Comments on: BBR Rankings: 2010-02-05 http://www.basketball-reference.com/blog/?p=4549 NBA & ABA Basketball Statistics & History Mon, 21 Nov 2011 20:56:04 +0000 hourly 1 https://wordpress.org/?v=4.6 By: DSMok1 http://www.basketball-reference.com/blog/?p=4549&cpage=1#comment-14357 Sat, 06 Feb 2010 03:02:17 +0000 http://www.basketball-reference.com/blog/?p=4549#comment-14357 Basically, I took the whole season and ran Solver to minimize the difference between the actual efficiency margins (from the box scores) and the "predicted" margin for each game (adjusted for home/away). Since this is SRS, it was not least squares (I think that would overweight blowouts unnecessarily anyway) but simply minimizing the sum of the differences.

It gives a lot of useful information, I think... I've run it for the last 20 seasons, too. I like such a method as the purest form of an adjusted efficiency differential.

]]> By: Neil Paine http://www.basketball-reference.com/blog/?p=4549&cpage=1#comment-14353 Fri, 05 Feb 2010 17:53:19 +0000 http://www.basketball-reference.com/blog/?p=4549#comment-14353 Oh, that's because our game-by-game team stats don't have "Team Turnovers" (shot clock/5-second inbound/8 second halfcourt violations/etc.) included in the team total.

As an aside, for your SRS, did you base it on the average of game-by-game efficiency differentials?

]]> By: DSMok1 http://www.basketball-reference.com/blog/?p=4549&cpage=1#comment-14352 Fri, 05 Feb 2010 16:07:33 +0000 http://www.basketball-reference.com/blog/?p=4549#comment-14352 An alternative ranking system: SRS of efficiencies (compiled off 4-factor box scores on this site--thanks!)
.
Home Court Advantage: 3.54
.
Team Margin/100 Poss
CLE 7.51
UTA 6.75
LAL 6.50
DEN 5.99
ATL 5.41
ORL 5.25
OKC 4.72
SAS 4.09
PHO 3.61
POR 3.49
BOS 3.05
DAL 1.08
HOU 0.75
TOR 0.03
MEM -0.07
MIA -0.49
MIL -0.59
NOH -1.32
CHA -1.55
CHI -2.02
SAC -3.12
LAC -3.31
GSW -3.79
PHI -4.25
DET -4.75
IND -4.78
NYK -5.42
WAS -5.61
MIN -7.06
NJN -12.18
.
As a side note: why, when I calculate the league average efficiency (by averaging the 4 factor box scores) do I come up with a different value than that shown in the league-wide season box score?

]]>