2017-18 RAPM/RPM/etc. Thread

[CommonerCoffee]

Something sorta interesting and strange.
Detroit holds the 6th DRTG in the league but their guys don't seem to be well liked by DRAPM much.

[dhsilv2]

Something sorta interesting and strange.
Detroit holds the 6th DRTG in the league but their guys don't seem to be well liked by DRAPM much.

It doesn't seem to dislike any of them either.

[CommonerCoffee]

Is RAPM down? =(

[Clyde Frazier]

Read on Twitter

[Clyde Frazier]

Read on Twitter

[Dr Spaceman]

Read on Twitter

I get what he’s saying here but I actually don’t see this as a positive.

This is actually Toronto’s best lineup by a mile and it’s one that has no chance of ever seeing the light of day in a playoff series. The individual players have been awesome of course,which I think was his point... but the core Raptors lineups haven’t been actually more effective than they were in previous years.

If lineups like this one actually go a long way to explaining the Raptors huge leap forward this year, which it seems like they do, I don’t see that as a positive sign.

[dontcalltimeout]

Read on Twitter

I get what he’s saying here but I actually don’t see this as a positive.

This is actually Toronto’s best lineup by a mile and it’s one that has no chance of ever seeing the light of day in a playoff series. The individual players have been awesome of course,which I think was his point... but the core Raptors lineups haven’t been actually more effective than they were in previous years.

If lineups like this one actually go a long way to explaining the Raptors huge leap forward this year, which it seems like they do, I don’t see that as a positive sign.

Might not be quite true that it's the main reason for their improvement. Their most-used starting lineup & most-used lineup overall is performing pretty dang well:

Read on Twitter

Their most used lineups in the RS last year was -10.4 (for 364 minutes) and +2.3 after the Ibaka trade (for 232 minutes). Compared to this year's most used lineup which is +12.4 (655 minutes). There's some reason for optimism.

[Jim Naismith]

RPM updated recently: http://www.espn.com/nba/statistics/rpm/_/sort/RPM

Harden now #1 in both RPM and WINS.

[Clyde Frazier]

Anyone familiar with "PIPM"?

Read on Twitter

Nylon Calculus: Introducing Player Impact Plus-Minus

Without further ado, I present Player Impact Plus-Minus. There are three components of Player Impact Plus-Minus: a box-score prior, luck-adjusted on-off data, and luck-adjusted net rating. Together these three components are able to provide a descriptive function of what has happened as well as add insight into future results.

https://fansided.com/2018/01/11/nylon-calculus-introducing-player-impact-plus-minus/

[eminence]

Anyone familiar with "PIPM"?

Read on Twitter

Nylon Calculus: Introducing Player Impact Plus-Minus

Without further ado, I present Player Impact Plus-Minus. There are three components of Player Impact Plus-Minus: a box-score prior, luck-adjusted on-off data, and luck-adjusted net rating. Together these three components are able to provide a descriptive function of what has happened as well as add insight into future results.

https://fansided.com/2018/01/11/nylon-calculus-introducing-player-impact-plus-minus/

I've seen Jacob using it before, and it certainly looks really good, but overall I worry it's a bit overtuned.

[SideshowBob]

For archival purposes RPM Updated for the full 2017 - 2018 RS on April 13, 2018. This is Single-Year.

Anyone spreadsheet this?

[thekdog34]

Anyone familiar with "PIPM"?

Read on Twitter

Nylon Calculus: Introducing Player Impact Plus-Minus

Without further ado, I present Player Impact Plus-Minus. There are three components of Player Impact Plus-Minus: a box-score prior, luck-adjusted on-off data, and luck-adjusted net rating. Together these three components are able to provide a descriptive function of what has happened as well as add insight into future results.

https://fansided.com/2018/01/11/nylon-calculus-introducing-player-impact-plus-minus/

My understanding is it's like RPM with a box-score prior but it improves it with player-tracking plus-minus.

Here's a little article on it:
https://fansided.com/2018/01/11/nylon-calculus-introducing-player-impact-plus-minus/

[thekdog34]

Vanilla RAPM is also working:
https://docs.google.com/spreadsheets/d/e/2PACX-1vSzp3G5rwP9xgCgluVGmR3Qj4-BMoGSYiuTKM6o_pzES6s95oQE1nQvB2CXed-4fRc_MMGgpULtDaJ_/pubhtml?gid=1825430955&single=true

Top 10 are Covington, Curry, Eric Gordon, Embiid, CP3, Lillard, Porter, Oladipo, Holiday.

