Monday , Jan , 05 , 2009 C.Y. Ellis

The Ultimate Stat Equivalator

How many points is a steal worth? How many rebounds equals a blocked shot? In normal basketball analysis, these questions are absurd. People would mock you for asking them, and rightly so. “You fool,” they would sneer. “You can’t equate different kinds of stats.” Then, for emphasis, they would make an inflammatory hand gesture, or maybe spit on your shoes. For some reason, the people I am thinking of are Marco Belinelli’s extended family.

The Ultimate Stat Equivalator



How many points is a steal worth?  How many rebounds equals a blocked shot?  In normal basketball analysis, these questions are absurd.  People would mock you for asking them, and rightly so.  “You fool,” they would sneer.  “You can’t equate different kinds of stats.”  Then, for emphasis, they would make an inflammatory hand gesture, or maybe spit on your shoes.  For some reason, the people I am thinking of are Marco Belinelli’s extended family.

Anyway, in normal basketball analysis, the practice of equating stats to each other is nonsensical.  But in fantasy basketball?  To quote Carl Weathers: baby, you got a stew goin’

Today I am unveiling the Ultimate Stat Equivalator (Linking Equivalated Stats in Summation), a.k.a. USE(LESS), an algorithm that determines, in fantasy hoops, exactly how many points a steal is worth, and how many three-pointers equals a rebound, and how many turnovers equals Shaq’s free-throw percentage, and so on.  Don’t be daunted by these technical-sounding words: the algorithm here is uncomplicated, and “equivalator” is simply an awesome-sounding word I made up.  Later in this column, the stats are used to give the Top 25 USE(LESS) Fantasy Rankings, a table that will be updated with every subsequent column by me.

The premise of USE(LESS) is actually quite simple: I took the total points scored this season by every player, and divided it by the season’s total rebounds, assists, steals, blocks, three-pointers, and turnovers.  Then, all you do is multiply  a player’s per-game averages by each of those, and you get the adjusted USE(LESS) value of their rebounds per game, assists per game, etc.  For example, in the case of steals, there have been 94813 points scored in the NBA as of last night (Jan. 4), and 7380 steals.  Thus, when Rajon Rondo picks Chris Duhon’s pocket, it has the same fantasy value as 13.335 points—and on Duhon’s side, because there have been 13865 turnovers in the league so far, he loses 7.098 USE(LESS) points on the exchange.

Two points: one, you should note that the premise is one of scarcity.  The rarer a category is—like blocks and steals—the more valuable each instance is.  This should seem intuitive—Jose Calderon’s 8.7 assists per game are way more valuable than his 13.1 points, because assists are scarcer than points.  Here, for the season so far, is the USE(LESS) conversion table:*

 

points

rebounds

assists

blocks

steals

3pointers

turnovers

total

98413

41590

20699

4972

7380

6434

13865

per point

1

2.366

4.754

19.793

13.335

15.296

7.098

Point number two: the formula for FG% and FT% is more complicated, because it takes into account the number of attempts as well as the percentage itself.  So we’ll save that for another column.  You have enough excitement in your life right now as it is.

Here, then, are the USE(LESS) Fantasy Power Rankings for the season so far.  I took each USE(LESS) adjusted stat and just totaled them up.  Remember, all per-game averages below are adjusted.  (The stats, I should also note, are courtesy of the superb site dougstats.com.) Feast your eyes on this:

