possessions formulas

Moderator: Doctor MJ

lorak
Head Coach
Posts: 6,317
And1: 2,231
Joined: Nov 23, 2009

possessions formulas 

Post#1 » by lorak » Tue Apr 10, 2012 11:09 am

Which one is better (more accurate):
nba.com: ((FGA - O-Rebs) + TO + (.436 x FTA))
or
basketball-reference:
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))

?
Chicago76
Rookie
Posts: 1,134
And1: 228
Joined: Jan 08, 2006

Re: possessions formulas 

Post#2 » by Chicago76 » Wed Apr 11, 2012 6:40 am

When you compare actual tracked possessions courtesy of basketballvalue, the latter form is much closer.

The NBA's version is pretty similar to the initial Basketball on Paper simplified form that Dean Oliver put out years ago (along with the more complex formula he also published in the same book):

FGA-ORB+0.4xFTA+TOV

This form is actually a bit more accurate than the NBA's simplified version too.

FWIW: last year the NBA's possession formula would have overestimated possessions pretty consistently at a rate of roughly 2 per game. The more complex formula on average is +/- 0.3 possessions per game away from the true number.
lorak
Head Coach
Posts: 6,317
And1: 2,231
Joined: Nov 23, 2009

Re: possessions formulas 

Post#3 » by lorak » Thu Apr 12, 2012 11:17 am

Thx Chicago76. That's the answer I was looking for.

Return to Statistical Analysis