Current stars and their historic similars (Mike G, 2011)
Posted: Mon Apr 18, 2011 7:29 pm
Author Message
Mike G
Joined: 14 Jan 2005
Posts: 3604
Location: Hendersonville, NC
PostPosted: Thu Jan 20, 2011 9:30 am Post subject: Current stars and their historic similars Reply with quote
Euclidean differences between some players' 2010-11 rates and others' entire career rates. Not surprisingly, many players are most similar to their own careers.
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 James,Lebron 27.0 7.2 7.8 2.1 1.4 3.6 .4
.22 Lebron James 28.8 7.2 6.2 1.8 1.6 3.0 .8
.42 Dwyane Wade 27.9 5.2 6.1 2.5 1.8 3.7 .9
.52 Oscar Robertson 21.9 5.3 8.0 2.6 2.0 3.4 .7
.60 Tracy McGrady 24.0 6.6 5.1 2.0 1.4 2.5 1.0
.64 Kobe Bryant 27.0 5.5 4.5 2.7 1.5 2.9 .6
.68 Larry Bird 22.8 9.9 5.7 2.5 1.7 3.0 .8
.71 Michael Jordan 31.2 6.5 5.1 2.6 2.3 2.7 .8
The 'diff' number (other than 2011, which merely indicates this season -- the diff is really zero) sums the square roots of the differences in the square roots of the other categories.
LeBron is having a subpar year in scoring, TO, Blk; up in Ast.
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Wade,Dwyane 27.7 6.9 4.8 2.5 1.4 2.9 1.1
.18 Kobe Bryant 27.0 5.5 4.5 2.7 1.5 2.9 .6
.18 Lebron James 28.8 7.2 6.2 1.8 1.6 3.0 .8
.19 Tracy McGrady 24.0 6.6 5.1 2.0 1.4 2.5 1.0
.23 Paul Pierce 24.6 6.4 3.8 2.9 1.5 3.0 .6
.24 Michael Jordan 31.2 6.5 5.1 2.6 2.3 2.7 .8
.28 Dwyane Wade 27.9 5.2 6.1 2.5 1.8 3.7 .9
.29 Vince Carter 24.4 5.7 4.2 3.0 1.2 2.3 .7
Wade's become a bigger-rebounding, lesser-passing-and-TO version of himself. Who knew?
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Bryant,Kobe 29.5 5.7 5.3 2.3 1.4 3.2 .1
.29 Kobe Bryant 27.0 5.5 4.5 2.7 1.5 2.9 .6
.48 Lebron James 28.8 7.2 6.2 1.8 1.6 3.0 .8
.49 Gilbert Arenas 23.8 4.3 5.3 3.1 1.7 3.3 .2
.50 Michael Jordan 31.2 6.5 5.1 2.6 2.3 2.7 .8
.52 Dwyane Wade 27.9 5.2 6.1 2.5 1.8 3.7 .9
.52 Allen Iverson 25.1 3.5 5.6 1.8 2.0 3.2 .2
.57 Manu Ginobili 23.3 5.6 4.7 3.0 1.9 2.7 .4
Now for someone completely different --
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Howard,Dwight 23.9 14.5 1.3 3.5 1.2 3.6 2.3
.17 Dwight Howard 20.7 14.2 1.4 3.4 .9 3.1 2.2
.34 Moses Malone 21.0 13.2 1.3 2.6 .9 3.3 1.4
.37 Patrick Ewing 23.0 11.0 2.0 3.7 1.0 3.1 2.6
.52 George Mikan 26.6 13.1 2.9 4.4 1.3 3.6 2.0
.54 Bob Pettit 24.9 11.7 2.8 3.1 1.3 3.3 1.8
.54 Yao Ming 25.6 11.1 1.8 3.7 .4 3.0 2.1
.55 Hakeem Olajuwon 22.9 11.6 2.6 3.7 1.8 3.1 3.2
Mikan was a beast! He once scored something like 84% of his team's points in a game (16/19?). Pre-shot clock.
Stl and Blk are made up for years before 1974, TO before '78.
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Rose,Derrick 26.0 4.7 7.7 1.7 1.0 3.4 .6
.37 Dwyane Wade 27.9 5.2 6.1 2.5 1.8 3.7 .9
.47 Oscar Robertson 21.9 5.3 8.0 2.6 2.0 3.4 .7
.52 Lebron James 28.8 7.2 6.2 1.8 1.6 3.0 .8
.57 Allen Iverson 25.1 3.5 5.6 1.8 2.0 3.2 .2
.64 Tracy McGrady 24.0 6.6 5.1 2.0 1.4 2.5 1.0
.66 Jerry West 24.8 4.2 6.0 2.6 2.7 3.3 .9
.66 Kobe Bryant 27.0 5.5 4.5 2.7 1.5 2.9 .6
Rose is in territory new to him.
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Nowitzki,Dirk 27.3 8.2 2.6 2.6 .7 2.2 .8
.09 Dirk Nowitzki 25.7 9.1 2.6 2.6 .9 1.9 1.0
.27 Dominique Wilkins 24.7 7.0 2.5 2.0 1.3 2.6 .6
.35 Carmelo Anthony 25.1 6.5 2.8 3.1 1.2 3.1 .4
.46 Paul Arizin 23.2 6.3 2.3 3.7 1.1 2.6 .9
.46 George Yardley 21.9 7.7 2.0 3.3 1.0 2.6 1.2
.48 Karl Malone 27.1 10.9 3.5 3.1 1.4 3.0 .8
.52 Chris Bosh 21.3 9.8 1.9 2.6 .8 2.2 1.2
Dirk is most Dirklike so far; or more like previous self.
The career numbers do not contain this season.
Yardley was the first player to score 2000 in a season.
_________________
`
36% of all statistics are wrong
Last edited by Mike G on Thu Jan 20, 2011 1:32 pm; edited 1 time in total
Back to top
View user's profile Send private message Send e-mail
Mike G
Joined: 14 Jan 2005
Posts: 3604
Location: Hendersonville, NC
PostPosted: Thu Jan 20, 2011 9:52 am Post subject: Reply with quote
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Griffin,Blake 21.6 13.3 3.0 2.8 .7 2.6 .6
.11 Carlos Boozer 21.0 12.7 2.5 3.5 1.1 2.5 .5
.28 Charles Barkley 22.8 12.4 3.8 3.1 1.6 3.2 .8
.29 Brad Daugherty 19.6 10.1 3.6 2.7 .8 2.8 .8
.31 Walt Bellamy 18.4 11.0 2.3 3.4 .8 2.5 .7
.32 Zach Randolph 20.5 11.2 1.8 2.8 .9 2.6 .3
.40 Jeff Ruland 19.9 11.3 3.3 4.0 .8 4.0 .9
.48 Kevin Garnett 21.5 11.6 4.3 2.6 1.3 2.4 1.6
The new kid is the best rebounder of the bunch, so far. But these are career rates, and guys cool it on the boards after a while.
Bellamy's Stl/TO/Blk are based on his late years' stats, not affected by any reputation.
