The Rankings: Split Three ways

Posted on 04/05/2011 by

25



Possessions  = Field Goal Attempts +.44* Free Throw Attempts + Turnovers

Where the hell did the offensive rebounds go?

Settle down people, I’m gonna explain.

I’ve spent a lot of time recently reading. I’ve re-read Basketball on Paper, Freakonomics and I’m reading scorecasting.

I thinks it’s had an effect, I’m trying to look at things a little differently.

So whereas generally Possessions are defined as:

Possessions  = Field Goal Attempts +.44* Free Throw Attempts + Turnovers – Offensive Rebounds

or alternatively

Possessions = 0.5 * ((Tm FGA + 0.4 * Tm FTA – 1.07 * (Tm ORB / (Tm ORB + Opp DRB)) * (Tm FGA – Tm FG) + Tm TOV) + (Opp FGA + 0.4 * Opp FTA – 1.07 * (Opp ORB / (Opp ORB + Tm DRB)) * (Opp FGA – Opp FG) + Opp TOV))

Which people generally use to come up with an Offense Rating and Defensive Rating.

But see, here’s the thing.  Shooting? Defense? Not Rebounding. So I’m using the first formula ( Possessions  = Field Goal Attempts +.44* Free Throw Attempts + Turnovers) because I decided I want to look at three distinct things:

  • How good each team is at getting the ball in the hoop on the offensive end (My version of the O-rating let’s call it the Offensive success rate)
  • How good each team is at keeping the ball out the hoop on the defensive end (My version of the O-rating let’s call it the Defensive success rate
  • How good each team is at getting the ball back after a shot and not giving it back after the opponent shoots. In essence, I want to know how many extra possessions each team generates or gives up. Well call this one the Rebounding Edge.

By treating each possession as ending on a shot or a turnover, I can separate out the effect of a Rebounding edge from how good you are on Defense and on Offense. The critical equations become:

Offensive Success Rate:  Pts per 100 Possessions Offense

Defensive Success Rate: Pts per 100 Possessions Defense

Rebounding Edge: Point Margin from Possesion Generation = Expected Possesion Delta Offense @ 200 Possessions* Offensive Success Rate + Expected Possesion Delta Defense @ 200 Possessions* Defense Success Rate

Point Margin @ 200 Possessions = Off.Success Rate + Defensive Success Rate + Rebounding Edge

Enough talk. Let’s put it on a pretty table:

Lots of fun things to see there. The numbers line up very well to WP48 results. Miami, Chicago, LA and Boston form your top 4. Defense seems to be the key with 8 of the top ten in D showing up in the overall top ten. 7 of the the top ten offenses show up in the top ten. Chicago and Memphis are the only two top rebounding teams that show up in the top ten.

But we know a lot of that is ancient history right?

The Invisible Hand has it's say on the Melo trade!

Let’s look at the last month of the season (Post Trade-deadline):

Denver and Chicago are kicking ass and taking names. Miami, LA and OKC round out that top five. Houston at six looks to be out but the bottom of the West looks downright deadly with Portland and Memphis surging while Dallas and San Antonio plummet. If Houston gets in and Denver stays at 5? Lot’s of potential upsets in the West.

Should the Nuggets fans start thinking parade?

Not quite yet George.

 

We know that the playoff lineups and minutes are not the same as the regular season (30 to 16 to 1 anyone?). Let’s shorten those lineups by looking only at players who plaed 24 or more minutes in a game and look at every team after the deadline:

The data does get a little skewed with the shortening of the lineups (at least for my ratings) but a few things seem clear:

  • Miami, Denver, LA and Chicago are really good.
  • Portland, Memphis and possibly Houston could be bad news for San Antonio and Dallas.
  • Rondo’s doldrums are killing the Celtics.
  • New York is very,very strange. If they play a team that’s bad at rebounding and short in rotation (say Boston with Baby at Center) they could beat them. Miami and Chicago would wreck them.
  • San Antonio has real problems.
  • Denver matches up well with San Antonio and Dallas. Not so well with OKC and LA. They need homecourt against OKC (or a Perk injury) and Bynum to not be healthy to have a really good shot at  Finals . Interestingly, these things are not longshots . Same Bynum logic applies to Portland. (Nuggets/Blazers are +3000 to win West and +6500 to win it all, hmmmmmmmmmmmmm).

Hope that was good for you too.

Weird to see the world differently

 

Advertisements
Posted in: Uncategorized