2018-2019 OKC Thunder Regular Season Pt 1: A Renewed Hope

[getrichordie]

Wanted to express some concern over the Adams hype. His per 100 numbers are super close to last years #s. Not much different except less fouling.


Sent from my iPhone using RealGM Forums

[Dadouv47]

Wanted to express some concern over the Adams hype. His per 100 numbers are super close to last years #s. Not much different except less fouling.


Sent from my iPhone using RealGM Forums

Whats your concern? He was great last year too.

[ThunderBolt]

Wanted to express some concern over the Adams hype. His per 100 numbers are super close to last years #s. Not much different except less fouling.


Sent from my iPhone using RealGM Forums

You need to play fantasy sports. It will help control these self destructive tendencies that you have towards the thunder roster.

[RalphSampsonJr]

Wanted to express some concern over the Adams hype. His per 100 numbers are super close to last years #s. Not much different except less fouling.


Sent from my iPhone using RealGM Forums

Ok. Explain in detail these concerns? The per 100 numbers are similar to last year so.. ?

[getrichordie]

Wanted to express some concern over the Adams hype. His per 100 numbers are super close to last years #s. Not much different except less fouling.


Sent from my iPhone using RealGM Forums

Ok. Explain in detail these concerns? The per 100 numbers are similar to last year so.. ?

A lot are saying he’s better this year and I’m not so sure that’s the case. If so, it’s slightly better.

His USG% is up, but shooting numbers are down.
Rebound #s are marginally better and that could be attributed to the easy schedule. So I’m not really sure what he’s doing better this year than last. Grant has been a big help for Adams on defense. Knowing you have Grant (and not Melo) helping you if you get beat is huge.

He’s 11th in RPM (1.42) even though “he’s played like the best center in the league.”

Out of the centers who play 30 minutes, Adams ranks 6th out of 10th. According to RPM, Adams is closer to Adebayo, Allen, Valanciunas level.

1. Jokic (30.9mp)
2. Vucecic (30.8mp)
3. M. Garsol (34.5 mp)
4. Gobert (30.9)
5. Embiid (33.8)
6. Adams (33.4)
7. KAT (33.6)
8. Horford (30.0)
9. Jordan (31.3)
10. Drummond (33.3)

It seems he has largely been doing what he has last year. He just has more help this year.




Sent from my iPhone using RealGM mobile app

[RalphSampsonJr]

Wanted to express some concern over the Adams hype. His per 100 numbers are super close to last years #s. Not much different except less fouling.


Sent from my iPhone using RealGM Forums

Ok. Explain in detail these concerns? The per 100 numbers are similar to last year so.. ?

A lot are saying he’s better this year and I’m not so sure that’s the case. If so, it’s slightly better.

His USG% is up, but shooting numbers are down.
Rebound #s are marginally better and that could be attributed to the easy schedule. So I’m not really sure what he’s doing better this year than last. Grant has been a big help for Adams on defense. Knowing you have Grant (and not Melo) helping you if you get beat is huge.

He’s 11th in RPM (1.42) even though “he’s played like the best center in the league.”

Out of the centers who play 30 minutes, Adams ranks 6th out of 10th. According to RPM, Adams is closer to Adebayo, Allen, Valanciunas level.

1. Jokic (30.9mp)
2. Vucecic (30.8mp)
3. M. Garsol (34.5 mp)
4. Gobert (30.9)
5. Embiid (33.8)
6. Adams (33.4)
7. KAT (33.6)
8. Horford (30.0)
9. Jordan (31.3)
10. Drummond (33.3)

It seems he has largely been doing what he has last year. He just has more help this year.




Sent from my iPhone using RealGM mobile app

Shooting numbers as in fg%? Of course you will struggle to shoot 60% when you are now the clear 3rd option and have a higher useage. Hes being double teamed in the post these days

His points rebs and assists are the highest they have ever been.. still collecting more oboards than dboards while helping the team be, once agiant an elite rebounding team.

His net rating is better than pg russ and even curry..

I dont get your concern with him. Are you just trying to find numbers that help your idea of trading him for a big name?