*rank

Player

*ppg

*rpg

*apg

*spg

*bpg

*3pg

*topg

*fg%

*ft%

*total

1

james,lebron

27.5

15.8

31.7

25.5

24.4

19.6

-20.4

29.6

2.0

155.7

2

paul,chris

20.1

12.8

54.8

38.2

3.3

10.8

-21.3

18.7

14.2

151.6

3

wade,dwyane

28.8

12.3

33.2

29.6

31.3

10.6

-26.8

17.2

-2.9

133.3

4

camby,marcus

12.0

31.9

10.8

13.8

54.8

0.0

-9.8

16.5

-1.5

128.5

5

granger,danny

25.1

12.2

16.0

12.4

28.1

38.5

-21.5

1.1

11.4

123.3

6

stoudemire,amare

21.8

20.2

10.4

14.6

21.5

1.5

-19.9

41.4

8.9

120.5

7

nowitzki,dirk

26.0

20.5

12.5

9.6

16.5

14.9

-14.2

12.8

17.5

116.1

8

garnett,kevin

16.1

20.7

12.5

16.5

28.3

0.4

-13.6

30.0

4.1

115.1

9

bryant,kobe

26.8

12.6

19.8

19.6

7.4

16.8

-18.6

15.7

14.7

114.7

10

hilario,nene

14.7

18.7

6.9

17.7

28.8

0.4

-12.9

43.3

-3.3

114.3

11

howard,dwight

20.1

32.2

6.8

11.3

68.0

0.0

-18.4

37.9

-45.5

112.5

12

jefferson,al

22.8

24.6

8.1

8.1

34.5

0.0

-13.3

25.1

0.2

110.1

13

ming,yao

19.8

23.2

7.4

5.1

33.5

0.0

-20.9

27.0

15.1

110.1

14

gasol,pau

17.6

21.0

17.7

5.6

20.2

0.5

-13.7

37.2

3.5

109.5

15

nelson,jameer

16.4

8.4

24.3

16.5

0.0

28.1

-14.2

25.2

3.8

108.7

16

calderon,jose

13.1

8.0

41.5

12.0

0.6

24.3

-15.3

11.0

11.1

106.4

17

kaman,chris

13.9

22.2

8.9

9.8

36.6

0.0

-19.4

35.2

-1.0

106.2

18

lewis,rashard

19.6

14.6

12.3

17.7

11.0

46.6

-15.4

-6.1

6.0

106.2

19

johnson,joe

22.8

11.5

28.7

14.9

7.7

34.0

-17.4

-1.6

5.0

105.7

20

duncan,tim

20.9

24.4

15.6

6.5

33.9

0.0

-15.5

30.7

-11.0

105.4

21

boozer,carlos

20.5

27.6

12.7

12.2

4.9

0.0

-14.2

45.3

-4.7

104.4

22

kidd,jason

8.6

15.2

40.4

33.5

8.9

22.8

-16.8

-8.4

0.0

104.3

23

terry,jason

21.0

6.2

17.3

17.0

7.1

34.0

-13.3

8.2

6.7

104.3

24

bibby,mike

16.2

8.6

25.0

16.6

2.4

40.1

-10.5

4.5

1.4

104.1

25

roy,brandon

23.0

10.6

25.0

13.8

7.6

15.4

-14.2

10.2

10.8

102.2

First of all, not to toot my own horn,** but this is a good-looking list.  LeBron is first, Chris Paul is second, and they’re close to each other and far from the rest.  Marcus Camby’s and Dwyane Wade’s monster seasons are also acknowledged as such.  Danny Granger at #5 makes perect sense to me—he contributes in all categories and has excellent percentages.  Chris Kaman is on this list because his per-game averages are good; the list ignores that he has played fifteen games.  Same with Carlos Boozer.  Note that the top five point guards are: Chris Paul, Jameer Nelson, Jose Calderon, Jason Kidd, and Mike Bibby.  Meditate on this list, and repent.

Finally, just for fun, the top ten players over the last ten games—not the last season—are these:

*rank

Player

Team

PS

*total

1

james,lebron

cle

SF

178.8

2

roy,brandon

por

SG

165.0

3

granger,danny

ind

SF

149.4

4

bryant,kobe

lal

SG

143.8

5

nelson,jameer

orl

PG

139.5

6

paul,chris

nor

PG

136.6

7

millsap,paul

uta

PF

136.0

8

o’neal,jermaine

tor

C

135.8

9

wallace,gerald

cha

SF

131.8

10

lewis,rashard

orl

PF

131.3

Yes, injured Jermaine O’Neal, Paul Millsap, and Brandon Roy are on this list.  Hmmm.  Perhaps USE(LESS) needs some tweaking.

*UNBELIEVABLY NERDY FOOTNOTE ALERT—the rate of conversion here might remind you of physics class, and not the part where Big Mike kept hitting you in the head until you let him copy your homework.  No, this should be reminiscent of unit standardization: each stat, after adjustment, ends up in the unit of points per game.  For example:

5 rebounds/game * (2.366 points/rebound) = 11.83 points*rebound/game*rebound = 11.83 points/game

You will remember that the operation of addition, in physics, only works if the quantities have the same units, and furthermOW BIG MIKE PLEASE STOP HITTING ME IN THE HEAD.

**This euphemism always sounds unsavory to me.  I regret having used it.