|
![[image loading]](http://static.lakana.com/bmg-wctitv-media-us-east-1/photo/2017/11/20/11-20-17%20Tim%20Duncan%20and%20John%20Stockton%20_OP_ESPN%20photo_CP__1511229579718_9403562_ver1.0_640_360.jpg)
http://www.espn.com/mens-college-basketball/story/_/id/21493893/tim-duncan-john-stockton-headline-college-hoops-hall-fame-class
Tim Duncan, John Stockton highlight college Hall of Fame class of 2017
KANSAS CITY, Mo. -- John Stockton and Tim Duncan made their names in college, polished their legacies in the NBA, and retired with résumés that made them a lock for induction into just about any Hall of Fame.
Hard to believe there was a time when a precious few wanted them.
Stockton was an undersized point guard in the Pacific Northwest with competing offers from Idaho and Montana when he decided to continue his family's legacy at Gonzaga. Duncan dreamed of becoming a swimmer before a hurricane wiped out the only Olympic-size pool in the Virgin Islands, turning him toward hoops in the ninth grade -- too late for many schools to recruit him but not for Wake Forest.
Yet they made the most of their opportunities, parlaying their college success into careers that surpassed their wildest imaginations, and resulted in spots in the Collegiate Basketball Hall of Fame.
"Somebody has to give you a chance somewhere along the line," Stockton said. "Obviously I don't look the part, but someone found something they appreciated in me."
|
Ok, let's begin at the beginning.
The published "ESPN Real PLus Minus" is technically called known as RAPM or xRAPM.
That is "Regularized Adjusted Plus Minus".
It gives you the "Plus Minus" stat of a player per 100 possessions. and it is "Adjusted" to who is on the floor with the player. So all other 9 guys are accounted for in the calculation.
The "root" of this stat is the basic boxscore plus-minus you see in every game boxscore.
For example, on offense James Harden produces 8.57 more points per 100 possesions than the theoretical "average player".
http://www.espn.com/nba/statistics/rpm/_/sort/ORPM
|
What is plus minus anyway?
And what is the formula when it is adjusted and the others are accounted for?
Can you do an indepth explanation of a specific scenario of RAPM of a team, it's very complicated
|
it is brutally complicated. i and 1 other posted have provided the Nylon Calculus simplified guide to how "Adjusted Plus Minus" accounts for every player on the floor a few times. here you go!
https://fansided.com/2014/09/25/glossary-plus-minus-adjusted-plus-minus/
Basic Boxscore Plus Minus
While Lebron James was on the floor for 38 minutes last game the Cavaliers scored 25 more points than they gave up.
Therefore, LBJ's plus minus for that game is +25.
Regularized Adjusted Plus Minus
remember 2 equations and 2 unknowns from grade 9 math?
welp RAPM is 60,000 equations and 450 unknowns. the 450 unknowns are all NBA players.
the 60,000 equations are the results on NBA games split up into 20 second "stints".
this creates a giant MOFO of a Matrix. if you have 2nd year Linear Algebra including knowing stuff like how to orthogonally diagonalize a matrix.. eigenvectors.. eigenvalues.. etc.. you can probably do the math for it.
|
|
|
Yea, ive read those and some others too, but could you please provide a comprehensive eli5.
how does one get from regular stats like points, rebounds, etc to plus minus?
And how about Adjusted? What is being adjusted?
And whats the difference between RAPM and xRAPM?
And you said something about taking into account 10 players, how does that happen? What is the formula?
Could you also explain how that works in a team, like how maybe Derozan gets his stats and the raptors team as a whole
|
On November 22 2017 09:40 JimmiC wrote:Show nested quote +On November 22 2017 09:28 JimmyJRaynor wrote:On November 22 2017 09:20 JimmiC wrote:
Like slow ass kelly olynyk is a better defender then butler by over 20 slots. Or maybe its all effort???