[spearsy23]

Wanted to express some concern over the Adams hype. His per 100 numbers are super close to last years #s. Not much different except less fouling.


Sent from my iPhone using RealGM Forums

This is factually incorrect.

[getrichordie]

Wanted to express some concern over the Adams hype. His per 100 numbers are super close to last years #s. Not much different except less fouling.


Sent from my iPhone using RealGM Forums

This is factually incorrect.

How so?


Sent from my iPhone using RealGM mobile app

[Dn4sty]

Wanted to express some concern over the Adams hype. His per 100 numbers are super close to last years #s. Not much different except less fouling.


Sent from my iPhone using RealGM Forums

You need to play fantasy sports. It will help control these self destructive tendencies that you have towards the thunder roster.

I was about to ask if he had the newest NBA 2K game? He could GM himself into oblivion.

[bondom34]

Ok. Explain in detail these concerns? The per 100 numbers are similar to last year so.. ?

A lot are saying he’s better this year and I’m not so sure that’s the case. If so, it’s slightly better.

His USG% is up, but shooting numbers are down.
Rebound #s are marginally better and that could be attributed to the easy schedule. So I’m not really sure what he’s doing better this year than last. Grant has been a big help for Adams on defense. Knowing you have Grant (and not Melo) helping you if you get beat is huge.

He’s 11th in RPM (1.42) even though “he’s played like the best center in the league.”

Out of the centers who play 30 minutes, Adams ranks 6th out of 10th. According to RPM, Adams is closer to Adebayo, Allen, Valanciunas level.

1. Jokic (30.9mp)
2. Vucecic (30.8mp)
3. M. Garsol (34.5 mp)
4. Gobert (30.9)
5. Embiid (33.8)
6. Adams (33.4)
7. KAT (33.6)
8. Horford (30.0)
9. Jordan (31.3)
10. Drummond (33.3)

It seems he has largely been doing what he has last year. He just has more help this year.




Sent from my iPhone using RealGM mobile app

Shooting numbers as in fg%? Of course you will struggle to shoot 60% when you are now the clear 3rd option and have a higher useage. Hes being double teamed in the post these days

His points rebs and assists are the highest they have ever been.. still collecting more oboards than dboards while helping the team be, once agiant an elite rebounding team.

His net rating is better than pg russ and even curry..

I dont get your concern with him. Are you just trying to find numbers that help your idea of trading him for a big name?

Instead, just take out the box score part. Try RAPM.

https://docs.google.com/spreadsheets/d/e/2PACX-1vQdG8Zv84zqKEzETDjd8KPsClcw9bPETX9v_x_KEAxjv9NrFaWikOoiSaciy1jbMiygg2D-V8DUQn0O/pubhtml?gid=0&single=true

Adams has performed at a level much higher than most of those names.

[getrichordie]

Is there any reason to think RAPM is any better than RPM? What’s the difference? I don’t trust any stat that has Danny Green at the top.

What does RAPM tell me that RPM doesn’t?

EDIT:

Conventional adjusted plus-minus does a poor job of predicting the outcome of future games, particularly when you have less than a season of data.

So this is a prediction stat?

Sent from my iPhone using RealGM Forums

[bondom34]

Edit: NVM, not worth it

[slick_watts]

Is there any reason to think RAPM is any better than RPM? What’s the difference? I don’t trust any stat that has Danny Green at the top.

What does RAPM tell me that RPM doesn’t?

EDIT:

Conventional adjusted plus-minus does a poor job of predicting the outcome of future games, particularly when you have less than a season of data.

So this is a prediction stat?

Sent from my iPhone using RealGM Forums

RAPM tells you 'less' than what RPM tells you because while RAPM uses priors to do ridge regression in order to control outliers, the priors are set to zero for everyone. RPM is a variant of xRAPM, which uses boxscore data among other things to set different priors for different players-- which makes sense intuitively, but some say it also introduces its own noise and biases.

my experience is that when people don't like what RPM says they will go to RAPM and see if they find it there and then go to non-prior informed RAPM to find it and so on.

your quote is about plain jane APM which is adjusted plus-minus, basically the starting point of all these metrics. on its own it is incredibly noisy which is why many of these other techniques were devised to account for it.