Harden is 17, Westbrook 19, Durant 78 and LeBron 36.

What's going on with Durant? He's dropping off in all these.

[dontcalltimeout]

What's going on with Durant? He's dropping off in all these.

I'm guessing it's related to this:

Read on Twitter

Both RPM and RAPM are single year (so his previous yr performances aren't helping him) and the team hasn't been good with him and without Curry. Plus minus doesn't care whether the team is "coasting" or not, if they play poorly on offense, he's going to look bad.

[SideshowBob]

What's going on with Durant? He's dropping off in all these.

I'm guessing it's related to this:

Read on Twitter

Both RPM and RAPM are single year (so his previous yr performances aren't helping him) and the team hasn't been good with him and without Curry. Plus minus doesn't care whether the team is "coasting" or not, if they play poorly on offense, he's going to look bad.

But Multi-Year or Prior-Year informed RAPM can (to an extent) proxy-account for this by having a concrete value to revert to, as opposed to 0, which is part of why those sets perform better ITO predictive value.

[dontcalltimeout]

What's going on with Durant? He's dropping off in all these.

I'm guessing it's related to this:

Read on Twitter

Both RPM and RAPM are single year (so his previous yr performances aren't helping him) and the team hasn't been good with him and without Curry. Plus minus doesn't care whether the team is "coasting" or not, if they play poorly on offense, he's going to look bad.

But Multi-Year or Prior-Year informed RAPM can (to an extent) proxy-account for this by having a concrete value to revert to, as opposed to 0, which is part of why those sets perform better ITO predictive value.

For sure. I do think it's interesting that Durant is consistently lower in RAPM (including multiyear) than box score stats would imply. It's been a consistent thing throughout his career from a variety of sources of RAPM.

Another guy that we see that with to an even larger extent is Anthony Davis. Last year he was like 52 in vanilla RAPM and this year 31. That's good, without a doubt, but not a level above guys like Towns, Jokic, Horford.

We can all draw our own conclusions on how much to weigh that, but it's an undeniable and consistent data point.

[RCM88x]

I'm guessing it's related to this:

Read on Twitter

Both RPM and RAPM are single year (so his previous yr performances aren't helping him) and the team hasn't been good with him and without Curry. Plus minus doesn't care whether the team is "coasting" or not, if they play poorly on offense, he's going to look bad.

But Multi-Year or Prior-Year informed RAPM can (to an extent) proxy-account for this by having a concrete value to revert to, as opposed to 0, which is part of why those sets perform better ITO predictive value.

For sure. I do think it's interesting that Durant is consistently lower in RAPM (including multiyear) than box score stats would imply. It's been a consistent thing throughout his career from a variety of sources of RAPM.

Another guy that we see that with to an even larger extent is Anthony Davis. Last year he was like 52 in vanilla RAPM and this year 31. That's good, without a doubt, but not a level above guys like Towns, Jokic, Horford.

We can all draw our own conclusions on how much to weigh that, but it's an undeniable and consistent data point.

Am I the only one who isn't really suprised by these two sort of underpreforming in these metrics?

You watch the play and personally I just don't see the impact that a Curry, LeBron, even a CP3, Gobert, Embiid etc... have. Especially outside of '15 for AD and after the foot injury for KD. I just don't think either guys have reached those levels since and probably never will.

I don't know if it really means anything, or if I'm just barking up the wrong tree, or perhaps it's just confirmation bias. But I just can't say I'm surprised by these numbers. The tall, athletic guy who can shoot always seems to be overrated in their actual impact and difference making on the court outside of putting up good box scores.

[thekdog34]

But Multi-Year or Prior-Year informed RAPM can (to an extent) proxy-account for this by having a concrete value to revert to, as opposed to 0, which is part of why those sets perform better ITO predictive value.

For sure. I do think it's interesting that Durant is consistently lower in RAPM (including multiyear) than box score stats would imply. It's been a consistent thing throughout his career from a variety of sources of RAPM.