ROLF ya, you are not using the stat properly. Centers are judged against other Centers. So, a Center with a Defensive RAPM of -0.6 is amongst the worst in the NBA while a Shooting Guard with a Defensive RAPM of -0.6 is below average with lots of SGs that are way way worse.. I probably need to create an article explaining how RAPM is used. When RAPM is used properly it has its uses. When used improperly its a case of a little bit of knowledge being a bad thing. you are using the stat improperly. lol my bad, but you do agree that T.Ross is the 6th best SG in the league at defense? And in the more "positionless era" is the stat not useless. Like Wiggins plays sg and sf. Simmons plays pg and sf, Greek freak plays everything, LBJ plays at least 3 spots and so on and so forth. What is the point of a stat that can only compare specific position groups? And what happened to Rubio, he was a great defender, now he's 33rd pg down with Lowry who you say is bad. Must have been a big effort drop off.
Rubio's ankle is fucked. Rubio palyed like 3495389 hard minutes as the primary ball carrier AND scorer in that EuroBasket thing in the summer. Ross has surprised me. Its too bad his shooting has gone to hell. He is working great in Vogel's system. If Ross continues like this all year he deserves the credit. I don't know what happened to Ross's shooting though.
Here is the thing with defense. Some players try harder than others. Lowry has stitches in his face 20 times a year and 5 black eyes every year. His level of effort is off the charts. If we trade him to Sacramento and give him 22 minutes a game behind D'aron Fox is Lowry going to be willing to lose 5 teeth every night like he is right now? Nah. Lowry will just "give up" like Hill has. And Lowry's #s will go to hell.. just like Hill's #s have.
That is why Hill's #s have tanked. The guy has just given up.
On November 22 2017 09:40 JimmiC wrote:
What is the point of a stat that can only compare specific position groups?
you are correct this limits the usability of RAPM/ESPN Real-Plus-Minus
|
|
|
On November 22 2017 09:49 JimmiC wrote:
Shouldn't hill be doing much better since he ages better since he's taller?
generally "big" PGs start to produce at peak levels at 26... much older than smaller PGs... and they stay at peak levels much older.. than smaller PGs.
However, there are other factors that go into it. if he is told he keeping his spot warm until Fox fully takes over a player in Sacramento might just give up. I think he has taken his foot off the gas.
always performing at peak effort is never 100% guaranteed. once effort is gone... that's that.
|
On November 22 2017 09:42 Twinkle Toes wrote:
Yea, ive read those and some others too, but could you please provide a comprehensive eli5.
how does one get from regular stats like points, rebounds, etc to plus minus?
And how about Adjusted? What is being adjusted?
And whats the difference between RAPM and xRAPM?
And you said something about taking into account 10 players, how does that happen? What is the formula?
Could you also explain how that works in a team, like how maybe Derozan gets his stats and the raptors team as a whole
RAPM = Regularized Adjusted Plus Minus. its method is published and publicly available.
xRAPM = EXpected Regularilzed Adjusted Plus Minus. It is a proprietary ESPN method made my Jeremias Engelmann. "Expected" means its a future prediction of future player behaviour. plain RAPM is just a statement about what actually happened. xRAPM tries to predict the future.
give me a day or two and i'll cover your questions.
|
|
|
How is Stockton just now getting in? Surely he retired way before Timmy
|
On November 22 2017 10:01 JimmiC wrote:
How can you use this stat to make decisions if effort is the biggest factor?
you can't. if basic effort disappears the stat won't work. this is true of any stat.
i think Lowry is going to throw himself into the stands 3 times next game and the stats describing his performance this year will do a good job of predicting his output in the next game.
However, if some family disaster happens to Lowry 2 hours before game time. Who knows what we'll get. Maybe Lowry will be distracted all game and play terribly.
Stats can never over ride the human factor.
On November 22 2017 10:01 JimmiC wrote:
I mean you stated Hill and Rubio were far superior to Teague and Lowry, with this stat and height as your reasoning.
please re-read my comment. i stated Hill is slightly better than Rubio and Rubio is slightly better than Teague. And the last few years that is how they've been.
Teague has performed as expected. He is an average to slightly above average PG. Hill has given up in Sacremento. Rubio is slowed by achilles tendonitis due to playing all summer long and he has no one to help him on offense in Utah.
