Comments on: BBR Rankings: Schedule-Adjusted Offensive and Defensive Ratings (December 24, 2010)

By: Ben

Ben — Fri, 31 Dec 2010 16:40:37 +0000

D.S. Mok, how big is the HCA and rest day adjustment in your rankings?

By: DSMok1

DSMok1 — Fri, 31 Dec 2010 02:35:11 +0000

It's similar to what Neil is doing here, not SRS (which is based on points rather than efficiency). But the effect is the same. I also included rest compensation.

The blue line does represent trend.

By: cbobcat3

cbobcat3 — Thu, 30 Dec 2010 23:00:32 +0000

Thanks DSMok1...does the adj eff relate to SRS? What does the blue line represent...trend?

By: DSMok1

DSMok1 — Thu, 30 Dec 2010 14:59:49 +0000

@ Cbobcat3

Here's something similar to what you'd like: it's Charlotte's performances in each game, adjusted for opponent, rest, and location. Whether the game was a win or not is also indicated with the darker bars: http://picasaweb.google.com/118396169256157940063/APBRCharts#5556489894993116242

By: cbobcat3

cbobcat3 — Wed, 29 Dec 2010 23:22:59 +0000

Is there a place/way to go and look at how a team's SRS has changed after each game? For instance...the Bobcats SRS after their 1st game, 10th game, etc?

By: kelfala m

kelfala m — Sun, 26 Dec 2010 22:19:46 +0000

It's great to see the miami heat on the rise they have too much talent to fail. Boston is doing great as well. I wonder who would be on top after the season is over.

By: dsong

dsong — Sun, 26 Dec 2010 01:00:31 +0000

Well, at least the Lakers are improving; a 16-point thumping at home vs. the Miami Heat after a 19-point shellacking at home vs. the Milwaukee Bucks.

I guess the Lakers are pulling their usual Randy Moss act and play when they feel like it. Kinda frustrating to see them treat every game like a preseason game until the playoffs, but hey, as long as they win championships they'll be forgiven.

By: danny95207

danny95207 — Fri, 24 Dec 2010 22:44:59 +0000

Hope Sacramento will get better ranking

By: DSMok1

DSMok1 — Fri, 24 Dec 2010 18:58:53 +0000

I think your rankings will be more indicative of overall team strength if you also minimize the errors of the overall margin. You might look into that.

By: Neil Paine

Neil Paine — Fri, 24 Dec 2010 18:52:48 +0000

I'm just minimizing the sum of the squared errors for each efficiency. So the error term for each game is:

(actual home rtg - home rtg predicted from SRS)^2 + (actual away rtg - away rtg predicted from SRS)^2

Then I minimize the average error per game, weighted by the # of possessions in each game. You have to do it that way to get the numbers to be centered around the actual league average pts/100 poss.