[RalphSampsonJr]

Why do we need to dive into all this RPM RAPM stuff?
Are we watching games? What Adams does is literally how the team is built to operate.

Terrible shooting team - need guys to rebound and get extra possesions. Check.

Have two superstars that need to have space created for them off solid screens. Check.

Great defense who get hands in passing lanes, deflect balls which generates downhill, transition offense. Check.

Adams is elite at kost things the team needs him to do. His ability to screen and stop the defending players from staying on Russ and PG is second to none.
His post game has improved to the point where he now has to be scouted for.

Try find any piece of data you want to make a case but from the games im watching, he is doing everything that has been asked of him at a very high level.

[slick_watts]

another thing is that any metric that is using priors isn't really designed to describe how a player has played in the current season. it's predictive and not descriptive.

[getrichordie]

another thing is that any metric that is using priors isn't really designed to describe how a player has played in the current season. it's predictive and not descriptive.

So, if I understand correctly, RPM is descriptive and RAPM is predictive. Is that correct?


Sent from my iPhone using RealGM mobile app

[bondom34]

Why do we need to dive into all this RPM RAPM stuff?
Are we watching games? What Adams does is literally how the team is built to operate.

Terrible shooting team - need guys to rebound and get extra possesions. Check.

Have two superstars that need to have space created for them off solid screens. Check.

Great defense who get hands in passing lanes, deflect balls which generates downhill, transition offense. Check.

Adams is elite at kost things the team needs him to do. His ability to screen and stop the defending players from staying on Russ and PG is second to none.
His post game has improved to the point where he now has to be scouted for.

Try find any piece of data you want to make a case but from the games im watching, he is doing everything that has been asked of him at a very high level.

When you can get an older, worse player who doesn't play with the pace you want, you gotta do it.

[slick_watts]

another thing is that any metric that is using priors isn't really designed to describe how a player has played in the current season. it's predictive and not descriptive.

So, if I understand correctly, RPM is descriptive and RAPM is predictive. Is that correct?

Sent from my iPhone using RealGM mobile app

they are both designed to be predictive. consider rpm a more finely tuned version of rapm. it probably eliminates more noise but in during so may develop 'preferences'.

the way rpm in particular is presented lends itself to this kind of 'ranking' during the season, deciding who is playing better than who, etc. when it's not really designed for that.

comparing steven adams' rpm apples for apples with bam or valanciunas is missing the point of rpm.

[thekaoswithin]

another thing is that any metric that is using priors isn't really designed to describe how a player has played in the current season. it's predictive and not descriptive.

So, if I understand correctly, RPM is descriptive and RAPM is predictive. Is that correct?

Sent from my iPhone using RealGM mobile app

they are both designed to be predictive. consider rpm a more finely tuned version of rapm. it probably eliminates more noise but in during so may develop 'preferences'.

the way rpm in particular is presented lends itself to this kind of 'ranking' during the season, deciding who is playing better than who, etc. when it's not really designed for that.

comparing steven adams' rpm apples for apples with bam or valanciunas is missing the point of rpm.

Slick, you seen pretty educated on how all these advanced stats work from the stats side of things. As someone very interested in that, is there somewhere I could learn how they work?

[bondom34]

So, if I understand correctly, RPM is descriptive and RAPM is predictive. Is that correct?

Sent from my iPhone using RealGM mobile app

they are both designed to be predictive. consider rpm a more finely tuned version of rapm. it probably eliminates more noise but in during so may develop 'preferences'.

the way rpm in particular is presented lends itself to this kind of 'ranking' during the season, deciding who is playing better than who, etc. when it's not really designed for that.

comparing steven adams' rpm apples for apples with bam or valanciunas is missing the point of rpm.

Slick, you seen pretty educated on how all these advanced stats work from the stats side of things. As someone very interested in that, is there somewhere I could learn how they work?

http://godismyjudgeok.com/DStats/2011/nba-stats/a-review-of-adjusted-plusminus-and-stabilization/

https://cornerthreehoops.wordpress.com/2014/04/17/explaining-espns-real-plus-minus/

He probably has more but those are 2.