Efficiency - reply

[Ryan Falcone <FalconeR@nvg.com>]

>>> Alex Wang wrote:

Seriously Ryan, why don't you tell me what Paul Pierce's FG% is on fast
breaks then? Or what Walker's 3pt FG% from the left side of the court is
when the shot is contested, vs. when he is open?  These are "relevant
statistics" but are certainly not available to anyone but obsessed
coaches like Pitino or video freaks.  Why is FG% used?  Because it's
easily measured <<<

             Per usual, you have misinterpreted my original point, which was
to illustrate the dichotomy between the usefulness of this ostensible
efficiency measurement and the usefulness of data trends that can be
inferred from more commonly utilized basketball statistics.  First of all,
there are plethora of idiosyncratic, not-easily calculated statistics that
provide a tremendous amount of useful data--such as a player's free
throw percentage during the last two minutes of the fourth quarter. 

             Because these statistics are recognized for the insight they
provide about players' performance, they are calculated and analyzed. 
The two percentages you listed above obviously do not fit into that
category, or someone would be inclined to compile data on those
performance indicators to calculate such a rating.

>>> How about this, do you prefer when a player takes 20 three pointers
a game and hits 9, or 20 two pointers a game and hits 10? At the very
least, you want "adjusted field goal percentage" to account for three
pointers weighing three pointers as 1.5*two pointers.  Guess what?  It's
not a "commonly used stat."  That doesn't make it automatically bad.<<<

             Gosh, Alex, either one would be a welcome change from what I
saw offensively from the Celtics this past season!  Point noted, although I
submit that this is perhaps the one example of where an adjusted field
goal percentage is a more useful measurement tool than straight FG%. 
Although there are a select few players in the NBA who shoot an
impressive 40% from three-point range, hitting 9 of 20 three-pointers
(45%) in one game would certainly be an anomaly--far more so than a
player shooting 10 for 20 on two-pointers.

            As such,  I don't think that the above information has much
applicable usefulness for coaches.  With the possible exception of
designing individual set plays, what other possible relevance would
there be for tracking that statistic?  Certainly not to determine lineups, as
many other important factors (rebounding, defense, chemistry) weigh
into that determination than simply the
points-as-a-function-of-shots-taken.  I don't think that calculating
efficiency is "automatically bad" as you have suggested, just that it isn't
particularly relevant / useful for coaches except for designing individually
situational set plays.

>>> I never said it was a magic number that summarizes a person's
offensive ability into one number but I stand by my assertion that it is a
superior measure than FG%.  <<<

             Only if one is interested in determining how many
points-as-a-function-of-shots-taken a player produces.  While this might
be interesting information (to some), I fail to see how this is a superior
measure to FG%.  I doubt that coaches would use this measurement to
fill their rosters (insert vivid recollections of Westhead's Denver Nuggets
team circa 1990-1991 here). 

             Further, if this is not a summarization of players' offensive ability
into a single number, then what is it?  In your earlier example, you
suggested pro-rating three pointers and free throws to reflect their
relative significance on the final calculation.  Does that not mitigate (or in
certain cases, emphasize) the overall significance of those scoring
opportunities, thereby creating a regression-like "score of best fit" to
describe the player's efficiency as a function of offensive ability?

>>> I mean, FG% doesn't even take into account a player getting fouled
and going to the line.  How do you explain how it is a good measure of
efficiency then? <<<

             You're right--but again, this information is only meaningful if
someone (i.e.-coaches) has use for it beyond fans' recreational
interpretation of pseudo-interesting statistics.

>>> This is your refutation of my "insinuation" of hyperbole? Wow.  I'm
not going to continue this type of discussion. <<<

             There was nothing for me to refute.  Let he without sin cast the
first stone, no?  I'm not sure whether to construe your non-dignification
as being tantamount to acknowledgement of my original claim of
hyperbolic hypocrisy.  Jim Meninno admonished me for being to harsh in
my post yesterday, so apologies to both the list and Alex.

Ryan