2015 APM/RAPM/etc.

[Doctor MJ]

Melo is a surprise. Knicks fans were saying all year that Melo is not the problem. I thought that he couldn't be having that much impact on a 5 win team, but RAPM does confirm that he is having impact.

What's fascinating about Melo is that the course of his career means that the narrative of him at any given time can't help but be messed up.

What I see: In New York Melo has pretty consistently impacted the game like a clear cut star in a way he did not in Denver. This probably says some about his maturity, but also says something specific about coach Karl in Denver having the ability to work wonders with a certain type of offense that was not what ran while Melo was at the focus.

This NY impact from Melo isn't enough that he should have been seen a true MVP candidate, but it's still plenty large, and yet nowhere near enough to make up for being on a horrendous team. Which is of course ironic, because so much of Melo's Denver reputation is based on the idea that he could guarantee you a pretty good team.

So his time in NY has had me revise my estimation of him upward, but for the mainstream, I get the sense that they really have no idea what to make of it and tend to basically just credit Melo when good things happen and make fun of his brand-building desires when bad things happen without really getting into what it means about his play other than to occasionally say he should be focusing on basketball more.

[Doctor MJ]

RAPM is supposed to be a better gauge then BPM or VORP or PER?
Cody Zellar huh. That's surprising. Morris/Kris/Miles. What's being done to get these numbers?

RAPM is a higher validity, lower reliability metric than the others.
That means that the other stats have outright biases that will ensure they will literally never get things right, but that the noise in RAPM can result in it getting things even more wrong.

As someone who uses RAPM, and similar stats, a lot, what you have to understand is that folks like me don't just swallow the numbers whole without looking at other context.

Example: If you see a guy with a huge RAPM while playing low minutes, that's a reason to think perhaps certain lineups are a great fit, and that perhaps that player should get more time, but it's not a reason to come out and say that player is actually a star-quality player. Being a star is about being able to have huge impact when the defense has to build its game around stopping you.

[Doctor MJ]

Anyone taken a look at Got Buckets' FourFactor APM? Its multiyear, they have data for the 2015 season (that is currently using data from Feb 2013-Feb 2015), but only in offensive/defensive splits. It'd be good to compare to previous seasons (they go back to 2005).

Offense

Defense

Fantastic. So glad to see someone else doing more 4 factor analysis with regression. First thing I always check is Defensive Rebounding.

And yes, once again, Nene destroys at impact defensive rebounding, while Dwight Howard is conspicuously more of an afterthought.

[Dr Spaceman]

Anyone taken a look at Got Buckets' FourFactor APM? Its multiyear, they have data for the 2015 season (that is currently using data from Feb 2013-Feb 2015), but only in offensive/defensive splits. It'd be good to compare to previous seasons (they go back to 2005).

Offense

Defense

Here is how anyone that got an MVP top 10 vote last year is doing according to this stat and I listed it from largest to smallest

Chris Paul = 9.07/3.07 = 12.14
LeBron James = 9.07/2.14 = 11.21
Stephen Curry = 12.93/-2.88 = 10.05
James Harden = 11.64/-2.29 = 9.35
Kevin Durant = 5.80/0.56 = 6.36
Lamarcus Aldridge = 2.64/1.95 = 4.59
Joakim Noah = -4.48/5.68 = 1.20
Blake Griffin = 0.10/-0.64 = -0.54
Kevin Love = -0.45/-0.78 = -1.23
Al Jefferson = -4.25/-1.61 = -5.86

Also is their a bias when it comes to big men on offense because both gasols, demarcus cousins are all in the negative, blake is barely above zero and LaMarcus is the first true big man ( PF, PF/C or C to be listed and he's only at 2.64, then Davis who is a great offensive player on a good offensive team is only at 2.08??? )

Frankly I'd just ignore the big number they've amalgamated there, it doesn't seem to say how it's weighted and the defensive version is particularly poor.

Here's what's important about this stat: it's a regression on Dean Oliver's four factors, so quite literally it's attempt to gauge "rebounding impact", "eFG impact" "TOV impact" and "FTR impact". In the same way RAPM is meant to quantify "scoreboard impact".

