basket (since opening up space near the basket is better than opening up space away from the basket).

Indeed, in my analysis, the inclusion of later years brought Steph down a good bit, so the above tweet probably *understates* Steph’s lead in the relevant timeframe. So yeah, this adds to the evidence that Steph’s “creation” outdoes other players, including LeBron. Which shouldn’t be surprising to those who see how his presence warps defenses and uniquely pulls defenders *away* from the basket, which is better than pulling defenders towards the basket (since opening up space near the basket is better than opening up space away from the basket).

That's not what creation is. Creation is made up of many things

Delivering the ball to your teammates (delivery quality)
Taking out defenders
The open-ness of shots

And... anyone who knows basketball, knows on-ball playmaking and scoring is far more impactful than off-ball playmaking. That's just how basketball works!

Take a look at any ATG off-ball payer like Dirk, Bird, Reggie. Then compare to on-ball legends like Magic, Nash, Luka, Bron.

Which one stands out more as PMs? The on-ball ones, and it shouldn't be surprising. Not that we shouldn't value off-ball PM, but not close to extent as on-ball PM, and I have yet to see evidence saying otherwise.

Um but they did post evidence. Shot quality is definitely important and part of why we value playmaking.

Saying that I don't really see how this helps Steph here. Yeah Steph still in his prime is 1st but Lebron way past it is only .2 off?

Also there's a weird choice of choosing 2011-2014 Lebron and filtering out role players. Which means we're talking mostly about Lebron not being able to create a bunch of space for Dwayne Wade which sure but that's not really too relevant in almost any situation.

And to me Lebron being that close to Steph as a playmaker when we include 2021 and 2022 goes a little against the story you're telling here.

Also excluding catch and shoot threes with Lebron is um meh

[BusywithBball]

Okay i say I don’t like just saying like fact how things are. I think they posted really useful info and even if you think it’s wrong you shouldn’t just ignore it because whatever.

Lebron score surprised me too though. I didn’t think he was a top 4 playmaker still. Both are great ofc and this is a tough year for sure. I think because of the winning and way warrior changed the sport i have to say Steph but it close. Lebron was awesome for sure and his finals was amazing.

We have a great finals now too

[lessthanjake]

Um but they did post evidence. Shot quality is definitely important and part of why we value playmaking.

Saying that I don't really see how this helps Steph here. Yeah Steph still in his prime is 1st but Lebron way past it is only .2 off?

That’s actually a big difference. It’s basically saying you cause your teammates to have a 2% higher expected FG%. There should be no question that that’s a big difference that we’d expect to have a major effect on a team’s ORTG. For instance, a 2% increase in eFG% this season would be enough for a team to go from an average team to the 4th best eFG% in the league, just above the Pacers. And the analysis I did that used PBPstats data for an earlier timeframe (which included the 2015 season that this thread is about) showed an even bigger difference.

Also there's a weird choice of choosing 2011-2014 Lebron and filtering out role players. Which means we're talking mostly about Lebron not being able to create a bunch of space for Dwayne Wade which sure but that's not really too relevant in almost any situation.

I definitely did not “filter[] out role players.” As noted in my post, the bar I used was that someone played 1,000 minutes with the player in the entire timespan. To take your LeBron-in-Miami example, that would include 13 players. The players that played most with LeBron in Miami that did not make the list were Bibby and Arroyo—who were both barely there.

And to me Lebron being that close to Steph as a playmaker when we include 2021 and 2022 goes a little against the story you're telling here.

Except that the PBPstats shot quality analysis shows LeBron far behind Steph in the earlier timeframes of his career too. This isn’t a conclusion that is exclusive to any one timeframe. And actually, if anything, LeBron looks closer in later years than he did in earlier years (which isn’t very surprising, since LeBron actually kept developing even more as a playmaker, and that 2018-2022 timeframe actually includes two of his top few best playmaking years IMO and was his 2nd highest five-year span in assists per game).

Also excluding catch and shoot threes with Lebron is um meh

The shot quality data does not exclude catch and shoot threes.

[lessthanjake]

Okay i say I don’t like just saying like fact how things are. I think they posted really useful info and even if you think it’s wrong you shouldn’t just ignore it because whatever.

Lebron score surprised me too though. I didn’t think he was a top 4 playmaker still. Both are great ofc and this is a tough year for sure. I think because of the winning and way warrior changed the sport i have to say Steph but it close. Lebron was awesome for sure and his finals was amazing.

We have a great finals now too

You didn’t think that LeBron was a top 4 playmaker over the course of 2018-2022? That’s very close to his best five-year timespan in terms of assists per game (second only to the nearby 2017-2021 span)! And it includes arguably his two best playmaking years (2018 and 2020). Would be a pretty remarkable indictment of his playmaking if he wasn’t even a top 4 playmaker in that span.

[Elpolo_14]

Indeed, in my analysis, the inclusion of later years brought Steph down a good bit, so the above tweet probably *understates* Steph’s lead in the relevant timeframe. So yeah, this adds to the evidence that Steph’s “creation” outdoes other players, including LeBron. Which shouldn’t be surprising to those who see how his presence warps defenses and uniquely pulls defenders *away* from the basket, which is better than pulling defenders towards the basket (since opening up space near the basket is better than opening up space away from the basket).

