Post by eric on Mar 12, 2021 12:57:47 GMT -5
There is a wild controversy about the basketball statistic PER. Critics claim anyone shooting more than 30% from the field will see higher PER just by taking more shots, and since 30 FG% shots are trashola that's not a good break point. Proponents claim the mechanics of the stat later compare to league average so the apparent 30% breakpoint is an illusion. Please note! This is not an argument about whether shot takers are good, whether players who can get their own low percentage look whenever they want are valuable despite the low percentage. This is an argument only about whether the PER statistic returns a higher value with such behavior as an input.
So let's test it, shall we? Currently Billy Williams for the Lakers has the following per 36 stat line:
More specifically, he has a uPER of .215 which when divided into the league average uPER of .359 and multiplied by 15 gives 9.0. O.K., if we double Billy Williams' FG and FGA from 205 and 487 to 410 and 974 respectively, he therefore goes from 511 to 921 PTS, and instead has the stat line:
Again, more specifically he has a uPER of .283, divide into the league average of .360 (note that his coming up necessarily increases the league average very slightly), *15 and there's our 11.8. Let's double his FG and FGA again, so he's 820 of 1948 and scores 1741 PTS:
A garbage shooter gets above league average PER just by shooting more garbage shots.
.
Now, Billy's PTS/TSA keep going down because he's not that bad from the free throw or three point lines. This actually makes the case worse for PER but just to put everyone's mind at ease let's redo the first and third examples with no free throws or threes at all. Billy "All Twos" Williams then goes:
The efficiency in Player Efficiency Rating refers only to production per time, not per shot. As such, comparing it to shots taken (for example USG) is required to get accurate reads on players with very disparate shot totals. It is an open source computation and really not even that complicated, it's all just arithmetic.
The truth is out there.
So let's test it, shall we? Currently Billy Williams for the Lakers has the following per 36 stat line:
234 Lakers Billy Williams 1 SG 60 34.6 8.9 5.1 2.7 1.2 0.2 1.5 .421 .775 .407 .961 9.0More specifically, he has a uPER of .215 which when divided into the league average uPER of .359 and multiplied by 15 gives 9.0. O.K., if we double Billy Williams' FG and FGA from 205 and 487 to 410 and 974 respectively, he therefore goes from 511 to 921 PTS, and instead has the stat line:
234 Lakers Billy Williams 1 SG 60 34.6 16.0 5.1 2.7 1.2 0.2 1.5 .421 .775 .407 .904 11.8Again, more specifically he has a uPER of .283, divide into the league average of .360 (note that his coming up necessarily increases the league average very slightly), *15 and there's our 11.8. Let's double his FG and FGA again, so he's 820 of 1948 and scores 1741 PTS:
234 Lakers Billy Williams 1 SG 60 34.6 30.2 5.1 2.7 1.2 0.2 1.5 .421 .775 .407 .874 17.4A garbage shooter gets above league average PER just by shooting more garbage shots.
.
Now, Billy's PTS/TSA keep going down because he's not that bad from the free throw or three point lines. This actually makes the case worse for PER but just to put everyone's mind at ease let's redo the first and third examples with no free throws or threes at all. Billy "All Twos" Williams then goes:
234 Lakers Billy Williams 1 SG 60 34.6 7.1 5.1 2.7 1.2 0.2 1.5 .421 .000 .000 .842 7.3
234 Lakers Billy Williams 1 SG 60 34.6 28.5 5.1 2.7 1.2 0.2 1.5 .421 .000 .000 .842 15.7The efficiency in Player Efficiency Rating refers only to production per time, not per shot. As such, comparing it to shots taken (for example USG) is required to get accurate reads on players with very disparate shot totals. It is an open source computation and really not even that complicated, it's all just arithmetic.
The truth is out there.