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Love,Kevin 19.7 15.7 2.1 2.2 .6 2.3 .3
.48 Carlos Boozer 21.0 12.7 2.5 3.5 1.1 2.5 .5
.49 Zach Randolph 20.5 11.2 1.8 2.8 .9 2.6 .3
.69 Walt Bellamy 18.4 11.0 2.3 3.4 .8 2.5 .7
.79 Moses Malone 21.0 13.2 1.3 2.6 .9 3.3 1.4
.88 Swen Nater 14.1 13.6 2.0 3.7 .6 2.9 .7
Boozer doesn't resemble either Love or Griffin, but whatever. Big rebounders who don't block shots.
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Duncan,Tim 18.1 12.2 4.0 2.2 .9 2.2 2.5
.32 Jerry Lucas 15.5 11.9 3.1 2.8 1.4 2.6 1.8
.33 Kevin Garnett 21.5 11.6 4.3 2.6 1.3 2.4 1.6
.42 Bill Walton 15.7 12.8 4.0 3.5 1.0 3.6 2.7
.42 Derrick Coleman 17.9 10.3 2.8 2.9 .8 2.8 1.5
.44 Elton Brand 20.5 10.4 2.5 3.1 .9 2.4 2.0
.46 Pau Gasol 21.0 9.5 3.1 2.4 .6 2.4 1.8
.50 Larry Nance 18.3 9.0 2.7 3.2 1.0 2.0 2.4
Tim used to be a much bigger scorer.
Note that he's better than Coleman at everything, but not by a lot at any.
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Westbrook,Russel 22.3 5.6 8.5 2.5 1.9 3.7 .4
.06 Oscar Robertson 21.9 5.3 8.0 2.6 2.0 3.4 .7
.29 Chris Paul 21.5 5.0 8.7 2.5 2.3 2.5 .1
.33 Bob Cousy 19.1 4.1 7.8 2.5 2.4 3.0 .4
.33 Magic Johnson 19.4 7.5 10.4 2.4 1.9 3.9 .4
.37 Steve Francis 19.2 6.0 6.3 3.2 1.5 3.5 .4
.44 Isiah Thomas 18.4 3.7 8.8 3.1 2.0 3.8 .3
.45 Kevin Johnson 19.0 3.6 9.0 2.3 1.5 3.3 .3
Hoo-wee, that's very close!
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Stoudemire,Amare 24.5 9.1 2.9 3.6 .9 3.5 2.3
.15 Kareem AbdulJabbar 24.7 10.6 3.4 3.1 1.0 3.0 2.7
.15 Bob McAdoo 22.5 9.9 2.3 3.6 1.1 3.3 1.7
.16 Neil Johnston 23.8 10.2 2.8 3.4 1.3 3.1 1.6
.22 Patrick Ewing 23.0 11.0 2.0 3.7 1.0 3.1 2.6
.26 Bob Pettit 24.9 11.7 2.8 3.1 1.3 3.3 1.8
.30 Bob Lanier 20.9 10.5 3.3 3.5 1.2 3.0 1.7
.32 Dolph Schayes 21.0 10.1 3.1 3.6 1.4 2.9 1.6
Amare' is passing and blocking more than usual. In fact, more like these guys.
Preponderance of pre-74 players his indicates Stl/TO/Blk are just about par considering his other rates.
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
kjb
Joined: 03 Jan 2005
Posts: 865
Location: Washington, DC
PostPosted: Thu Jan 20, 2011 10:25 am Post subject: Reply with quote
Not a star, but what says your statistical doppleganger machine about Nick Young?
And what's the official name of your "similars" machine? I'd like to use the info for a blog entry I'm doing about Young at the Wash Post.
Back to top
View user's profile Send private message AIM Address Yahoo Messenger
Mike G
Joined: 14 Jan 2005
Posts: 3604
Location: Hendersonville, NC
PostPosted: Thu Jan 20, 2011 1:56 pm Post subject: Reply with quote
It's the Euclidean Similizer, and this is unusual to compare season rates with career rates. But it occurred to me that some players may be having 'breakout' seasons that are going to end up being close to their career rates.
And so, it's a "what if" situation. Blake Griffin may be the next Barkley, and Derrick Rose may be the 2nd Oscar.
It's a bit dicier for a 29 mpg player such as Nick Young. But he's among my Most Improved, so maybe he'll stick. In which case ...
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Young,Nick 20.1 3.4 1.3 2.5 .8 1.5 .3
.15 Dale Ellis 18.8 4.4 1.7 2.5 1.0 1.8 .2
.18 Kiki Vandeweghe 21.2 4.0 2.1 2.3 .7 1.9 .4
.29 Glen Rice 20.1 4.9 2.2 2.5 1.0 1.8 .3
.31 Jeff Malone 20.1 3.0 2.5 2.1 .7 1.9 .1
.35 Peja Stojakovic 20.1 5.3 1.8 2.0 1.0 1.5 .2
.40 Allan Houston 20.8 3.4 2.7 2.6 .7 2.3 .2
.45 Dell Curry 17.1 3.8 2.6 2.5 1.4 1.8 .3
.45 Cazzie Russell 18.0 4.0 2.7 2.7 1.2 2.1 .4
.47 Michael Redd 22.7 4.6 2.3 1.9 1.0 1.8 .1
.51 Wally Szczerbiak 18.1 5.0 2.7 2.5 .7 1.8 .3
Ha -- I expected to see Jeff Malone. And due to Young's nearly nonexistent assist rate, he resembles as many forwards as guards. Non-rebounding forwards, that is.
How one-dimensional is Nick Young? If I insert him among my All-time 650 'substantial' careers, these are the lowest in my 'versatility index':
Code:
.52 Young,Nick (2011 only)
.56 Chris Dudley
.57 Mark West
.57 Mark Eaton
.57 Eddy Curry
.58 Greg Ostertag
.58 Samuel Dalembert
.58 Jeff Malone
.58 Steve Kerr
VI = (Sco*(Reb+Blk)*(Ast+Stl))^(1/2)/T
where T is the sum of the others (-TO)
Jeff Malone's lowest Ast/36 was 1.8, his last year in Utah. A veritable passing whiz next to Nick Young.
I see Young has had picked up his assists of late. Good for him.