That's not what creation is. Creation is made up of many things

Delivering the ball to your teammates (delivery quality)
Taking out defenders
The open-ness of shots

And... anyone who knows basketball, knows on-ball playmaking and scoring is far more impactful than off-ball playmaking. That's just how basketball works!

Take a look at any ATG off-ball payer like Dirk, Bird, Reggie. Then compare to on-ball legends like Magic, Nash, Luka, Bron.

Which one stands out more as PMs? The on-ball ones, and it shouldn't be surprising. Not that we shouldn't value off-ball PM, but not close to extent as on-ball PM, and I have yet to see evidence saying otherwise.

My post actually didn’t say anything about off-ball or on-ball playmaking. You read that in yourself and then argued with it. Everyone creates to at least some degree on-ball and off-ball, and shot quality data tells us that Steph’s combination of the two is better than anyone else’s at getting his teammates good looks. That’s what matters.

But, since you raise the issue, I do want to address this idea that on-ball playmaking is better than off-ball playmaking. This is something I’ve seen discussed a fair bit elsewhere recently, and I think there’s some important concepts to note. A player will almost always draw more attention on the ball than off the ball, because he actually is a more immediate threat when he has the ball. Whether it’s Steph Curry or anyone else, they will almost certainly draw more attention with the ball. That’s the intuition you’re running with here. But you’re not reasoning through the whole picture here. In particular, you’re ignoring that *someone* is on the ball. If Steph Curry is off the ball, he might individually be drawing a little less attention than he draws when he’s on the ball, but someone on his team is on the ball and is drawing more attention than they normally would. The net result for the team might warp the defense more than if Steph were on the ball, particularly if Steph draws almost as much attention off the ball as on it and his teammate draws far more attention on the ball than off it. It’s basically the concept of comparative advantage. And since very few players warp defenses off the ball anything like Steph does, it is easy to imagine it being the case that Steph being off the ball allows for a unique maximization of the overall pressure on defenses, beyond what on-ball playmakers can have their teams achieve (and perhaps beyond what other off-ball playmakers could). The other thing to remember here is that even the most ball-dominant guys don’t actually *always* have the ball. Even if you thought the pressure on the defense is always maximized when the superstar has the ball (which, again, isn’t necessarily right), the overall picture will be maximized throughout a possession if the superstar is able to exert lots of pressure on the defense during the inevitable time when he doesn’t have the ball.

In any event, the above is all fairly abstract, but when the shot quality data tells us that Steph’s presence on the court improves the shot quality of his teammates more than any other player’s presence does (including LeBron), then the conclusion here is obvious about who is creating the most for their teammates, regardless of how you think that creation works.

I think the concepts of On-ball playmaking being better than Off-ball Playmaking come down to the fundamental aspect of who initiat the Overall offense and make the Decision of the team. With an On-ball engine the offense will be organize by the best Offensive player in the team - he will be the one who brings the attention on him ( like you mentioned onball bring more defensive attention than offball due to the scoring aspect ) and he's the one to force defense into an disadvantageous situation by collapsing or rotation ( by Driving/ Posting/ PnR ect. ) then the creation or Opportunities will be operate by best player whom have elite traits there as Passing / Timing / Court awareness to deliver the ball so it less Prone to mistake or lower quality due to another teammates being the decision maker or the ball destributor ( who could easier miss a read or not usetalise the play that was made by the off-ball player ). The on-ball PM can be generate more sustainable in almost any situation and not too dependented on teammates or Scheme running by the team. Off-ball Playmaking is elite in a Ceilings rising situation with many Elite teammates who have traits to be the one facilliting the offense around that off-ball player who creates separation for some else but it too variable in any other situation without a great ball carrier with elite decision making. BOTH TYPES of playmaking sure does create opportunities for the team but the difference will be dependent on Delivery and how the advantage is put to action as fast as possible which I think an On-ball playmaker is more reliable ( Will more frequently capitalize on his creation due to being the one deciding those flow of the offense )on that aspect while still maintaining equal/if not more attention ( depends on player but the point is there )

Anyway I do think the mix of both worlds is the best tho

[Elpolo_14]

This thread stays open till the end of the Finals. More people should vote in the coming days.

I thought the time limit was 14-15 June or something like that. Didn't know it would go longer but that great because there not enough people voting on this thread IMO. Hope we will get more points of view and people who share their thoughts on this POY thread ( would be interesting to read )

[Elpolo_14]

Okay i say I don’t like just saying like fact how things are. I think they posted really useful info and even if you think it’s wrong you shouldn’t just ignore it because whatever.

Lebron score surprised me too though. I didn’t think he was a top 4 playmaker still. Both are great ofc and this is a tough year for sure. I think because of the winning and way warrior changed the sport i have to say Steph but it close. Lebron was awesome for sure and his finals was amazing.

We have a great finals now too

I wanna know what is the reasoning or criteria to not have lebron in top 4 playmaker?

