Efficient Shooting Percentage

Basketball is arguably the most difficult sport of the Big 3 to develop complete statistics.  Baseball works on a play-by-play basis meaning that the only aspect of the game that is difficult to capture is defensive ability.  Football revolves around ball movement on a play-by-play basis so the only gap involves the off-ball interactions.  Unlike its peers, basketball involves fluid gameplay.  Inherently, a lack of separation between actions increases the difficulty of effectively capturing the quality of play with statistics.

In sports, nearly all statistics receive criticism.  More than any other sport, basketball statistics incite the most debate.  This is probably due to a number of factors: there are a high number of separate aspects of gameplay, different aspects can be difficult to differentiate at times, not all aspects of the game can be measured, how should different aspects be weighted against one another, and many more.  While large-scale statistics receive the most argumentation, shooting statistics are often the most misused.  The validity of standard shooting percentages is non-debatable, but the effectiveness of measuring players by them certainly is.  With increasingly developed shooting statistics comes decreased consensus over effectiveness.

While many different statistics exist, all seeming to take different approaches at assessing offensive efficiency, none attempt to assess offensive intelligence.  I decided to attempt to fill this gap.  It is nearly impossible to measure true offensive intelligence—can't measure whether a player "makes the right play or not"—but it is certainly possible to measure whether a player sets himself up to be successful.  A player who effectively sets himself up to succeed would be expect to shoot the most from the places where he shoots the best.  Using this logic, to measure offensive efficiency I used shooting percentages and percentage of shots taken from certain distances to calculate a total efficiency value.

First, I used hoopdata.com to break down a player's overall field goal percentage into selected shooting percentages, at the rim, in the paint (<10 feet), mid-range (10-15 feet), long 2-pointers (16-23 feet), and 3-point field goals (I used effective 3-point field goal percentage to add weight for difficulty).  I then calculated the percentage of total shots taken in each of the regions.  These two break downs told me how good a player was from a certain area and how often they shot from that area.  Next, I multiplied the two percentages to obtain a "efficiency value" for each area of the court.  I knew that it was efficient for a good free-throw shooter to get to the free-throw line more often so I then multiplied free throw percentage by total FT/FG to obtain an "efficiency value" for getting to the line.  Finally, I added the respective "values" to create an "Efficient Shooting Percentage."

As a sample set, I calculated the Efficient Shooting Percentage (ES%) for 40 of the NBA's top players using percentages at the time of the All-Star break.  Here is what I found (sorted by ES%):

While the results exhibit few major surprises, there are also a few enlightenments (See Eric Gordon and Derrick Rose).

What do you think?