Another guy that we see that with to an even larger extent is Anthony Davis. Last year he was like 52 in vanilla RAPM and this year 31. That's good, without a doubt, but not a level above guys like Towns, Jokic, Horford.

We can all draw our own conclusions on how much to weigh that, but it's an undeniable and consistent data point.

Am I the only one who isn't really suprised by these two sort of underpreforming in these metrics?

You watch the play and personally I just don't see the impact that a Curry, LeBron, even a CP3, Gobert, Embiid etc... have. Especially outside of '15 for AD and after the foot injury for KD. I just don't think either guys have reached those levels since and probably never will.

I don't know if it really means anything, or if I'm just barking up the wrong tree, or perhaps it's just confirmation bias. But I just can't say I'm surprised by these numbers. The tall, athletic guy who can shoot always seems to be overrated in their actual impact and difference making on the court outside of putting up good box scores.

I'm not sure what the profile is of these guys, but Dirk always had large impact, so not sure it's just about tall scorers who don't do much else.

I think it often comes down to some sort of "gravity" these guys have and whether they can exploit it. Durant should have much more gravity than he seems to actually have on the court.

[bondom34]

For sure. I do think it's interesting that Durant is consistently lower in RAPM (including multiyear) than box score stats would imply. It's been a consistent thing throughout his career from a variety of sources of RAPM.

Another guy that we see that with to an even larger extent is Anthony Davis. Last year he was like 52 in vanilla RAPM and this year 31. That's good, without a doubt, but not a level above guys like Towns, Jokic, Horford.

We can all draw our own conclusions on how much to weigh that, but it's an undeniable and consistent data point.

Am I the only one who isn't really suprised by these two sort of underpreforming in these metrics?

You watch the play and personally I just don't see the impact that a Curry, LeBron, even a CP3, Gobert, Embiid etc... have. Especially outside of '15 for AD and after the foot injury for KD. I just don't think either guys have reached those levels since and probably never will.

I don't know if it really means anything, or if I'm just barking up the wrong tree, or perhaps it's just confirmation bias. But I just can't say I'm surprised by these numbers. The tall, athletic guy who can shoot always seems to be overrated in their actual impact and difference making on the court outside of putting up good box scores.

I'm not sure what the profile is of these guys, but Dirk always had large impact, so not sure it's just about tall scorers who don't do much else.

I think it often comes down to some sort of "gravity" these guys have and whether they can exploit it. Durant should have much more gravity than he seems to actually have on the court.

I think its physical profile too. Looking at him Durant looks like he has all the athletic ability and tools to be a great defender and can score with the best ever. Since his injury his handles have been shaky and he doesn't consistently defend that way though. He's a great player but at this point I don't even see him as a top 5 guy this year. Maybe top 10 at best.

[SideshowBob]

For sure. I do think it's interesting that Durant is consistently lower in RAPM (including multiyear) than box score stats would imply. It's been a consistent thing throughout his career from a variety of sources of RAPM.

Another guy that we see that with to an even larger extent is Anthony Davis. Last year he was like 52 in vanilla RAPM and this year 31. That's good, without a doubt, but not a level above guys like Towns, Jokic, Horford.

We can all draw our own conclusions on how much to weigh that, but it's an undeniable and consistent data point.

Am I the only one who isn't really suprised by these two sort of underpreforming in these metrics?

You watch the play and personally I just don't see the impact that a Curry, LeBron, even a CP3, Gobert, Embiid etc... have. Especially outside of '15 for AD and after the foot injury for KD. I just don't think either guys have reached those levels since and probably never will.

I don't know if it really means anything, or if I'm just barking up the wrong tree, or perhaps it's just confirmation bias. But I just can't say I'm surprised by these numbers. The tall, athletic guy who can shoot always seems to be overrated in their actual impact and difference making on the court outside of putting up good box scores.

I'm not sure what the profile is of these guys, but Dirk always had large impact, so not sure it's just about tall scorers who don't do much else.

I think it often comes down to some sort of "gravity" these guys have and whether they can exploit it. Durant should have much more gravity than he seems to actually have on the court.

Back and forth on this from the 2017 POTY thread. dontcalltimeout had some great Synergy data on the following pages.