He one of the best defensive manipulator and put defense in disadvantageous position ( all time as methodical possession and tempo controller ). His Scoring traits and rim pressure always force defense to collapse on him and leaving Bron teammates open/at least have a larger room to operate. He's elite passing skills with many varietion to deliver the pass off. He one of the best ( if not the best ) Transition initiator ever. His court vision and decision making on opportunities he created are elite. He's a good off-ball with cutting/ Slashing + good Screener he even have position awareness offball.

I just wanna have inside on your way of viewing Playmaking. Have a nice day

[AEnigma]

Kept this open for the Finals, but now setting a firmer closing deadline of 12pm EST on Friday, June 20 (~48 hours from now). Either the Finals will be over, or I would rather full attention be on a hypothetical Game 7 anyway.

[metta-tonne]

Um but they did post evidence. Shot quality is definitely important and part of why we value playmaking.

Saying that I don't really see how this helps Steph here. Yeah Steph still in his prime is 1st but Lebron way past it is only .2 off?

That’s actually a big difference. It’s basically saying you cause your teammates to have a 2% higher expected FG%. There should be no question that that’s a big difference that we’d expect to have a major effect on a team’s ORTG. For instance, a 2% increase in eFG% this season would be enough for a team to go from an average team to the 4th best eFG% in the league, just above the Pacers. And the analysis I did that used PBPstats data for an earlier timeframe (which included the 2015 season that this thread is about) showed an even bigger difference.

Also there's a weird choice of choosing 2011-2014 Lebron and filtering out role players. Which means we're talking mostly about Lebron not being able to create a bunch of space for Dwayne Wade which sure but that's not really too relevant in almost any situation.

I definitely did not “filter[] out role players.” As noted in my post, the bar I used was that someone played 1,000 minutes with the player in the entire timespan. To take your LeBron-in-Miami example, that would include 13 players. The players that played most with LeBron in Miami that did not make the list were Bibby and Arroyo—who were both barely there.

And to me Lebron being that close to Steph as a playmaker when we include 2021 and 2022 goes a little against the story you're telling here.

Except that the PBPstats shot quality analysis shows LeBron far behind Steph in the earlier timeframes of his career too. This isn’t a conclusion that is exclusive to any one timeframe. And actually, if anything, LeBron looks closer in later years than he did in earlier years (which isn’t very surprising, since LeBron actually kept developing even more as a playmaker, and that 2018-2022 timeframe actually includes two of his top few best playmaking years IMO and was his 2nd highest five-year span in assists per game).

Also excluding catch and shoot threes with Lebron is um meh

The shot quality data does not exclude catch and shoot threes.

I don't know man.

https://web.archive.org/web/20210707132926/https://backpicks.com/2017/06/27/star-player-effects-on-teammate-efficiency/

2015 to 2017 matters more here, right? Lebron's leads all the players in this one and there's that one with the catch and shoots which also includes 2015? I'm not trying to say this proves Lebron is the better playmaker but I don't think you should be confident either. Assists underrate both probably and I think it's alot harder to tell who would be more let down by the APG than just picking this or that on/off shot quality thing and talking about gravity. And whoever we want to say is the better playmaker I don't feel that means when Lebron is on his own really we should be looking at a 3 or 5 years and say that applies to when he's kind of just left as the whole offense.

[trelos6]

OPOY

1.Stephen Curry. #1 playmaker, got better as a passer. 26.6 pp75 on +10.4 rTS%. Playoffs: 27.7, +7.3%. Team rOrtg of +6.

2.Chris Paul. Top 3 playmaker, leagues best passer. 20.8 pp75 on +6.2 rTS%. Playoffs: 21.7, +9.3%. Team rOrtf of +6.8.

3.Lebron James. Top 5 playmaker. 27.2 pp75 on +4.3 rTS%. Playoffs: 28.3, -4.7%. Team rOrtg of +5.5.

HM: James Harden. Top 8 playmaker. 27.8 pp75 on +7.1 rTS%. Playoffs: 25.7, +8.6%. Team rOrtg of +1.4.

DPOY

1.Draymond Green. Anchored the best D in the league.

2.Rudy Gobert. Great rim protection.

3.Kawhi Leonard. One of the best perimeter D seasons, from the wing.

POY

1.Steph Curry. +6.41 OPIPM, +1.73 DPIPM. +8.14 PIPM. 23.5 Wins Added. Dominant regular season team based off solid defense and Curry driving the offense. Played well in playoffs, but they faced some strong defenses in Memphis and Cleveland. So it makes sense that his efficiency took a hit in those series.

2.Lebron James. +4.79 OPIPM, +1.83 DPIPM. +6.62 PIPM. 19.9 Wins Added. Strong regular season. Cruised through the East, then lost all help. Put up some amazing counting stats against a great defense, and there were talks about winning FMVP from the losing side.

3.Chris Paul. +6.48 OPIPM, +0.57 DPIPM. +7.05 PIPM. 20.28 Wins Added. Strong regular season. Was great vs the Spurs in one of the best first round series of all time. Missed a few games vs Houston, but was generally pretty good. Not his fault that Smith / Brewer went off in the 4th.