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
Mike G
Joined: 14 Jan 2005
Posts: 3604
Location: Hendersonville, NC
PostPosted: Thu Jan 20, 2011 4:54 pm Post subject: Reply with quote
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Randolph,Zach 20.6 14.1 1.6 2.2 .7 1.9 .4
.27 Zach Randolph 20.5 11.2 1.8 2.8 .9 2.6 .3
.38 Carlos Boozer 21.0 12.7 2.5 3.5 1.1 2.5 .5
.54 Walt Bellamy 18.4 11.0 2.3 3.4 .8 2.5 .7
.59 Moses Malone 21.0 13.2 1.3 2.6 .9 3.3 1.4
.67 Chris Bosh 21.3 9.8 1.9 2.6 .8 2.2 1.2
.75 Al Jefferson 18.9 11.7 1.2 3.5 .9 2.0 1.6
.77 Drew Gooden 15.8 11.2 1.3 3.5 .9 2.1 .9
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Gasol,Pau 19.1 10.9 4.2 2.2 .7 1.8 1.9
.22 Pau Gasol 21.0 9.5 3.1 2.4 .6 2.4 1.8
.27 Kevin Garnett 21.5 11.6 4.3 2.6 1.3 2.4 1.6
.35 Derrick Coleman 17.9 10.3 2.8 2.9 .8 2.8 1.5
.37 Elton Brand 20.5 10.4 2.5 3.1 .9 2.4 2.0
.42 Brad Daugherty 19.6 10.1 3.6 2.7 .8 2.8 .8
.43 Chris Webber 20.4 9.9 4.2 3.3 1.4 2.8 1.4
.43 Larry Nance 18.3 9.0 2.7 3.2 1.0 2.0 2.4
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Paul,Chris 19.3 4.6 9.6 2.6 2.8 2.5 .1
.12 Chris Paul 21.5 5.0 8.7 2.5 2.3 2.5 .1
.31 Tim Hardaway 18.3 3.6 8.3 2.4 1.7 3.0 .2
.33 Bob Cousy 19.1 4.1 7.8 2.5 2.4 3.0 .4
.40 Isiah Thomas 18.4 3.7 8.8 3.1 2.0 3.8 .3
.40 Gary Payton 17.5 4.4 7.0 2.6 1.9 2.4 .2
.42 Kevin Johnson 19.0 3.6 9.0 2.3 1.5 3.3 .3
.43 Terrell Brandon 17.6 4.0 7.7 2.2 1.9 2.3 .4
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Williams,Deron 22.4 4.0 9.3 2.8 1.1 3.2 .2
.18 Deron Williams 19.1 3.7 8.1 2.8 1.1 3.1 .2
.23 Kevin Johnson 19.0 3.6 9.0 2.3 1.5 3.3 .3
.24 Stephon Marbury 19.6 3.1 7.7 2.4 1.2 3.0 .1
.27 Steve Nash 18.9 3.5 8.8 2.0 .8 3.3 .1
.28 Sam Cassell 19.5 4.0 7.1 3.3 1.2 2.9 .2
.33 Tim Hardaway 18.3 3.6 8.3 2.4 1.7 3.0 .2
.37 Isiah Thomas 18.4 3.7 8.8 3.1 2.0 3.8 .3
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Durant,Kevin 26.9 6.3 2.8 1.9 1.0 3.1 .9
.14 Dominique Wilkins 24.7 7.0 2.5 2.0 1.3 2.6 .6
.26 Carmelo Anthony 25.1 6.5 2.8 3.1 1.2 3.1 .4
.30 Paul Pierce 24.6 6.4 3.8 2.9 1.5 3.0 .6
.31 George Gervin 24.3 5.5 2.6 3.3 1.3 3.2 1.1
.40 Kobe Bryant 27.0 5.5 4.5 2.7 1.5 2.9 .6
.41 Vince Carter 24.4 5.7 4.2 3.0 1.2 2.3 .7
.43 Adrian Dantley 24.7 5.9 2.8 2.8 1.0 3.0 .2
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Ginobili,Manu 23.5 4.8 5.6 2.1 2.1 2.8 .5
.16 Manu Ginobili 23.3 5.6 4.7 3.0 1.9 2.7 .4
.19 Jerry West 24.8 4.2 6.0 2.6 2.7 3.3 .9
.19 Rick Barry 22.1 5.5 4.5 3.0 2.1 2.8 .5
.20 Gilbert Arenas 23.8 4.3 5.3 3.1 1.7 3.3 .2
.22 Allen Iverson 25.1 3.5 5.6 1.8 2.0 3.2 .2
.30 Kobe Bryant 27.0 5.5 4.5 2.7 1.5 2.9 .6
.30 John Havlicek 19.5 5.2 4.7 2.7 1.8 2.6 .7
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
erivera7
Joined: 19 Jan 2009
Posts: 184
Location: Chicago, IL
PostPosted: Thu Jan 20, 2011 5:13 pm Post subject: Reply with quote
This is pretty cool, Mike. Thanks for sharing.
Is it possible you could do these similarities for stars in previous years?
_________________
@erivera7
I cover the Orlando Magic - Magic Basketball
Back to top
View user's profile Send private message
deepak
Joined: 26 Apr 2006
Posts: 665
PostPosted: Thu Jan 20, 2011 6:35 pm Post subject: Reply with quote
How did you arrive at steals and blocks and turnovers for players from the 50s/60s?
Back to top
View user's profile Send private message
Mike G
Joined: 14 Jan 2005
Posts: 3604
Location: Hendersonville, NC
PostPosted: Thu Jan 20, 2011 8:36 pm Post subject: Reply with quote
erivera7 wrote:
Is it possible you could do these similarities for stars in previous years?
I have done these for a few years on another forum.
2010 - http://apbr.org/forum/viewtopic.php?t=3157
2009 - http://apbr.org/forum/viewtopic.php?t=2337
2008 - http://apbr.org/forum/viewtopic.php?t=1838 -- just a handful
2007 - http://apbr.org/forum/viewtopic.php?t=398
deepak wrote:
How did you arrive at steals and blocks and turnovers for players from the 50s/60s?
Very crudely: Per36 --
Stl = sqrt(Ast) - 0.4
Blk = sqrt(Reb) - 1.6
TO = .08*Sco + .07*Reb + .16*Ast + .05*Stl + .10*Blk - .005*Min/G
The advantage of having something other than zero in these categories, of course, is that we can at least imagine comparable players across eras.
The drawback is that if a modern player happens to have Stl/Blk/TO close to what these estimators produce, he'll be guessed to be similar to a lot of old-timers.
Sometimes I list most similar player-season from each decade.
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
Mike G
Joined: 14 Jan 2005
Posts: 3604
Location: Hendersonville, NC
PostPosted: Fri Jan 21, 2011 9:06 am Post subject: Reply with quote
Moving toward borderline All-Star types.
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Nash,Steve 18.6 4.3 11.9 1.2 .7 3.9 .0
.41 Steve Nash 18.9 3.5 8.8 2.0 .8 3.3 .1
.68 Kevin Johnson 19.0 3.6 9.0 2.3 1.5 3.3 .3
.73 Mark Price 19.0 3.2 7.9 1.6 1.4 3.0 .1
.88 Tim Hardaway 18.3 3.6 8.3 2.4 1.7 3.0 .2
.89 Deron Williams 19.1 3.7 8.1 2.8 1.1 3.1 .2
.90 Stephon Marbury 19.6 3.1 7.7 2.4 1.2 3.0 .1
.96 Rod Strickland 15.0 4.5 8.6 2.2 1.7 3.1 .2
The Nashster is now the most-unique player seen thus far. Extremely high Ast, along with high TO and extremely low PF, Stl, Blk.
This is the highest Reb rate he's ever had.
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Pierce,Paul 22.8 6.2 3.6 3.0 1.1 1.9 .6
.09 Vince Carter 24.4 5.7 4.2 3.0 1.2 2.3 .7
.12 Alex English 21.3 5.9 3.6 2.9 1.0 2.7 .8
.15 Mark Aguirre 22.2 6.0 3.4 3.3 .9 2.9 .4
.19 Paul Pierce 24.6 6.4 3.8 2.9 1.5 3.0 .6
.21 Marques Johnson 20.4 7.5 3.6 2.7 1.4 2.5 .8
.21 Ed Macauley 20.4 6.4 3.4 2.6 1.4 2.6 .9
.24 James Worthy 19.3 5.8 3.1 2.4 1.3 2.2 .7
Classic SF -- 48 similars within .50 diff.
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Millsap,Paul 19.6 9.4 2.6 3.8 1.5 1.6 1.0
.11 Terry Cummings 19.3 9.3 2.2 4.0 1.3 2.3 .7
.20 Dan Issel 20.7 8.5 2.2 3.1 1.1 2.3 .6
.23 Rasheed Wallace 17.7 8.1 2.1 3.4 1.1 1.7 1.5
.24 Josh Howard 18.5 7.6 1.9 3.1 1.2 1.8 .7
.24 Vern Mikkelsen 18.2 8.5 2.4 4.6 1.2 2.5 1.3
.24 Bailey Howell 19.1 8.2 2.0 4.2 1.0 2.4 1.3
.25 David West 19.1 8.8 1.9 2.9 .8 1.9 .9
The classic PF -- 58 players within .50 diff.
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 West,David 22.1 8.2 2.5 3.1 .9 2.0 .9
.08 Dan Issel 20.7 8.5 2.2 3.1 1.1 2.3 .6
.09 George Yardley 21.9 7.7 2.0 3.3 1.0 2.6 1.2
.12 Shareef AbdurRahim 20.1 8.5 2.8 3.0 1.1 2.7 .8
.16 Chris Bosh 21.3 9.8 1.9 2.6 .8 2.2 1.2
.18 David West 19.1 8.8 1.9 2.9 .8 1.9 .9
.19 Dirk Nowitzki 25.7 9.1 2.6 2.6 .9 1.9 1.0
.21 Paul Arizin 23.2 6.3 2.3 3.7 1.1 2.6 .9
West has moved into rather elite Sco/Reb territory.
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Smith,Josh 18.0 10.1 3.8 2.9 1.4 2.8 2.0
.14 Chris Webber 20.4 9.9 4.2 3.3 1.4 2.8 1.4
.18 Bob Lanier 20.9 10.5 3.3 3.5 1.2 3.0 1.7
.18 Derrick Coleman 17.9 10.3 2.8 2.9 .8 2.8 1.5
.20 Arvydas Sabonis 18.3 10.8 3.0 3.9 1.1 2.4 1.4
.21 Jerry Lucas 15.5 11.9 3.1 2.8 1.4 2.6 1.8
.22 Dolph Schayes 21.0 10.1 3.1 3.6 1.4 2.9 1.6
.26 Josh Smith 15.5 8.9 2.7 3.2 1.4 2.7 2.5
Smoove's career keeps evolving. More scoring and rebounding, less shotblocking.
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
kjb
Joined: 03 Jan 2005
Posts: 865
Location: Washington, DC
PostPosted: Fri Jan 21, 2011 10:20 am Post subject: Reply with quote
Thanks for the info on Nick Young. Good to know that Young is one of the least versatile players ever. Smile
The consensus among Wizards fans seems to be that Young has arrived as a player. I feel like Crash Davis in Bull Durham. Yeah, he scores, and he will play defense a bit. But he doesn't do anything else.
Back to top
View user's profile Send private message AIM Address Yahoo Messenger
acollard
Joined: 22 Sep 2010
Posts: 56
Location: MA
PostPosted: Fri Jan 21, 2011 3:47 pm Post subject: Reply with quote
Thanks for that!
I did some Euclidean distance similarities for teams this year in an old Daily Thunder post.
Just curious as to your reasoning for /36min stats instead of /100 possessions, which could help compare players from different eras and differently paced teams. Also, FG% 3PT% and other scoring efficiency metrics seem like they'd be easy to add, and would help when comparing players. Westbrook and Paul have similar normal stats, but Russ has a much different scoring efficiency compared to Paul, and that makes a huge difference.
Each player's third season, for example: http://bkref.com/tiny/i679G
Back to top
View user's profile Send private message
Mike G
Joined: 14 Jan 2005
Posts: 3604
Location: Hendersonville, NC
PostPosted: Fri Jan 21, 2011 4:56 pm Post subject: Reply with quote
Per36, unless I specify 'raw' or 'undadjusted' refers to per 36 minutes, per 100 points per team, per 44 rebounds per team. These are the standards (historic norms) to which these numbers are scaled.
The numbers don't care if they were accrued in high-pace and low-efficiency, or high-eff and low-pace. Just as rebounds as a % of all rebounds are indicative of rebounding ability, points as a % of all points are indicative of scoring ability.
While it's arguable that high% shooting is better than lesser % shooting, it's also arguable that just being able to score in a defensive milieu, when others can't, is worth something: 45% is very good when others are shooting 40%.
So 15 pts/36 Min/100 pts/team is average. Whether your coach pursues a high pace or high efficiency, it's the fraction of points you score that determines your effectiveness.
Shooting % is built into the scoring (Sco) portion of these per36 rates. But the more factors you include in these similarities, the less each factor can count. Since I've endeavored to capture production, I don't care about the means of achieving these productions.
It's quite possible that if you traded Paul for Westbrook, CP would shoot more, at a lower %, and vice versa. Both their Sco stats would likely not change much.
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
acollard
Joined: 22 Sep 2010
Posts: 56
Location: MA
PostPosted: Fri Jan 21, 2011 9:14 pm Post subject: Reply with quote
apologies! that explains a lot, and makes a lot of sense.
Back to top
View user's profile Send private message
acollard
Joined: 22 Sep 2010
Posts: 56
Location: MA
PostPosted: Fri Jan 21, 2011 9:18 pm Post subject: Reply with quote
Mike G wrote:
The numbers don't care if they were accrued in high-pace and low-efficiency, or high-eff and low-pace. Just as rebounds as a % of all rebounds are indicative of rebounding ability, points as a % of all points are indicative of scoring ability.
Whoa wait... I just read it again. How does that make sense? If you're on a team with other bad scorers, that would effect the per 100 pts calculation. Also, shouldn't we care if someone scores high-pace low efficiency compared to high efficiency and low pace? Those are fundamentally different, almost the opposite in fact.
Back to top
View user's profile Send private message
Mike G
Joined: 14 Jan 2005
Posts: 3604
Location: Hendersonville, NC
PostPosted: Sat Jan 22, 2011 7:57 am Post subject: Reply with quote
Good point, and you've caught me oversimplifying the explanation.
I've refined Reb% to be closer to % of opponent rebounds. Even though you aren't literally grabbing the opponents' rebounds, your rebounds (a number) are still a known fraction of that total.
The theory is that rebounds grabbed by your teammates are not 'available'; you should be vying for those the opposition is vying for.
Same deal for scoring. Scoring 20 in a 100-90 win is better than scoring 20 for the losers. One is .200 of the opponent total, and the other is .222 .
Basically, every player's Sco rate includes the factor (100/OppPPG).
Yes, a coach should prefer high pace or high efficiency players to match his style. Different players will flourish or languish, depending. David Lee isn't flourishing in GS, for example. Amare' loves NY, etc.
Michael Beasley shoots .530, averaging 22.3 Pts per 36 min. for a team that allows 108.4 ppg.
David West shoots .562, 19.5 Pts/36, for a team giving up 91.4 .
Given that these team PPG allowed are what the NBA at large scores in the games these guys are in, it makes some sense to scale their scoring to these parameters.
This adjustment for West is = 19.5 * 100/91.4 = 21.3 Pts/36/100ppg
For Beasley, its 22.3 * 100/108.4 = 20.6
In the simplest terms, I'd expect West to outscore Beasley in a game between their teams. I'd say he's the better scorer vs anyone.