So basically look here if you want to see "who increases their team's rebounding rate the most when entring the game?"

[RSCD3_]

Anyone taken a look at Got Buckets' FourFactor APM? Its multiyear, they have data for the 2015 season (that is currently using data from Feb 2013-Feb 2015), but only in offensive/defensive splits. It'd be good to compare to previous seasons (they go back to 2005).

Offense

Defense

Here is how anyone that got an MVP top 10 vote last year is doing according to this stat and I listed it from largest to smallest

Chris Paul = 9.07/3.07 = 12.14
LeBron James = 9.07/2.14 = 11.21
Stephen Curry = 12.93/-2.88 = 10.05
James Harden = 11.64/-2.29 = 9.35
Kevin Durant = 5.80/0.56 = 6.36
Lamarcus Aldridge = 2.64/1.95 = 4.59
Joakim Noah = -4.48/5.68 = 1.20
Blake Griffin = 0.10/-0.64 = -0.54
Kevin Love = -0.45/-0.78 = -1.23
Al Jefferson = -4.25/-1.61 = -5.86

Also is their a bias when it comes to big men on offense because both gasols, demarcus cousins are all in the negative, blake is barely above zero and LaMarcus is the first true big man ( PF, PF/C or C to be listed and he's only at 2.64, then Davis who is a great offensive player on a good offensive team is only at 2.08??? )

Frankly I'd just ignore the big number they've amalgamated there, it doesn't seem to say how it's weighted and the defensive version is particularly poor.

Here's what's important about this stat: it's a regression on Dean Oliver's four factors, so quite literally it's attempt to gauge "rebounding impact", "eFG impact" "TOV impact" and "FTR impact". In the same way RAPM is meant to quantify "scoreboard impact".

So basically look here if you want to see "who increases their team's rebounding rate the most when entring the game?"

Oh ok so the individual factors are the important part and the combination is just some kind of PER like weighting system of each factor?

[Dr Spaceman]

Here is how anyone that got an MVP top 10 vote last year is doing according to this stat and I listed it from largest to smallest

Chris Paul = 9.07/3.07 = 12.14
LeBron James = 9.07/2.14 = 11.21
Stephen Curry = 12.93/-2.88 = 10.05
James Harden = 11.64/-2.29 = 9.35
Kevin Durant = 5.80/0.56 = 6.36
Lamarcus Aldridge = 2.64/1.95 = 4.59
Joakim Noah = -4.48/5.68 = 1.20
Blake Griffin = 0.10/-0.64 = -0.54
Kevin Love = -0.45/-0.78 = -1.23
Al Jefferson = -4.25/-1.61 = -5.86

Also is their a bias when it comes to big men on offense because both gasols, demarcus cousins are all in the negative, blake is barely above zero and LaMarcus is the first true big man ( PF, PF/C or C to be listed and he's only at 2.64, then Davis who is a great offensive player on a good offensive team is only at 2.08??? )

Frankly I'd just ignore the big number they've amalgamated there, it doesn't seem to say how it's weighted and the defensive version is particularly poor.

Here's what's important about this stat: it's a regression on Dean Oliver's four factors, so quite literally it's attempt to gauge "rebounding impact", "eFG impact" "TOV impact" and "FTR impact". In the same way RAPM is meant to quantify "scoreboard impact".

So basically look here if you want to see "who increases their team's rebounding rate the most when entring the game?"

Oh ok so the individual factors are the important part and the combination is just some kind of PER like weighting system of each factor?

Correct.

[SideshowBob]

Oh ok so the individual factors are the important part and the combination is just some kind of PER like weighting system of each factor?

Yes, but those weights are derived via regression on the Four Factors. On that note, BPM/ASPM and SPM also fit into that mold; BPM/ASPM is based on regression of box-score metrics on RAPM (currently J.E.'s 14 year set) and SPM is typically associated with team MOV.

[GYK]

RAPM is supposed to be a better gauge then BPM or VORP or PER?
Cody Zellar huh. That's surprising. Morris/Kris/Miles. What's being done to get these numbers?

RAPM is a higher validity, lower reliability metric than the others.
That means that the other stats have outright biases that will ensure they will literally never get things right, but that the noise in RAPM can result in it getting things even more wrong.