4.Kawhi Leonard. Starts to blossom as a scorer, alongside his amazing perimeter defense. 20 pp75 on +3.3 rTS%. Slightly upped scoring volume and efficiency in one playoff series. +2.31 OPIPM, +2.77 DPIPM, +5.08 PIPM. 11.76 Wins Added.

5.Anthony Davis. Strongly considered Draymond, Westbrook and Harden. 26.6 pp75 on +5.7 rTS%. Great rim protection. Was great vs Warriors despite the sweep: 28.6, +7.9%, 3 bpg. +3.42 OPIPM. +2.52 DPIPM. +5.94 PIPM. 14.73 Wins Added.

[lessthanjake]

Um but they did post evidence. Shot quality is definitely important and part of why we value playmaking.

Saying that I don't really see how this helps Steph here. Yeah Steph still in his prime is 1st but Lebron way past it is only .2 off?

That’s actually a big difference. It’s basically saying you cause your teammates to have a 2% higher expected FG%. There should be no question that that’s a big difference that we’d expect to have a major effect on a team’s ORTG. For instance, a 2% increase in eFG% this season would be enough for a team to go from an average team to the 4th best eFG% in the league, just above the Pacers. And the analysis I did that used PBPstats data for an earlier timeframe (which included the 2015 season that this thread is about) showed an even bigger difference.

Also there's a weird choice of choosing 2011-2014 Lebron and filtering out role players. Which means we're talking mostly about Lebron not being able to create a bunch of space for Dwayne Wade which sure but that's not really too relevant in almost any situation.

I definitely did not “filter[] out role players.” As noted in my post, the bar I used was that someone played 1,000 minutes with the player in the entire timespan. To take your LeBron-in-Miami example, that would include 13 players. The players that played most with LeBron in Miami that did not make the list were Bibby and Arroyo—who were both barely there.

And to me Lebron being that close to Steph as a playmaker when we include 2021 and 2022 goes a little against the story you're telling here.

Except that the PBPstats shot quality analysis shows LeBron far behind Steph in the earlier timeframes of his career too. This isn’t a conclusion that is exclusive to any one timeframe. And actually, if anything, LeBron looks closer in later years than he did in earlier years (which isn’t very surprising, since LeBron actually kept developing even more as a playmaker, and that 2018-2022 timeframe actually includes two of his top few best playmaking years IMO and was his 2nd highest five-year span in assists per game).

Also excluding catch and shoot threes with Lebron is um meh

The shot quality data does not exclude catch and shoot threes.

I don't know man.

https://web.archive.org/web/20210707132926/https://backpicks.com/2017/06/27/star-player-effects-on-teammate-efficiency/

2015 to 2017 matters more here, right? Lebron's leads all the players in this one and there's that one with the catch and shoots which also includes 2015? I'm not trying to say this proves Lebron is the better playmaker but I don't think you should be confident either. Assists underrate both probably and I think it's alot harder to tell who would be more let down by the APG than just picking this or that on/off shot quality thing and talking about gravity. And whoever we want to say is the better playmaker I don't feel that means when Lebron is on his own really we should be looking at a 3 or 5 years and say that applies to when he's kind of just left as the whole offense.

A few important things on this:

1. There’s some definite mistakes in that analysis. It says it looked at players who played 1,000 minutes with the players in those seasons, but if we go to PBPstats we find that Ben included teammates that shouldn’t be included by that criteria and didn’t include teammates that should’ve been included. For instance, Ezeli was included but did not play 1,000 minutes with Steph, while Pachulia was not included but did. Jefferson played 1,000 minutes with LeBron and wasn’t included while Korver didn’t play that many minutes and was included.

2. That page is not about shot quality, but rather just comparing scoring efficiency with and without a player. I was focused on shot quality in part because it does away with the potential randomness of how a player’s shots happen to go. That’s especially an issue when looking at relatively small samples—which this is for some players, particularly when only looking at a three-year span. For instance, you have a guy like Channing Frye, who took less than 250 shots without LeBron in that timeframe, or Bogut who shot just 73 shots without Steph. The other reason I was focused on shot quality is that it’s the most comparable to the “creation” analysis others have been doing, in the sense that both analyses are looking at the quality of the looks created, rather than the actual result of those shots.

3. If we did look at scoring efficiency and actually included the right players, then we’d find that, in total, from 2015-2017, the teammates that played 1,000 minutes with Steph scored 1.197 points per TS attempt with Steph on the court, and 1.006 points per TS attempt with Steph off the court. That is a difference of +0.191 points per TS attempt for Steph’s teammates when Steph was on the floor. Meanwhile, in 2015-2017, the teammates that played 1,000 minutes with LeBron scored 1.164 points per TS attempt with LeBron on the court, and 1.065 points per TS attempt with LeBron off the court. That is a difference of +0.099 points per TS attempt for LeBron’s teammates when LeBron was on the floor. So the scoring efficiency of their teammates improved a lot more with Steph on the floor than they did with LeBron on the floor.

So yeah, the analysis you linked to has definite mistakes, and is also measuring something different than shot quality. But even if we looked at what that link is looking at using the minutes threshold that page claimed to be using over the seasons it says it was using (2015-2017), Steph’s teammates improved their scoring efficiency *a lot* more with Steph on the court than LeBron’s teammates did with LeBron on the floor. This is a strong data point for Steph!