_________________
`
36% of all statistics are wrong
Mike G
Joined: 14 Jan 2005
Posts: 3604
Location: Hendersonville, NC
PostPosted: Thu Jan 20, 2011 9:30 am Post subject: Current stars and their historic similars Reply with quote
Euclidean differences between some players' 2010-11 rates and others' entire career rates. Not surprisingly, many players are most similar to their own careers.
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 James,Lebron 27.0 7.2 7.8 2.1 1.4 3.6 .4
.22 Lebron James 28.8 7.2 6.2 1.8 1.6 3.0 .8
.42 Dwyane Wade 27.9 5.2 6.1 2.5 1.8 3.7 .9
.52 Oscar Robertson 21.9 5.3 8.0 2.6 2.0 3.4 .7
.60 Tracy McGrady 24.0 6.6 5.1 2.0 1.4 2.5 1.0
.64 Kobe Bryant 27.0 5.5 4.5 2.7 1.5 2.9 .6
.68 Larry Bird 22.8 9.9 5.7 2.5 1.7 3.0 .8
.71 Michael Jordan 31.2 6.5 5.1 2.6 2.3 2.7 .8
The 'diff' number (other than 2011, which merely indicates this season -- the diff is really zero) sums the square roots of the differences in the square roots of the other categories.
LeBron is having a subpar year in scoring, TO, Blk; up in Ast.
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Wade,Dwyane 27.7 6.9 4.8 2.5 1.4 2.9 1.1
.18 Kobe Bryant 27.0 5.5 4.5 2.7 1.5 2.9 .6
.18 Lebron James 28.8 7.2 6.2 1.8 1.6 3.0 .8
.19 Tracy McGrady 24.0 6.6 5.1 2.0 1.4 2.5 1.0
.23 Paul Pierce 24.6 6.4 3.8 2.9 1.5 3.0 .6
.24 Michael Jordan 31.2 6.5 5.1 2.6 2.3 2.7 .8
.28 Dwyane Wade 27.9 5.2 6.1 2.5 1.8 3.7 .9
.29 Vince Carter 24.4 5.7 4.2 3.0 1.2 2.3 .7
Wade's become a bigger-rebounding, lesser-passing-and-TO version of himself. Who knew?
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Bryant,Kobe 29.5 5.7 5.3 2.3 1.4 3.2 .1
.29 Kobe Bryant 27.0 5.5 4.5 2.7 1.5 2.9 .6
.48 Lebron James 28.8 7.2 6.2 1.8 1.6 3.0 .8
.49 Gilbert Arenas 23.8 4.3 5.3 3.1 1.7 3.3 .2
.50 Michael Jordan 31.2 6.5 5.1 2.6 2.3 2.7 .8
.52 Dwyane Wade 27.9 5.2 6.1 2.5 1.8 3.7 .9
.52 Allen Iverson 25.1 3.5 5.6 1.8 2.0 3.2 .2
.57 Manu Ginobili 23.3 5.6 4.7 3.0 1.9 2.7 .4
Now for someone completely different --
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Howard,Dwight 23.9 14.5 1.3 3.5 1.2 3.6 2.3
.17 Dwight Howard 20.7 14.2 1.4 3.4 .9 3.1 2.2
.34 Moses Malone 21.0 13.2 1.3 2.6 .9 3.3 1.4
.37 Patrick Ewing 23.0 11.0 2.0 3.7 1.0 3.1 2.6
.52 George Mikan 26.6 13.1 2.9 4.4 1.3 3.6 2.0
.54 Bob Pettit 24.9 11.7 2.8 3.1 1.3 3.3 1.8
.54 Yao Ming 25.6 11.1 1.8 3.7 .4 3.0 2.1
.55 Hakeem Olajuwon 22.9 11.6 2.6 3.7 1.8 3.1 3.2
Mikan was a beast! He once scored something like 84% of his team's points in a game (16/19?). Pre-shot clock.
Stl and Blk are made up for years before 1974, TO before '78.
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Rose,Derrick 26.0 4.7 7.7 1.7 1.0 3.4 .6
.37 Dwyane Wade 27.9 5.2 6.1 2.5 1.8 3.7 .9
.47 Oscar Robertson 21.9 5.3 8.0 2.6 2.0 3.4 .7
.52 Lebron James 28.8 7.2 6.2 1.8 1.6 3.0 .8
.57 Allen Iverson 25.1 3.5 5.6 1.8 2.0 3.2 .2
.64 Tracy McGrady 24.0 6.6 5.1 2.0 1.4 2.5 1.0
.66 Jerry West 24.8 4.2 6.0 2.6 2.7 3.3 .9
.66 Kobe Bryant 27.0 5.5 4.5 2.7 1.5 2.9 .6
Rose is in territory new to him.
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Nowitzki,Dirk 27.3 8.2 2.6 2.6 .7 2.2 .8
.09 Dirk Nowitzki 25.7 9.1 2.6 2.6 .9 1.9 1.0
.27 Dominique Wilkins 24.7 7.0 2.5 2.0 1.3 2.6 .6
.35 Carmelo Anthony 25.1 6.5 2.8 3.1 1.2 3.1 .4
.46 Paul Arizin 23.2 6.3 2.3 3.7 1.1 2.6 .9
.46 George Yardley 21.9 7.7 2.0 3.3 1.0 2.6 1.2
.48 Karl Malone 27.1 10.9 3.5 3.1 1.4 3.0 .8
.52 Chris Bosh 21.3 9.8 1.9 2.6 .8 2.2 1.2
Dirk is most Dirklike so far; or more like previous self.
The career numbers do not contain this season.
Yardley was the first player to score 2000 in a season.
_________________
`
36% of all statistics are wrong
Last edited by Mike G on Thu Jan 20, 2011 1:32 pm; edited 1 time in total
Back to top
View user's profile Send private message Send e-mail
Mike G
Joined: 14 Jan 2005
Posts: 3604
Location: Hendersonville, NC
PostPosted: Thu Jan 20, 2011 9:52 am Post subject: Reply with quote
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Griffin,Blake 21.6 13.3 3.0 2.8 .7 2.6 .6
.11 Carlos Boozer 21.0 12.7 2.5 3.5 1.1 2.5 .5
.28 Charles Barkley 22.8 12.4 3.8 3.1 1.6 3.2 .8
.29 Brad Daugherty 19.6 10.1 3.6 2.7 .8 2.8 .8
.31 Walt Bellamy 18.4 11.0 2.3 3.4 .8 2.5 .7
.32 Zach Randolph 20.5 11.2 1.8 2.8 .9 2.6 .3
.40 Jeff Ruland 19.9 11.3 3.3 4.0 .8 4.0 .9
.48 Kevin Garnett 21.5 11.6 4.3 2.6 1.3 2.4 1.6
The new kid is the best rebounder of the bunch, so far. But these are career rates, and guys cool it on the boards after a while.
Bellamy's Stl/TO/Blk are based on his late years' stats, not affected by any reputation.
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Love,Kevin 19.7 15.7 2.1 2.2 .6 2.3 .3
.48 Carlos Boozer 21.0 12.7 2.5 3.5 1.1 2.5 .5
.49 Zach Randolph 20.5 11.2 1.8 2.8 .9 2.6 .3
.69 Walt Bellamy 18.4 11.0 2.3 3.4 .8 2.5 .7
.79 Moses Malone 21.0 13.2 1.3 2.6 .9 3.3 1.4
.88 Swen Nater 14.1 13.6 2.0 3.7 .6 2.9 .7
Boozer doesn't resemble either Love or Griffin, but whatever. Big rebounders who don't block shots.