As someone who uses RAPM, and similar stats, a lot, what you have to understand is that folks like me don't just swallow the numbers whole without looking at other context.

Example: If you see a guy with a huge RAPM while playing low minutes, that's a reason to think perhaps certain lineups are a great fit, and that perhaps that player should get more time, but it's not a reason to come out and say that player is actually a star-quality player. Being a star is about being able to have huge impact when the defense has to build its game around stopping you.

Thanks Doc. I just watch a lot of the SCAC videos. I'm honestly fascinated by the idea of analytics but it's just not there yet. Im more interested in what's effective. Which direction should a corner pin down screen setter be standing, because it's leading to a moving three that has % of which direction offensive rebounds. I see a huge game planning potential. Having good players is great chance to win. But other probabilities arise as well. The potential is absolutely game changing.

[fuzzy_dunlop]

Bumping this to humbly request some updated figures. Anyone mind badgering JE? He's given us a taste, but it's been like 30 games...

[SideshowBob]

Bumping this to humbly request some updated figures. Anyone mind badgering JE? He's given us a taste, but it's been like 30 games...

I requested it, hopefully we get something.

[colts18]

Jeremias Engelmann @JerryEngelmann · Mar 9

Leaders in single year RAPM: Curry, LBJ (gaining steam), DrGreen, Korver, Harden, Middleton, Thompson, Leonard, CP3, Mozgov, M.Barnes, ZachR

[SideshowBob]

Jeremias Engelmann @JerryEngelmann · Mar 9

Leaders in single year RAPM: Curry, LBJ (gaining steam), DrGreen, Korver, Harden, Middleton, Thompson, Leonard, CP3, Mozgov, M.Barnes, ZachR

I'm guessing this is NPI, as opposed to this:

http://stats-for-the-nba.appspot.com/ratings/RAPM_2015.html

[ceiling raiser]

Jeremias Engelmann @JerryEngelmann · Mar 9

Leaders in single year RAPM: Curry, LBJ (gaining steam), DrGreen, Korver, Harden, Middleton, Thompson, Leonard, CP3, Mozgov, M.Barnes, ZachR

No AD again.

[Dr Spaceman]

Jeremias Engelmann @JerryEngelmann · Mar 9

Leaders in single year RAPM: Curry, LBJ (gaining steam), DrGreen, Korver, Harden, Middleton, Thompson, Leonard, CP3, Mozgov, M.Barnes, ZachR

No AD again.

Yeah any thoughts on why that is? IIRC he was a negative last year, which I just attributed to second years being second years. But from what I remember even this year Holiday looks like the team star. What is going on that these massive boxscore stats aren't translating to the scoreboard?

IIRC Stoudemire never looked great by RAPM either, who I'd probably his closest stylistic analogue offensively.

[SideshowBob]

GotBuckets 2014 PI set has him at break even (+0.7 Off/-0.7 Def), so perhaps that prior is holding him down a little.

Davis was at +1.8 Offense, +0.7 Defense in the last 2015 PI RAPM update (the one that's linked on this page and the first). I'm assuming that with a decent chunk of added games, he's probably broken free from that and is trending towards the +4 overall range (which isn't too far from where I would peg him, though still lower), so perhaps he's getting there.

[RSCD3_]

GotBuckets 2014 PI set has him at break even (+0.7 Off/-0.7 Def), so perhaps that prior is holding him down a little.

Davis was at +1.8 Offense, +0.7 Defense in the last 2015 PI RAPM update (the one that's linked on this page and the first). I'm assuming that with a decent chunk of added games, he's probably broken free from that and is trending towards the +4 overall range (which isn't too far from where I would peg him, though still lower), so perhaps he's getting there.

Positive on defense is good in this measure right?

[G35]

But Korver is NOT the #1 offensive player by traditional measures.

He's only the 4th leading scorer on his team.
His FGA/36 is 13th on his team.

We're literally talking about a guy who does one thing so well...that he barely does that thing. An offense based not on "You can't stop our best shooter" but rather based on "Our shooter is so good, you simply have to commit to stop him, and when you do you can't cover the rest of our guys well."