It’s not clear to me how Ben came to a different conclusion. Maybe the above-mentioned mistakes changed the analysis, but I *think* it might largely be a result of him simply taking an average of the differences for each player (not clear if that’s what he did, so this is just a guess), which in a sample size that small can be super noisy for some of the less used players (and then that super noisy result gets weighted a lot). Again, a player like Frye shot less than 250 shots with LeBron off, Bogut had 73 shots with Steph off, etc. I did a similar method in my analysis using PBPstats shot quality (while I believe the tweet using Synergy data was based on a weighted average), but shot quality at least inherently takes away random shooting variance, and I looked at longer timespans (i.e. generally with bigger samples for the players), so it was less of an issue (though, admittedly, my PBPstats analysis probably could be improved by doing a weighted average of shot quality, weighted by number of attempts—it would just take a long time for me to do that).

[OldSchoolNoBull]

Alright, I'll get a vote in.

POY

1. Stephen Curry

The top 2 has been between LeBron and Steph for everyone, and going by the votes that have already been cast, it appears that LeBron will win this, but I'm going Steph. I just think he has the numbers, in addition to winning the ring.

RAPM(via thebasketballdatabase.com)
Steph RS: 5.39
LeBron RS: 4.73
Steph PO: 3.01
LeBron PO: 2.43

On/Off
Steph RS: +18.1
LeBron RS: +16.6
Steph PO: +7.6
LeBron PO: +0.8

(Although here I will stipulate that I don't believe that +0.8 is an accurate reflection of LeBron's playoff performance and almost certainly undersells his impact.)

Box Composites
Steph RS: .288 WS/48, 9.9 BPM
LeBron RS: .199 WS/48, 7.1 BPM
Steph PO: .228 WS/48, 8.8 BPM
LeBron PO: .173 WS/48, 7.9 BPM

Scoring Efficiency
Steph RS: 63.8% TS, 44.3% 3P, 48.7% FG, 91.4% FT
LeBron RS: 57.7% TS, 35.4% 3P, 48.8% FG, 71% FT
Steph PO: 60.7% TS, 42.2% 3P, 45.6% FG, 83.5% FT
LeBron PO: 48.7% TS, 22.7% 3P, 41.7% FG, 73.1% FT

It's just a lot of stats that appear to indicate that Steph performed better not just in the regular season, but in the playoffs. And another thing to consider is that Steph's playoff performance came against, imo, a tougher path to the Finals.

In Round 1, the Warriors defeated an AD-led Pelicans team(with Jrue Holiday a #2) with 45 wins, 1.13 SRS, and +0.9 Net Rtg, while the Cavs defeated an Isaiah Thomas-led Celtics team with 40 wins, -0.40 SRS, and +0.2 Net Rtg.

In Round 2, the Warriors defeated the Grit'n'Grind Grizzlies led by Marc Gasol and Mike Conley with 55 wins, 3.62 SRS, and +3.5 Net Rtg, while the Cavs defeated a Bulls team led by older Pau Gasol, just-emerging Jimmy Butler, and post-injury Derrick Rose, with 50 wins, 2.54 SRS, and +3.2 Net Rtg(and were a shot away from going down 3-1 to that Bulls team).

In the CF, the Warriors defeated a Rockets team led by Harden and Dwight with 56 wins, 3.82 SRS, and +3.5 Net Rtg, while the Cavs defeated the #1 seeded Hawks with 60 wins, 4.75 SRS, and +5.8 Net Rtg. While the Hawks were the better team by those numbers, they are remembered as paper tigers for a reason, and I would suggest the Rockets were a tougher out in the playoffs. Look at the two teams' respective scoring numbers in RS vs PO:

Rockets RS: 103.9ppg on 54.8% TS
Hawks RS: 102.5ppg on 56.3% TS
Rockets PO: 108.2ppg on 54.4% TS
Hawks PO: 98ppg on 52.1% TS

So while the Rockets scored over 4ppg more on less than half a point less efficiency in the playoffs, the Hawks scored 4.5ppg less on over four points less efficiency in the playoffs.

It's also worth noting who the Warriors' opponents themselves beat before playing the Warriors. The Grizzlies defeated Damian Lillard's Blazers who had won 51 games with 4.41 SRS and +4.5 Net Rtg; the Rockets had defeated CP3's Clippers who had won 56 games with 6.80 SRS and +6.9 Net Rtg(and who themselves had eliminated the defending champion Spurs) and before that, a Mavs team led by old Dirk that had won 50 games with 3.36 SRS and +3.0 Net Rtg.

In contrast, the Bulls(whom the Cavs defeated in the second round) had defeated a Brandon Jennings-led Bucks team that had won 41 games with -0.09 SRS and +0.4 Net Rtg; the Hawks had defeated a Wizards team that won 46 games with 0.17 SRS and +0.8 Net Rtg, and before that a Raptors team that had won 49 games with 2.45 SRS and +3.3 Net Rtg.

In short, the West was a bloodbath and Steph and the Warriors came out of it with Steph still putting up all the above numbers.