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Duncan,Tim 18.1 12.2 4.0 2.2 .9 2.2 2.5
.32 Jerry Lucas 15.5 11.9 3.1 2.8 1.4 2.6 1.8
.33 Kevin Garnett 21.5 11.6 4.3 2.6 1.3 2.4 1.6
.42 Bill Walton 15.7 12.8 4.0 3.5 1.0 3.6 2.7
.42 Derrick Coleman 17.9 10.3 2.8 2.9 .8 2.8 1.5
.44 Elton Brand 20.5 10.4 2.5 3.1 .9 2.4 2.0
.46 Pau Gasol 21.0 9.5 3.1 2.4 .6 2.4 1.8
.50 Larry Nance 18.3 9.0 2.7 3.2 1.0 2.0 2.4
Tim used to be a much bigger scorer.
Note that he's better than Coleman at everything, but not by a lot at any.
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Westbrook,Russel 22.3 5.6 8.5 2.5 1.9 3.7 .4
.06 Oscar Robertson 21.9 5.3 8.0 2.6 2.0 3.4 .7
.29 Chris Paul 21.5 5.0 8.7 2.5 2.3 2.5 .1
.33 Bob Cousy 19.1 4.1 7.8 2.5 2.4 3.0 .4
.33 Magic Johnson 19.4 7.5 10.4 2.4 1.9 3.9 .4
.37 Steve Francis 19.2 6.0 6.3 3.2 1.5 3.5 .4
.44 Isiah Thomas 18.4 3.7 8.8 3.1 2.0 3.8 .3
.45 Kevin Johnson 19.0 3.6 9.0 2.3 1.5 3.3 .3
Hoo-wee, that's very close!
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Stoudemire,Amare 24.5 9.1 2.9 3.6 .9 3.5 2.3
.15 Kareem AbdulJabbar 24.7 10.6 3.4 3.1 1.0 3.0 2.7
.15 Bob McAdoo 22.5 9.9 2.3 3.6 1.1 3.3 1.7
.16 Neil Johnston 23.8 10.2 2.8 3.4 1.3 3.1 1.6
.22 Patrick Ewing 23.0 11.0 2.0 3.7 1.0 3.1 2.6
.26 Bob Pettit 24.9 11.7 2.8 3.1 1.3 3.3 1.8
.30 Bob Lanier 20.9 10.5 3.3 3.5 1.2 3.0 1.7
.32 Dolph Schayes 21.0 10.1 3.1 3.6 1.4 2.9 1.6
Amare' is passing and blocking more than usual. In fact, more like these guys.
Preponderance of pre-74 players his indicates Stl/TO/Blk are just about par considering his other rates.
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
kjb
Joined: 03 Jan 2005
Posts: 865
Location: Washington, DC
PostPosted: Thu Jan 20, 2011 10:25 am Post subject: Reply with quote
Not a star, but what says your statistical doppleganger machine about Nick Young?
And what's the official name of your "similars" machine? I'd like to use the info for a blog entry I'm doing about Young at the Wash Post.
Back to top
View user's profile Send private message AIM Address Yahoo Messenger
Mike G
Joined: 14 Jan 2005
Posts: 3604
Location: Hendersonville, NC
PostPosted: Thu Jan 20, 2011 1:56 pm Post subject: Reply with quote
It's the Euclidean Similizer, and this is unusual to compare season rates with career rates. But it occurred to me that some players may be having 'breakout' seasons that are going to end up being close to their career rates.
And so, it's a "what if" situation. Blake Griffin may be the next Barkley, and Derrick Rose may be the 2nd Oscar.
It's a bit dicier for a 29 mpg player such as Nick Young. But he's among my Most Improved, so maybe he'll stick. In which case ...
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Young,Nick 20.1 3.4 1.3 2.5 .8 1.5 .3
.15 Dale Ellis 18.8 4.4 1.7 2.5 1.0 1.8 .2
.18 Kiki Vandeweghe 21.2 4.0 2.1 2.3 .7 1.9 .4
.29 Glen Rice 20.1 4.9 2.2 2.5 1.0 1.8 .3
.31 Jeff Malone 20.1 3.0 2.5 2.1 .7 1.9 .1
.35 Peja Stojakovic 20.1 5.3 1.8 2.0 1.0 1.5 .2
.40 Allan Houston 20.8 3.4 2.7 2.6 .7 2.3 .2
.45 Dell Curry 17.1 3.8 2.6 2.5 1.4 1.8 .3
.45 Cazzie Russell 18.0 4.0 2.7 2.7 1.2 2.1 .4
.47 Michael Redd 22.7 4.6 2.3 1.9 1.0 1.8 .1
.51 Wally Szczerbiak 18.1 5.0 2.7 2.5 .7 1.8 .3
Ha -- I expected to see Jeff Malone. And due to Young's nearly nonexistent assist rate, he resembles as many forwards as guards. Non-rebounding forwards, that is.
How one-dimensional is Nick Young? If I insert him among my All-time 650 'substantial' careers, these are the lowest in my 'versatility index':
Code:
.52 Young,Nick (2011 only)
.56 Chris Dudley
.57 Mark West
.57 Mark Eaton
.57 Eddy Curry
.58 Greg Ostertag
.58 Samuel Dalembert
.58 Jeff Malone
.58 Steve Kerr
VI = (Sco*(Reb+Blk)*(Ast+Stl))^(1/2)/T
where T is the sum of the others (-TO)
Jeff Malone's lowest Ast/36 was 1.8, his last year in Utah. A veritable passing whiz next to Nick Young.
I see Young has had picked up his assists of late. Good for him.
