Saturday, January 3, 2009

Converting off of Defensive Rebounds, Part II


So, a little while back I put up a post looking at the success of teams after defensive rebounds. In the comments to that post, reader Marshmallow Lips gave the helpful suggestion to present the data by position. So, below, I've split out the data for the 2007-2008 season by position, by team, and by individual player (separated out on various tabs based on position).

The positional classifications here come from the player data at dougstats, while the rest of the data was derived from play by play data that I downloaded from basketballvalue. I realize that positions aren't always fixed (for instance, in this data Monta Ellis is considered a point guard and Tim Duncan, despite his protestations, is a center), but hopefully in general the classifications will make sense. 

For the individual player tabs, I've added a color-coded column that looks at whether or not the individual's defensive rebounds resulted in the team's creating offense more or less quickly than what that team's average is. Green means faster (ie, a negative difference between the player's average and the team's average), and red means slower. 

As a reminder, I've included only defensive rebounds that immediately resulted in a shot attempt, a foul, or a turnover. I've removed those defensive rebounds that were followed by a timeout, jump ball, end of quarter, kicked ball or deflected out of bounds, and so forth. Also, the "success rate" metric does not count missed shots that resulted in offensive rebounds as "successes" -- the idea is to look at the success rate of the play immediately after the defensive rebound. Note also that the success rate is not a measure of points produced -- all drawn fouls are treated equally regardless of whether they resulted in made free throws, and made three pointers aren't given any extra credit.

For what it's worth, the individual player breakdown is separated out by team, so players who were traded midseason will have multiple entries. It might be interesting to compare the numbers for one player with different teams . . ..

The positional breakdown, I think, highlights some of the added value of a point guard who can rebound -- notice not only the higher success rate but the faster turnaround time (the "average time" column measures number of seconds from the rebound to the offensive play). Also note that, even here, Jason Kidd, Kobe Bryant, Lebron James, Dirk Nowitzki, and Tim Duncan show up as extremely valuable. Let me know in the comments if you see any other interesting results or areas for further research. 



(On a mostly unrelated note -- I am writing this while watching the Nuggets play the Thunder, and keep wondering about how to quanitfy the value of the sorts of offensive rebounds that Russell Westbrook retrieves. Nene and Carmelo Anthony have been prolific in terms of rebounding their own misses because the move that results in their shot attempt puts them in rebounding position, but Westbrook comes out of nowhere to extend Thunder possessions that, without his presence, ought to have been over, rebounding misses by centers when most point guards would have been retreating to prevent a fast break at the other end. It's like he's committing the opposite of a turnover. Hmmn . . .). 

No comments:

Post a Comment