Some player analysis using several metrics
Posted: Mon Jul 18, 2011 4:09 am
Using the side by side data is this article
http://www.basketball-reference.com/blog/?p=9786
I made a few calculations for the PGs who played over 2000 minutes.
RAPM-SPM, which probably is pretty good at capturing the amount of what PGs do to help or hurt their team that does not show up in boxscore metrics.
(player WSper48 / this group's avg WSper48) / (player SPM / group avg SPM), which probably can help identify players who got notable credit in WSper48 for defense beyond their individual boxscore data vs those lagging in this regard. (Offensive and defensive RAPM can do this too, but this measure is a non-Adjusted +/- alternative.)
(player PER / this group's avg PER) / (player SPM / this group's avg SPM), which may help identify players who are getting a relatively high level of their value from usage.
The leader on RAPM-SPM is Steve Nash at +5.2 and more than twice the level of Conley, Lowry and Fisher. Kidd is only other PG in this group of 32 to contribute more than +1 on RAPM-SPM. Just 5 guys over +1 on this metric difference of 32. Almost 2/3rds of the group are negative with Derrick Rose having the most negative score at -3.4. (Sometimes personal heroics may not be enough and / or have hidden costs.)
The leader on (player WSper48 / this group's avg WSper48) / (player SPM / group avg SPM) is Udrih followed by Holiday, Conley and Lawson. Ridnour is a far far negative outlier with only Collison and Augustin halfway to him.
The leaders on (player PER / this group's avg PER) / (player SPM / this group's avg SPM) in descending order are Udrih, Holiday and Conley. Ridnour is again the farthest out negative score but closely followed this time by Collison and Augustin again third most negative.
I might writeup some other position results later.
http://www.basketball-reference.com/blog/?p=9786
I made a few calculations for the PGs who played over 2000 minutes.
RAPM-SPM, which probably is pretty good at capturing the amount of what PGs do to help or hurt their team that does not show up in boxscore metrics.
(player WSper48 / this group's avg WSper48) / (player SPM / group avg SPM), which probably can help identify players who got notable credit in WSper48 for defense beyond their individual boxscore data vs those lagging in this regard. (Offensive and defensive RAPM can do this too, but this measure is a non-Adjusted +/- alternative.)
(player PER / this group's avg PER) / (player SPM / this group's avg SPM), which may help identify players who are getting a relatively high level of their value from usage.
The leader on RAPM-SPM is Steve Nash at +5.2 and more than twice the level of Conley, Lowry and Fisher. Kidd is only other PG in this group of 32 to contribute more than +1 on RAPM-SPM. Just 5 guys over +1 on this metric difference of 32. Almost 2/3rds of the group are negative with Derrick Rose having the most negative score at -3.4. (Sometimes personal heroics may not be enough and / or have hidden costs.)
The leader on (player WSper48 / this group's avg WSper48) / (player SPM / group avg SPM) is Udrih followed by Holiday, Conley and Lawson. Ridnour is a far far negative outlier with only Collison and Augustin halfway to him.
The leaders on (player PER / this group's avg PER) / (player SPM / this group's avg SPM) in descending order are Udrih, Holiday and Conley. Ridnour is again the farthest out negative score but closely followed this time by Collison and Augustin again third most negative.
I might writeup some other position results later.