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
Mike G
Joined: 14 Jan 2005
Posts: 3604
Location: Hendersonville, NC
PostPosted: Thu Jan 20, 2011 4:54 pm Post subject: Reply with quote
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Randolph,Zach 20.6 14.1 1.6 2.2 .7 1.9 .4
.27 Zach Randolph 20.5 11.2 1.8 2.8 .9 2.6 .3
.38 Carlos Boozer 21.0 12.7 2.5 3.5 1.1 2.5 .5
.54 Walt Bellamy 18.4 11.0 2.3 3.4 .8 2.5 .7
.59 Moses Malone 21.0 13.2 1.3 2.6 .9 3.3 1.4
.67 Chris Bosh 21.3 9.8 1.9 2.6 .8 2.2 1.2
.75 Al Jefferson 18.9 11.7 1.2 3.5 .9 2.0 1.6
.77 Drew Gooden 15.8 11.2 1.3 3.5 .9 2.1 .9
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Gasol,Pau 19.1 10.9 4.2 2.2 .7 1.8 1.9
.22 Pau Gasol 21.0 9.5 3.1 2.4 .6 2.4 1.8
.27 Kevin Garnett 21.5 11.6 4.3 2.6 1.3 2.4 1.6
.35 Derrick Coleman 17.9 10.3 2.8 2.9 .8 2.8 1.5
.37 Elton Brand 20.5 10.4 2.5 3.1 .9 2.4 2.0
.42 Brad Daugherty 19.6 10.1 3.6 2.7 .8 2.8 .8
.43 Chris Webber 20.4 9.9 4.2 3.3 1.4 2.8 1.4
.43 Larry Nance 18.3 9.0 2.7 3.2 1.0 2.0 2.4
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Paul,Chris 19.3 4.6 9.6 2.6 2.8 2.5 .1
.12 Chris Paul 21.5 5.0 8.7 2.5 2.3 2.5 .1
.31 Tim Hardaway 18.3 3.6 8.3 2.4 1.7 3.0 .2
.33 Bob Cousy 19.1 4.1 7.8 2.5 2.4 3.0 .4
.40 Isiah Thomas 18.4 3.7 8.8 3.1 2.0 3.8 .3
.40 Gary Payton 17.5 4.4 7.0 2.6 1.9 2.4 .2
.42 Kevin Johnson 19.0 3.6 9.0 2.3 1.5 3.3 .3
.43 Terrell Brandon 17.6 4.0 7.7 2.2 1.9 2.3 .4
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Williams,Deron 22.4 4.0 9.3 2.8 1.1 3.2 .2
.18 Deron Williams 19.1 3.7 8.1 2.8 1.1 3.1 .2
.23 Kevin Johnson 19.0 3.6 9.0 2.3 1.5 3.3 .3
.24 Stephon Marbury 19.6 3.1 7.7 2.4 1.2 3.0 .1
.27 Steve Nash 18.9 3.5 8.8 2.0 .8 3.3 .1
.28 Sam Cassell 19.5 4.0 7.1 3.3 1.2 2.9 .2
.33 Tim Hardaway 18.3 3.6 8.3 2.4 1.7 3.0 .2
.37 Isiah Thomas 18.4 3.7 8.8 3.1 2.0 3.8 .3
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Durant,Kevin 26.9 6.3 2.8 1.9 1.0 3.1 .9
.14 Dominique Wilkins 24.7 7.0 2.5 2.0 1.3 2.6 .6
.26 Carmelo Anthony 25.1 6.5 2.8 3.1 1.2 3.1 .4
.30 Paul Pierce 24.6 6.4 3.8 2.9 1.5 3.0 .6
.31 George Gervin 24.3 5.5 2.6 3.3 1.3 3.2 1.1
.40 Kobe Bryant 27.0 5.5 4.5 2.7 1.5 2.9 .6
.41 Vince Carter 24.4 5.7 4.2 3.0 1.2 2.3 .7
.43 Adrian Dantley 24.7 5.9 2.8 2.8 1.0 3.0 .2
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Ginobili,Manu 23.5 4.8 5.6 2.1 2.1 2.8 .5
.16 Manu Ginobili 23.3 5.6 4.7 3.0 1.9 2.7 .4
.19 Jerry West 24.8 4.2 6.0 2.6 2.7 3.3 .9
.19 Rick Barry 22.1 5.5 4.5 3.0 2.1 2.8 .5
.20 Gilbert Arenas 23.8 4.3 5.3 3.1 1.7 3.3 .2
.22 Allen Iverson 25.1 3.5 5.6 1.8 2.0 3.2 .2
.30 Kobe Bryant 27.0 5.5 4.5 2.7 1.5 2.9 .6
.30 John Havlicek 19.5 5.2 4.7 2.7 1.8 2.6 .7
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
erivera7
Joined: 19 Jan 2009
Posts: 184
Location: Chicago, IL
PostPosted: Thu Jan 20, 2011 5:13 pm Post subject: Reply with quote
This is pretty cool, Mike. Thanks for sharing.
Is it possible you could do these similarities for stars in previous years?
_________________
@erivera7
I cover the Orlando Magic - Magic Basketball
Back to top
View user's profile Send private message
deepak
Joined: 26 Apr 2006
Posts: 665
PostPosted: Thu Jan 20, 2011 6:35 pm Post subject: Reply with quote
How did you arrive at steals and blocks and turnovers for players from the 50s/60s?
Back to top
View user's profile Send private message
Mike G
Joined: 14 Jan 2005
Posts: 3604
Location: Hendersonville, NC
PostPosted: Thu Jan 20, 2011 8:36 pm Post subject: Reply with quote
erivera7 wrote:
Is it possible you could do these similarities for stars in previous years?
I have done these for a few years on another forum.
2010 - http://apbr.org/forum/viewtopic.php?t=3157
2009 - http://apbr.org/forum/viewtopic.php?t=2337
2008 - http://apbr.org/forum/viewtopic.php?t=1838 -- just a handful
2007 - http://apbr.org/forum/viewtopic.php?t=398
deepak wrote:
How did you arrive at steals and blocks and turnovers for players from the 50s/60s?
Very crudely: Per36 --
Stl = sqrt(Ast) - 0.4
Blk = sqrt(Reb) - 1.6
TO = .08*Sco + .07*Reb + .16*Ast + .05*Stl + .10*Blk - .005*Min/G
The advantage of having something other than zero in these categories, of course, is that we can at least imagine comparable players across eras.
The drawback is that if a modern player happens to have Stl/Blk/TO close to what these estimators produce, he'll be guessed to be similar to a lot of old-timers.
Sometimes I list most similar player-season from each decade.
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
Mike G
Joined: 14 Jan 2005
Posts: 3604
Location: Hendersonville, NC
PostPosted: Fri Jan 21, 2011 9:06 am Post subject: Reply with quote
Moving toward borderline All-Star types.
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Nash,Steve 18.6 4.3 11.9 1.2 .7 3.9 .0
.41 Steve Nash 18.9 3.5 8.8 2.0 .8 3.3 .1
.68 Kevin Johnson 19.0 3.6 9.0 2.3 1.5 3.3 .3
.73 Mark Price 19.0 3.2 7.9 1.6 1.4 3.0 .1
.88 Tim Hardaway 18.3 3.6 8.3 2.4 1.7 3.0 .2
.89 Deron Williams 19.1 3.7 8.1 2.8 1.1 3.1 .2
.90 Stephon Marbury 19.6 3.1 7.7 2.4 1.2 3.0 .1
.96 Rod Strickland 15.0 4.5 8.6 2.2 1.7 3.1 .2
The Nashster is now the most-unique player seen thus far. Extremely high Ast, along with high TO and extremely low PF, Stl, Blk.
This is the highest Reb rate he's ever had.
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Pierce,Paul 22.8 6.2 3.6 3.0 1.1 1.9 .6
.09 Vince Carter 24.4 5.7 4.2 3.0 1.2 2.3 .7
.12 Alex English 21.3 5.9 3.6 2.9 1.0 2.7 .8
.15 Mark Aguirre 22.2 6.0 3.4 3.3 .9 2.9 .4
.19 Paul Pierce 24.6 6.4 3.8 2.9 1.5 3.0 .6
.21 Marques Johnson 20.4 7.5 3.6 2.7 1.4 2.5 .8
.21 Ed Macauley 20.4 6.4 3.4 2.6 1.4 2.6 .9
.24 James Worthy 19.3 5.8 3.1 2.4 1.3 2.2 .7
Classic SF -- 48 similars within .50 diff.