Of course the obvious counter to all of this, at least as far as the playoffs go, is that Steph had a team around him and LeBron's #2 and #3 were hurt. That's totally fair. I just think there's a lot of numbers to overcome here.

And for the record, with all due respect to Iggy, who was a very good player, I thought him winning the Finals MVP over Steph was silly then, and I still do.

2. LeBron James

With all of the above said, there's still no way to argue anyone else over LeBron. What he did in the Finals alone - taking two games off the champs by himself - is enough to justify the #2 spot.

3. James Harden

Looked at Harden and CP3 here, and it's tough. CP3's got the higher RAPM, he beat the defending champion Spurs in the first round, and came within a game of defeating Harden and the Rockets(and looking at CP3's stats in the last three games of that series, you can't say it's his fault the Clippers blew a 3-1 lead). I think what's got me leaning Harden here is he had less help.

CP3 had Griffin and Redick and Crawford and DAJ.

In the regular season, Dwight Howard only played 41 games. And the WOWY doesn't suggest a great deal of impact from him - 29-12 with, 27-14 without. And after Dwight, Harden's teammates were Patrick Beverley, Trevor Ariza, Donatas Motijunas, Terrance Jones, Corey Brewer, old Jason Terry(four years after his Mavs title).

I just feel like that roster had no business getting as far as it did, so I'm giving #3 to Harden.

4. Chris Paul

It was a fantastic season with sparkling impact metrics, great team numbers, and a classic series-winning shot against the defending champs. I'd have made him #3 if I didn't feel like Harden had less help.

5. Russell Westbrook

This'll be the most surprising choice on my list for sure. I've never been much of a Westbrook fan, but in a season where Durant only played 27 games, Westbrook kept the team afloat with 45 wins, 2.47 SRS, and +2.3 Net Rtg, and very nearly made the playoffs - they tied with AD's Pelicans for #8, but the Pelicans had won the season series and thus got the ticket.

Westbrook played 67 games, and the Thunder were 40-27(.597) with him and 5-10(.333) without.

His best teammates with KD out were Ibaka, second-year Steven Adams, and the other players getting over 20mpg? Enes Kanter, Dion Waiters, Reggie Jackson, Anthony Morrow, DJ Augustin, Lance Thomas, and Sebastian Telfair, with Kendrick Perkins and defensive specialist Andre Roberson getting a shade under 20mpg.

AD at least had Jrue Holiday(for half the season anyway), Eric Gordon, Tyreke Evans, and defensive anchor Omer Asik; Kawhi was part of a championship-defending ensemble featuring multiple HOFers.

[lessthanjake]

I wonder if it’s really that airtight to put LeBron above Harden. I think I ultimately would, because he made the Finals and Harden didn’t, and I value that achievement in the context of POY. But I think there’s a pretty good argument that Harden played better in the regular season, and played as well or better in the playoffs but just happened to face the Warriors earlier.

Some regular season impact numbers:

EPM

Harden: 6.9
LeBron: 5.8

RAPTOR

Harden: 7.5
LeBron: 5.0

MAMBA

Harden: 8.473
LeBron: 9.239

LEBRON

Harden: 5.24
LeBron: 4.99

GitLab one-year RAPM

Harden: 4.5111
LeBron: 4.8377

BasketballDatabase one-year RAPM

Harden: 3.56
LeBron: 4.73

And here’s some regular season box numbers:

BPM

Harden: 8.8
LeBron: 7.1

Thinking Basketball BPM

Harden: 6.1
LeBron: 5.7

WS/48

Harden: 0.265
LeBron: 0.199

Overall, this is not entirely conclusive, since there’s data going both ways, but the data picture does go in Harden’s direction overall IMO. And when we account for the fact that Harden played 20% more minutes than LeBron during the regular season, it’s pretty easy to conclude that Harden was better in the regular season. Which, FWIW, was something that the MVP voters agreed with—putting Harden in 2nd, and well above LeBron in 3rd. This conclusion is, of course, bolstered by the fact that Harden lifted the Rockets to 56 wins, despite Dwight missing half the season.

Okay, but what about the playoffs? Normally, this would be a massive advantage for LeBron. But was it a massive advantage for LeBron in this particular year? Well, no. Here’s some playoff data:

Playoff EPM

Harden: 4.7
LeBron: 2.9

Playoff BPM

Harden: 7.6
LeBron: 7.9

Playoff Thinking Basketball BPM

Harden: 5.5
LeBron: 6.5

Playoff WS/48

Harden: 0.202
LeBron: 0.173

That’s basically a wash, and Harden is ahead in the metric that is probably the best one (EPM).

And then we think about who they actually played, and probably more importantly, who they actually beat. The Rockets had *easily* the most impressive series win of these two teams—beating the 6.80 SRS Chris Paul Clippers. Yes, the Cavaliers easily beat the Hawks while not being full strength, but those Clippers were definitely a much better team IMO. That was a really high-quality series win, against a genuinely great team (who had just beaten the defending-champion Spurs)—which is not something the Cavaliers have on their resume (though, to be fair, we don’t know if they were capable of it or not—the East was just weak).

