APBRmetrics

KD · Joined: 30 Jan 2005 Posts: 163

Exactly what you'd expect: entertaining, insightful, not much that hasn't already been discussed at length here, but well worth sending along to other hoop fans.

http://www.nytimes.com/2005/04/10/sports/basketball/10score.html?

Eli W · Joined: 01 Feb 2005 Posts: 402

Very interesting. Does anyone have a current season ranking of adjusted plus/minus? If Dan's too busy, is it possible to replicate the rankings using data from 82games.com?

kjb · Joined: 03 Jan 2005 Posts: 865 Location: Washington, DC

Unfortunately, NY Times didn't put online Dan's list of the top most underrated and overrated players. Dan, please post when you get a chance.

KD · Joined: 30 Jan 2005 Posts: 163

kjb · Joined: 03 Jan 2005 Posts: 865 Location: Washington, DC

Dan Rosenbaum · Posted: Mon Apr 11, 2005 12:52 pm Post subject:

One thing to note about the underrated/overrated list is that it equals:

Adjusted plus/minus rating per 40 minutes

minus

Glory statistics (points plus rebounds plus assists) per game

Note that the glory statistics are normalized to have the same mean and standard deviation as the adjusted plus/minus statistics, so that subtracting one from the other makes sense.

Note that adjusted plus/minus is a per 40 minutes measure and the glory statistics are per game. Thus, the underrated/overated measure is a combination of differences in productivity and in playing time.

This may seem like an apples and oranges comparison, but it appears that salaries are a function of per game measures, whereas productivity I think is best measured per 40 minutes. So I think this is an alright comparison. Overrated/underrated lists are always very difficult to come up with, because it is difficult to come up with how a player is "rated."

HoopStudies · Posted: Mon Apr 11, 2005 1:23 pm Post subject:

kjb · Joined: 03 Jan 2005 Posts: 865 Location: Washington, DC

Had a pure fan type question that maybe Bob Chaikin could best answer.

What happens if the 10 players on the underrated list play against the 10 players on the overrated list?

What happens if both teams were entered into this year's NBA and played a full schedule?

bchaikin · Joined: 27 Jan 2005 Posts: 689 Location: cleveland, ohio

Had a pure fan type question that maybe Bob Chaikin could best answer.

What happens if the 10 players on the underrated list play against the 10 players on the overrated list?

What happens if both teams were entered into this year's NBA and played a full schedule?

give me the minutes per game you want each player to play (in multiples of 4 minutes, i.e. 4, 8, 12, 16, 20, 24, 28, 32, 36, 40 min/g), make sure each team adds up to 240, and i'll run the simulation...

gabefarkas · Joined: 31 Dec 2004 Posts: 1313 Location: Durham, NC

gabefarkas · Joined: 31 Dec 2004 Posts: 1313 Location: Durham, NC

kjb · Joined: 03 Jan 2005 Posts: 865 Location: Washington, DC

gabe's minutes distribution looks fine to me.

Dan Rosenbaum · Posted: Tue Apr 12, 2005 10:54 am Post subject:

gabefarkas · Joined: 31 Dec 2004 Posts: 1313 Location: Durham, NC

gotcha. i'm guessing that you did the sum of Pts+Rbs+Ast and then normalized?

also, i'm guessing that taking Glory/game vs Glory/40 wouldn't affect the outcome, since one is basically a ratio of the other?

bchaikin · Joined: 27 Jan 2005 Posts: 689 Location: cleveland, ohio

here's the simulation for dan's most underrated versus overrated players, each team playing 8200 games versus an equal number of games (or close to equal) against the other NBA teams:

underated players
W 61 L 21 per average 82 games
min def
24 +2.0% n.collison (24 C) 9 pts/g
40 -5.0% a.kirilenko (28 PF, 12 SF) 21 pts/g, 317 BS, 347 ast
24 -3.0% r.horry (16 C, 8 PF) 9 pts/g
36 -4.0% m.ginobili (8 SF, 16 SG, 12 PG) 22 pts/g, 440 ast
36 -2.0% b.barry (36 PG) 14 pts/g, 333 ast
4 +1.5% m.ruffin (4 C) 1 pt/g
28 -0.5% b.cardinal (4 C, 12 PF, 12 SF) 12 pts/g, 211 ast
12 +2.0% g.buckner (12 SF) 4 pts/g
24 -1.5% f.hoiberg (24 SG) 10 pts/g
12 -2.5% sh.anderson (4 SF, 8 SG) 4 pts/g

as you can see having kirilenko for 40 min/g makes a huge difference...

off FG% - .459
def FG% - .408

stat/team/opp

ST 10.0/7.5
TO 13.6/16.0
FTM 28/20
FTA 20/26
reb 42/41
ast 23/20
BS 7.4/4.6

overrated players
W 44 L 38 per average 82 games
min def
40 -4.0% j.o'neal (40 C) 18 pts/g, 184 BS
32 +2.5% z.randolph (32 PF) 12 pts/g
32 -2.5% r.jefferson (24 SF, 8 SG, 4 PG) 11 pts/g
28 +2.5% m.redd (24 SG, 4 PG) 10 pts/g
40 -3.0% a.iverson (40 PG) 20 pts/g, 460 ast
20 +3.0% t.murphy (8 C, 12 PF) 7 pts/g
24 +3.5% a.walker (4 PF, 20 SF) 8 pts/g
8 +4.5% a.harrington (4 SF, 4 SG) 3 pts/g
12 +2.5% c.anthony (12 SG) 5 pts/g
4 +2.5% l.nailon (4 SG) 1 pt/g

i had to play some players out of position, like c.anthony at SG, and jefferson and redd at PG for short stretches...

off FG% - .430
def FG% - .448

stat/team/opp
ftm 25/15
fta 32/20
reb 44/41
ast 15/22
st 7/6
to 12.8/13.3
bs 4.1/4.8