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Millsap,Paul 19.6 9.4 2.6 3.8 1.5 1.6 1.0
.11 Terry Cummings 19.3 9.3 2.2 4.0 1.3 2.3 .7
.20 Dan Issel 20.7 8.5 2.2 3.1 1.1 2.3 .6
.23 Rasheed Wallace 17.7 8.1 2.1 3.4 1.1 1.7 1.5
.24 Josh Howard 18.5 7.6 1.9 3.1 1.2 1.8 .7
.24 Vern Mikkelsen 18.2 8.5 2.4 4.6 1.2 2.5 1.3
.24 Bailey Howell 19.1 8.2 2.0 4.2 1.0 2.4 1.3
.25 David West 19.1 8.8 1.9 2.9 .8 1.9 .9
The classic PF -- 58 players within .50 diff.
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 West,David 22.1 8.2 2.5 3.1 .9 2.0 .9
.08 Dan Issel 20.7 8.5 2.2 3.1 1.1 2.3 .6
.09 George Yardley 21.9 7.7 2.0 3.3 1.0 2.6 1.2
.12 Shareef AbdurRahim 20.1 8.5 2.8 3.0 1.1 2.7 .8
.16 Chris Bosh 21.3 9.8 1.9 2.6 .8 2.2 1.2
.18 David West 19.1 8.8 1.9 2.9 .8 1.9 .9
.19 Dirk Nowitzki 25.7 9.1 2.6 2.6 .9 1.9 1.0
.21 Paul Arizin 23.2 6.3 2.3 3.7 1.1 2.6 .9
West has moved into rather elite Sco/Reb territory.
Code:
diff career per36 Sco Reb Ast PF Stl TO Blk
2011 Smith,Josh 18.0 10.1 3.8 2.9 1.4 2.8 2.0
.14 Chris Webber 20.4 9.9 4.2 3.3 1.4 2.8 1.4
.18 Bob Lanier 20.9 10.5 3.3 3.5 1.2 3.0 1.7
.18 Derrick Coleman 17.9 10.3 2.8 2.9 .8 2.8 1.5
.20 Arvydas Sabonis 18.3 10.8 3.0 3.9 1.1 2.4 1.4
.21 Jerry Lucas 15.5 11.9 3.1 2.8 1.4 2.6 1.8
.22 Dolph Schayes 21.0 10.1 3.1 3.6 1.4 2.9 1.6
.26 Josh Smith 15.5 8.9 2.7 3.2 1.4 2.7 2.5
Smoove's career keeps evolving. More scoring and rebounding, less shotblocking.
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
kjb
Joined: 03 Jan 2005
Posts: 865
Location: Washington, DC
PostPosted: Fri Jan 21, 2011 10:20 am Post subject: Reply with quote
Thanks for the info on Nick Young. Good to know that Young is one of the least versatile players ever. Smile
The consensus among Wizards fans seems to be that Young has arrived as a player. I feel like Crash Davis in Bull Durham. Yeah, he scores, and he will play defense a bit. But he doesn't do anything else.
Back to top
View user's profile Send private message AIM Address Yahoo Messenger
acollard
Joined: 22 Sep 2010
Posts: 56
Location: MA
PostPosted: Fri Jan 21, 2011 3:47 pm Post subject: Reply with quote
Thanks for that!
I did some Euclidean distance similarities for teams this year in an old Daily Thunder post.
Just curious as to your reasoning for /36min stats instead of /100 possessions, which could help compare players from different eras and differently paced teams. Also, FG% 3PT% and other scoring efficiency metrics seem like they'd be easy to add, and would help when comparing players. Westbrook and Paul have similar normal stats, but Russ has a much different scoring efficiency compared to Paul, and that makes a huge difference.
Each player's third season, for example: http://bkref.com/tiny/i679G
Back to top
View user's profile Send private message
Mike G
Joined: 14 Jan 2005
Posts: 3604
Location: Hendersonville, NC
PostPosted: Fri Jan 21, 2011 4:56 pm Post subject: Reply with quote
Per36, unless I specify 'raw' or 'undadjusted' refers to per 36 minutes, per 100 points per team, per 44 rebounds per team. These are the standards (historic norms) to which these numbers are scaled.
The numbers don't care if they were accrued in high-pace and low-efficiency, or high-eff and low-pace. Just as rebounds as a % of all rebounds are indicative of rebounding ability, points as a % of all points are indicative of scoring ability.
While it's arguable that high% shooting is better than lesser % shooting, it's also arguable that just being able to score in a defensive milieu, when others can't, is worth something: 45% is very good when others are shooting 40%.
So 15 pts/36 Min/100 pts/team is average. Whether your coach pursues a high pace or high efficiency, it's the fraction of points you score that determines your effectiveness.
Shooting % is built into the scoring (Sco) portion of these per36 rates. But the more factors you include in these similarities, the less each factor can count. Since I've endeavored to capture production, I don't care about the means of achieving these productions.
It's quite possible that if you traded Paul for Westbrook, CP would shoot more, at a lower %, and vice versa. Both their Sco stats would likely not change much.
_________________
`
36% of all statistics are wrong
Back to top
View user's profile Send private message Send e-mail
acollard
Joined: 22 Sep 2010
Posts: 56
Location: MA
PostPosted: Fri Jan 21, 2011 9:14 pm Post subject: Reply with quote
apologies! that explains a lot, and makes a lot of sense.
Back to top
View user's profile Send private message
acollard
Joined: 22 Sep 2010
Posts: 56
Location: MA
PostPosted: Fri Jan 21, 2011 9:18 pm Post subject: Reply with quote
Mike G wrote:
The numbers don't care if they were accrued in high-pace and low-efficiency, or high-eff and low-pace. Just as rebounds as a % of all rebounds are indicative of rebounding ability, points as a % of all points are indicative of scoring ability.
Whoa wait... I just read it again. How does that make sense? If you're on a team with other bad scorers, that would effect the per 100 pts calculation. Also, shouldn't we care if someone scores high-pace low efficiency compared to high efficiency and low pace? Those are fundamentally different, almost the opposite in fact.
Back to top
View user's profile Send private message
Mike G
Joined: 14 Jan 2005
Posts: 3604
Location: Hendersonville, NC
PostPosted: Sat Jan 22, 2011 7:57 am Post subject: Reply with quote
Good point, and you've caught me oversimplifying the explanation.
I've refined Reb% to be closer to % of opponent rebounds. Even though you aren't literally grabbing the opponents' rebounds, your rebounds (a number) are still a known fraction of that total.
The theory is that rebounds grabbed by your teammates are not 'available'; you should be vying for those the opposition is vying for.
Same deal for scoring. Scoring 20 in a 100-90 win is better than scoring 20 for the losers. One is .200 of the opponent total, and the other is .222 .
Basically, every player's Sco rate includes the factor (100/OppPPG).
Yes, a coach should prefer high pace or high efficiency players to match his style. Different players will flourish or languish, depending. David Lee isn't flourishing in GS, for example. Amare' loves NY, etc.
Michael Beasley shoots .530, averaging 22.3 Pts per 36 min. for a team that allows 108.4 ppg.
David West shoots .562, 19.5 Pts/36, for a team giving up 91.4 .
Given that these team PPG allowed are what the NBA at large scores in the games these guys are in, it makes some sense to scale their scoring to these parameters.
This adjustment for West is = 19.5 * 100/91.4 = 21.3 Pts/36/100ppg
For Beasley, its 22.3 * 100/108.4 = 20.6
In the simplest terms, I'd expect West to outscore Beasley in a game between their teams. I'd say he's the better scorer vs anyone.
_________________
`
36% of all statistics are wrong