So what about when they faced the Warriors? The Cavs managed to win 2 games, while the Rockets only won 1 game. However, for one thing, the average MOV for the Warriors was very similar (8.2 vs. 7.2). Perhaps more importantly, Harden put up 28/8/6 on 63% TS% in that series. LeBron’s counting stats were higher (36/13/9), but it was on 47.7% TS%. Of course, the Cavs’ injuries surely inflated LeBron’s counting stats while deflating the efficiency (though LeBron was shooting poorly in these playoffs before the injuries). Yes, LeBron did play better defense than Harden. Overall, if someone wants to look at rosters and compare number of wins in the series, then they could convince themselves LeBron played better against the Warriors than Harden did, but I think Harden did actually play better individually. And the Rockets having better players than the Cavs had in those Finals doesn’t mean they actually played better in that small sample of one series. I think to conclude LeBron played better than Harden against the Warriors (or certainly meaningfully better) you’d have to really just think that teammate performance is a constant, which is really not at all true.

____________

So yeah, my conclusion here is that Harden was better in the regular season and played more, and that Harden played equally well or better in the playoffs, actually had a series win against a great team, and probably played better individually against the Warriors. I think I’d still probably put LeBron ahead, because the sheer achievement of getting to the Finals matters to me. But I think there’s a good argument for Harden over LeBron, and certainly a good argument that he actually played better individually in this year. I’ll conclude by noting that this shouldn’t really be a controversial idea—after all, like 40% of voters put Harden above LeBron in the original POY vote for this season.

[jalengreen]

And then we think about who they actually played, and probably more importantly, who they actually beat. The Rockets had *easily* the most impressive series win of these two teams—beating the 6.80 SRS Chris Paul Clippers. Yes, the Cavaliers easily beat the Hawks while not being full strength, but those Clippers were definitely a much better team IMO. That was a really high-quality series win, against a genuinely great team (who had just beaten the defending-champion Spurs)—which is not something the Cavaliers have on their resume (though, to be fair, we don’t know if they were capable of it or not—the East was just weak).

Hmph

Would start by throwing in the pre-series implied probabilities to get a sense of how the market viewed the respective series:

- CLE 63.45% - 36.55% ATL

- HOU 57.98% - 42.02% LAC

Okay, so Cleveland was the bigger favorite.

Injuries complicate this both ways but I think ignoring them is the simplest thing and is probably relatively fair. Love/CP3 injuries were known going in. Korver missing two games and Kyrie playing only 49 minutes were definitely not known, but whatever.

What seems more notable is how the series were won:

- The Cavaliers won 4-0 with a +13.3 MOV against a +4.75 SRS team while having a 63% market implied pre-series win probability

- The Rockets won 4-3 with a -3.1 MOV against a +6.80 SRS team while having a 58% market implied pre-series win probability

I personally wouldn't say that the Rockets had *easily* the more impressive series win.

And if we got into the individual component, worth mentioning that in Game 6, James Harden got benched with the Rockets down 17 in with 1:33 to go in the third quarter while down 3-2 in the series. Houston proceeded to finish the game on a 47-18 run without Harden to force a Game 7. He had a -21 +/- in the game and the second lowest +/- on the team was -1. We wouldn't be talking about a 'series win' at all without the efforts of Josh Smith and Corey Brewer while Harden had a towel on his head on the bench and likely presumed that his season was over, fwiw.

[lessthanjake]

And then we think about who they actually played, and probably more importantly, who they actually beat. The Rockets had *easily* the most impressive series win of these two teams—beating the 6.80 SRS Chris Paul Clippers. Yes, the Cavaliers easily beat the Hawks while not being full strength, but those Clippers were definitely a much better team IMO. That was a really high-quality series win, against a genuinely great team (who had just beaten the defending-champion Spurs)—which is not something the Cavaliers have on their resume (though, to be fair, we don’t know if they were capable of it or not—the East was just weak).

Hmph

Would start by throwing in the pre-series implied probabilities to get a sense of how the market viewed the respective series:

- CLE 63.45% - 36.55% ATL

- HOU 57.98% - 42.02% LAC

Okay, so Cleveland was the bigger favorite.

Injuries complicate this both ways but I think ignoring them is the simplest thing and is probably relatively fair. Love/CP3 injuries were known going in. Korver missing two games and Kyrie playing only 49 minutes were definitely not known, but whatever.

What seems more notable is how the series were won:

- The Cavaliers won 4-0 with a +13.3 MOV against a +4.75 SRS team while having a 63% market implied pre-series win probability

- The Rockets won 4-3 with a -3.1 MOV against a +6.80 SRS team while having a 58% market implied pre-series win probability

I personally wouldn't say that the Rockets had *easily* the more impressive series win.

And if we got into the individual component, worth mentioning that in Game 6, James Harden got benched with the Rockets down 17 in with 1:33 to go in the third quarter while down 3-2 in the series. Houston proceeded to finish the game on a 47-18 run without Harden to force a Game 7. He had a -21 +/- in the game and the second lowest +/- on the team was -1. We wouldn't be talking about a 'series win' at all without the efforts of Josh Smith and Corey Brewer while Harden had a towel on his head on the bench and likely presumed that his season was over, fwiw.