I never said anything about Lowry in my previous comments.
both Hill and Rubio could potentially be better than Teague, but they've got other issues messing with their performance.
Rubio Playing All Summer
If Thibs knew Rubio was dead set on "representing his country" all summer it was a good move by Thibs to deal him. The Raptors have had big problems in the past with European players who basically "carry" their lousy teams to big summer Tourney wins. Its not like Lowry playing 11 minutes a game for team USA. Rubio and Dragic played like 40 minutes a game and were far and away their country's best players. They scored, made plays, did it all. Hard minutes.
On the Raptors Garbojosa , Calderone and Türkoğlu showed up to training camp dead tired and injured because they played all summer.
|
|
|
On November 22 2017 10:14 JimmiC wrote:
But if the stat goes up or down how could you know if its effort or not? You have basically described everyone who has "improved" that their effort has gone up. Like for irving you said his effort went way up this year compared to with the cavs even in the playoffs.
Wouldn't it be more likely that the stat does not get rid of all the "noise" and the roster changes in Min for wiggins and Irving playing in a completely different system with diff players is a bigger factor then effort?
the weakness of the stat is it doesn't give you the "why" it only gives you the #.
The "Regularlized" part of RAPM attempts to get rid of statistical noise. Its not perfect.. but it works well.
On November 22 2017 10:14 JimmiC wrote:
Wouldn't it be more likely that the stat does not get rid of all the "noise" and the roster changes in Min for wiggins and Irving playing in a completely different system with diff players is a bigger factor then effort?
Great #s with Low Minutes in a Great Defensive System
Yes, if you play in a different system where you thrive your #s will improve. Also, say you are Kyle Anderson and you play for the #1 coach genius Greg Popovich. Pop can surgically use Anderson only in situations where he shines. Giving Anderson great #s in a low # of minutes. Its like a 1-inning left handed relief pitcher than only gets used against left handed hitters. Give Kyle Anderson 30+ minutes a game and he is no where near the defensive wizard he was at 16 minutes a game.
Wiggins
Personally, i just see Wiggins trying harder on defense. If you don't see that.. that's ok. I'm just saying what i'm seeing. Thibs "system" has not changed. Wiggins is probably more knowledgeable as a defender than he was 1 year ago. Players in their early 20s learn a lot if they are willing to try hard.
More effort + more knowledge = improved play
Irving
Irving is trying harder from what i've seen. he is eccentric to say the least. Stevens is a master of player psychology and he has got Irving playing for him. Good coaches can take weird, eccentric personalities and make them work for them. That's my theory on why Irving is trying harder and playing better on Defense. Coach Stevens has done a nice job with him.
Danny Ainge recently acknowledged the earth might be flat.... clearly all Celtics Management want Irving on their side.
|
On November 22 2017 10:01 JimmyJRaynor wrote:Show nested quote +On November 22 2017 09:42 Twinkle Toes wrote:
Yea, ive read those and some others too, but could you please provide a comprehensive eli5.
how does one get from regular stats like points, rebounds, etc to plus minus?
And how about Adjusted? What is being adjusted?
And whats the difference between RAPM and xRAPM?
And you said something about taking into account 10 players, how does that happen? What is the formula?
Could you also explain how that works in a team, like how maybe Derozan gets his stats and the raptors team as a whole RAPM = Regularized Adjusted Plus Minus. its method is published and publicly available. xRAPM = EXpected Regularilzed Adjusted Plus Minus. It is a proprietary ESPN method made my Jeremias Engelmann. "Expected" means its a future prediction of future player behaviour. plain RAPM is just a statement about what actually happened. xRAPM tries to predict the future. give me a day or two and i'll cover your questions.
Ok please. I hear it so often and I hope to know it as well.
|
|
|
That's low key racism in the NBA.
On the other hand, I've also read an article last week about Manute Bol's hs coach or something claiming to fake his age, saying he could have been 30-40 when he entered the NBA.
|
|
|
Thunder crushing the Warriors!
|
|
|
|
|
|
EPT