Reggie, though he seems unusual for an offensive star, was still your #1 scorer, and hence far more of a normal situation.

What I will say is that Atlanta with Korver can be seen as a descendant of Indiana with Miller in terms of a "How far can we push this?" type thing.

The one difference I see with Korver is his defense. Yes, he is nowhere near a #1 option but he is as you say the weapon teams have to account for. Typically anyone who shoots that well from 3pt land is usually not that good of a defender. Most great 3pt shooters are hidden on defense and offenses can attack them when they have the ball. I think Korver is an above average defender, plays well within the Hawks defense from what I have seen. He's really a unique player to find a comparison for...how many players can shoot 50% from the three at a high volume (5+ per game) and still be a positive on defense.

I looked at Mike Miller who was a better scorer but not as good 3P% and he would be a similar type during his Memphis days. The best single season 3P% you are talking about Tim Legler, Steve Kerr, Kapono, Detlef, Craig Hodges. Detlef Schrempf is a good comparison but he is a little taller, more PF and he didn't have nearly the volume of three's.

The top five 3P shooters are Curry (208), Thompson (190), Korver (189), Matthews (173), and Harden (161).

Only Thompson and Korver are in the top five of 3P% and it's not even close with Korver shooting .501 and Thompson at .434 and Currey is at .418.

What Korver does is a game changer. Now I don't think he can be a #1 option but on a team like the Hawks he makes them a top tier team. I wonder how Korver would look on the Warriors, or Cavaliers, or Clippers. I think those teams would become significantly better......

[ceiling raiser]

Yeah any thoughts on why that is? IIRC he was a negative last year, which I just attributed to second years being second years. But from what I remember even this year Holiday looks like the team star. What is going on that these massive boxscore stats aren't translating to the scoreboard?

IIRC Stoudemire never looked great by RAPM either, who I'd probably his closest stylistic analogue offensively.

I do think the box score numbers perhaps overrate his defense, but he was looking solid in a lot of the +/- stats a few weeks back. The team did pretty well while he was out I believed, maybe that hurt his numbers.

I'm wondering though (and maybe those who watch him more closely can answer this), are the box score stats overrating his offensive impact as well? +1.8 off in PI is very solid for a big, so I'm not sure if this is the case.

His defensive shot chart as of a couple days ago is pretty impressive:

[tweet]https://twitter.com/ShotAnalytics/status/575358639789228034[/tweet]

I'm not sure, really (only problem I see from the above, other than the corner 3s, is that people are getting to the rim a good deal, even if they're not scoring well there). Maybe there's a legitimate issue with his help defense? Alternatively, maybe there's some interference going on with teammates that's clouding up the numbers?

EDIT: It's also possible that J.E. is referring to PI RAPM (though I think he'd refer to that as multi-year), in which case I'd agree with most others that he's being buoyed by his prior.

[SideshowBob]

Positive on defense is good in this measure right?

Yup. Negative is only good when speaking relative to league average or relative to opponents' DRTG.

[drza]

Yeah any thoughts on why that is? IIRC he was a negative last year, which I just attributed to second years being second years. But from what I remember even this year Holiday looks like the team star. What is going on that these massive boxscore stats aren't translating to the scoreboard?

IIRC Stoudemire never looked great by RAPM either, who I'd probably his closest stylistic analogue offensively.

His raw +/- numbers from B-R have improved greatly since last season and are now strong, even if not "best in the league" kind of strong:

On +/-: +5.7
On/off net: +11.0

(Up from -1.8 and +1.9, respectively, last season). He's certainly putting up much better on and on/off +/- numbers than his primary teammates, and generally speaking the player with clearly the highest marks there do well in the RAPM studies. So I'll join the crowd and suggest that Davis is maybe being held back by his prior, and even if he wouldn't be topping the league in RAPM I think he'll eventually measure much better there than the early returns suggested

I think the Amare offensive comp is actually a pretty good one, on the whole. The difference is that he's producing at Amare-with-Nash level efficiency while playing with Holiday/Tyreke level floor generals which has to be more impressive. But stylistically and on the whole I'd say they definitely seem to be more in the same family offensively than say, Davis and Duncan.