All these are fair points, and I don’t want to go down a rabbit hole about comparing those two opponents, since that was a pretty minor part of my post. I will just say that I do not think teams like the 2015 Hawks are to be taken very seriously in the playoffs. I really think playoff paper tigers are a thing, that they’re not all that hard to identify, and that that team was one of them. It’s a qualitative judgment that someone could disagree with, but I feel pretty confident in my assessment on this. Leaving aside playoff-paper-tiger stuff (which is all very debatable and rabbit-holey), the Clippers did also have a 6.80 SRS, which is also just usually quite a lot scarier than a 4.75 SRS team (particularly if the 4.75 SRS team doesn’t have a major star and the 6.80 SRS team does—but that gets into indicators of being a paper tiger). The Clippers had also just gotten a series win against a great Spurs team, while the Hawks had barely put up a positive SRS against very mediocre opponents in the first two rounds. I feel pretty confident that the Clippers were a lot better.

In any event, though, I think one could take a more charitable view towards the 2015 Hawks than I do (or a more pessimistic view of the 2015 Clippers) and still not find it particularly difficult to conclude that Harden was better than LeBron in this particular year. There’s a lot of info in my post that goes to that point. Again, I still would put LeBron ahead because I just put independent value on the achievement of making the Finals, but the argument for Harden isn’t very hard to make IMO.

EDIT: Had somehow forgotten about CP3 missing games. While the Clippers still won a game without him (so the Rockets went 3-2 in the games CP3 did play), I’d say that that does mute this particular point a fair bit. Still think the Clippers were better than the Hawks, but CP3 having an injury reduces the gap such that I don’t really think this point about the Clippers being a great team is all that significant in this comparison.

[Elpolo_14]

Overall, this is not entirely conclusive, since there’s data going both ways, but the data picture does go in Harden’s direction overall IMO. And when we account for the fact that Harden played 20% more minutes than LeBron during the regular season, it’s pretty easy to conclude that Harden was better in the regular season. Which, FWIW, was something that the MVP voters agreed with—putting Harden in 2nd, and well above LeBron in 3rd. This conclusion is, of course, bolstered by the fact that Harden lifted the Rockets to 56 wins, despite Dwight missing half the season.

I'm sorry but I don't think you read the data in best way.

WOWY RAPM and on/off say Lebron better so I think it must be said the numbers like Lebron for impact.

There are also problem maybe in one RAPM inside https://www.apbr.org/metrics/viewtopic.php?t=9491. Mamba maker says coding is wrong.

[Elpolo_14]

And then we think about who they actually played, and probably more importantly, who they actually beat. The Rockets had *easily* the most impressive series win of these two teams—beating the 6.80 SRS Chris Paul Clippers. Yes, the Cavaliers easily beat the Hawks while not being full strength, but those Clippers were definitely a much better team IMO. That was a really high-quality series win, against a genuinely great team (who had just beaten the defending-champion Spurs)—which is not something the Cavaliers have on their resume (though, to be fair, we don’t know if they were capable of it or not—the East was just weak).

Hmph

Would start by throwing in the pre-series implied probabilities to get a sense of how the market viewed the respective series:

- CLE 63.45% - 36.55% ATL

- HOU 57.98% - 42.02% LAC

Okay, so Cleveland was the bigger favorite.

Injuries complicate this both ways but I think ignoring them is the simplest thing and is probably relatively fair. Love/CP3 injuries were known going in. Korver missing two games and Kyrie playing only 49 minutes were definitely not known, but whatever.

What seems more notable is how the series were won:

- The Cavaliers won 4-0 with a +13.3 MOV against a +4.75 SRS team while having a 63% market implied pre-series win probability

- The Rockets won 4-3 with a -3.1 MOV against a +6.80 SRS team while having a 58% market implied pre-series win probability

I personally wouldn't say that the Rockets had *easily* the more impressive series win.

And if we got into the individual component, worth mentioning that in Game 6, James Harden got benched with the Rockets down 17 in with 1:33 to go in the third quarter while down 3-2 in the series. Houston proceeded to finish the game on a 47-18 run without Harden to force a Game 7. He had a -21 +/- in the game and the second lowest +/- on the team was -1. We wouldn't be talking about a 'series win' at all without the efforts of Josh Smith and Corey Brewer while Harden had a towel on his head on the bench and likely presumed that his season was over, fwiw.

Like you mentioned a bit
The big factor that make the Houston rockets win that series would definitely be because Chris Paul enter the series with injury from previous round ( Since game 7 at the first Quarter against the Spurs ) and Chris-Paul still miss a significant amount of game in this series ( 2 game ) even when he came back his impact got hinder by being less on court ( limit minutes/ minutes restrictions for the 2-3 game back ) so the team who just loses ( at least the not best version)their best assessment on CP who is one of the best offensive engine with his mid volume good efficiency scorer with elite traits as a floor general/playmaker that Years and an elite POA defender to contain the Houston offense would be a Insane feats to win that series in this condition. ( The regular season SRS isn't a good indicator when the team have major changes due to injury and rotation )