Wednesday, November 25, 2009

New Basketball statistic?

I've never seen this stat used before by espn or anyone... I refer to this as: "Points-per-PT Attempted."


Total Points scored
_________________

Total Points attempted
[2pt FGA (x2) +3ptAtt (x3) + FT attempts (x1)]


I calculated this statistic for some NBA players from the 2008-09 NBA season. Here is how it ended up:


Shaq ------------------------ 60.6 %
Pau Gasol -------------------60.4 %
Andrew Bynum -------------58.6 %
Dwyane Wade -------------- 52.7 %
LeBron James -------------- 52.7 %
Kobe Bryant ----------------50.9 %
Artest ----------------------- 43.4 %


That's just a small pool. But I like this statistic more than I do the traditional "FG%"
All "FG%" is, is:



2pt FGM + 3pt FGM
_______________

2pt FGA + 3pt FGA


It doesn't take Free Throws into account at all. It does not benefit slashers who get to the FT line for easy points. Nor does it benefit 3 point shooters (who are attempting more points per shot attempt than most banger/C/PF players are taking).

My stat (pts/pts attempted) takes into account the amount of points attempted on FT (1 point apiece) and 3-point attempts (3 points apiece).


So not only does this benefit good 3 point shooters and good FT shooters, but it also punishes the prototypical "Antoine Walker" type of player. Those who miss easy free throws (i.e. Ben Wallace, Shaq, Dwight Howard, Biedrins, Odom) would be taken into account. And those players who take too many 3 point attempts (i.e. Rasheed Wallace, Jamal Crawford, Chris Duhon) would be punished for attempting too many 3PA for the relatively small amount they make by comparison.

This stat is broad yet inclusive. It directly measures scoring/shot efficiency. It takes into account 2FGM, 3FGM, FTM, FTA, 2PA, and 3PA, and the widely used FG%.

The essential difference between "FG%" and "Pts/PtAttempted" is that "pts/ptatt"incorporates FT%, 2ptFG%, and 3ptFG% into "FG%."

Points per Field Goal Attempt would be a great stat to determine scoring efficiency in terms of on-floor shot selection (without involving FTs). Or you could use "points per PA" and subtract points scored with FTs off the numerator and FTA off the bottom of the formula.


So let's say Kobe Bryant has a game where he hits 6/14 2-pointers, 2/6 3-pointers, and 8/9 FTs. His stats in each category would be as follows:

"points-per-PA" ----47%
2 pointers: 12/28
3 pointers: 6/18
1 pointers: 8/9
total points scored: 26 divided-by 55 (total points attempted)

"points-per FGA" ------90% (300% maximum possible)
2 pointers: 12 points
3 pointers: 6 points
total points scored from 2's and 3's: 18 divided-by 20 (total number of 2 and 3 pt FGA)

"points-per-PA (not inclusive of FTs)" -----39%
2 pointers: 12/28
3 pointers: 6/18
total points scored from 2's and 3's: 18 divided-by 46 (number of points attempted from 2's and 3's)

No comments: