A New Feature at Basketball-Statistics.com: Charting Defense
As I’ve mentioned before, there are many issues with the current stats out there intended to measure individual defense (including mine). The best way to really analyze a player is to watch the games and record your observations. Here at Basketball-Statistics.com, that’s just what we’re going to try to do.
Schuyler Davis will be tracking and recording the defense of Chicago Bulls’ players for this season. He has a great method of doing this, and hopefully this kind of research catches on elsewhere. All of his work will be available at the main site (http://basketball-statistics.com) in the Articles section. Without any further ado, I present to you Schuyler’s first entry…
Defense: A look into the other side of the game
Kobe Bryant picked up at midcourt by Ron Artest, he moves right off Gasol’s screen, room for Bryant in the lane as he goes up for the throw down! Artest gives up the bucket to Kobe, but is that really fair? What about Yao, should he have hedged the screen? Maybe Artest could have fought through or around it and stayed in front of Bryant. What about after that — shouldn’t Carl Landry have stepped in to contest Bryant’s shot or take a charge? Or was McGrady in position to step in? This is the problem with the defensive metrics available today. They mean relatively little when looking at individual plays, where every defender has a job to do, and one failure leads to the failure of the entire defense.
That is the key. Who made that one mistake or successful play (hopefully the latter in my testing)? Over the course of the rest of this 2008-2009 NBA season I intend to chart the defensive statistics of my hometown Chicago Bulls. No current statistics will be used, just a charting system I developed to give positive and negative credit to the individual defenders of the Chicago Bulls. Six categories will be charted: Pick and/or Roll Coverage, One-on-one, Help Defense, Turnover Attempts, Rebounds, and Leaving Open/Getting Back. Here’s a breakdown:
Pick and/or roll Coverage - For both guards and bigs, how they cover a pick and roll, pick and pop, or running through off-ball screens (i.e. Rip Hamilton stuff).
One-on-one - Just like it sounds, one-on-one situations, only recorded when a player is obviously looking to score, drive, or create somehow.
Help Defense - Extremely important in basketball, but sadly this is rarely ever recorded other than when a blocked shot or charge occurs. Negatives are given for late help, no help, or help when it is not needed. If it is evident there is a designed double team, something the coach put in, then I don’t penalize. Positives are given for good help defense, like getting over in time, and in some situations also recovering to your assignment.
Turnover Attempts - A positive is given for a play leading to a turnover like a tipped ball, you don’t have to be the actual person who gets the steal. A negative is given for a failed attempt, leaving the rest of your defense vulnerable.
Defensive Rebounds - Normal rebounds are not counted, only contested rebounds, when it is clear the offense is crashing the boards. Good box outs, or timing to get the ball are recorded as positives.
Getting back/Leaving open - Mostly negative, not hustling back on D, or a mental lapse (i.e. a backdoor cut) or leaving a player open when there is no reason to do it. Positives are given for great hustle to get back, like a guard hustling back to disrupt a fastbreak, or something in that vain.
Only positive and negative plays are recorded. Also, the end result is not important. I don’t care if Elton Brand drains that jumper, if Noah has a hand right in his face and good position he will get positive credit. Plays I deem as average are not recorded, and as you can imagine several marks can be recorded on each play. The assigning of fault or success can get tough at times, especially when I am trying to see Derrick Rose break Andre Miller’s ankles at the other end of the court. So far I am six games in with this method, and the results are pretty interesting. Drew Gooden can’t play help defense to save his life, and Tyrus Thomas actually has his head on straight on the defensive end. Better, more reliable data will be available after I have charted more games. I find myself really getting into these games as well, much more so than I used to. Hopefully this study can open some insight into different and better ways defense can be evaluated in basketball. I have learned a lot already, and I hope anyone reading this will too. Here is an example of the charting. A couple of things to note: the categories are abbreviated obviously, and the column heading with the actual listing of the category is a list of the positive plays while the heading with negative indicates the number of negative plays.
http://spreadsheets.google.com/ccc?key=pD2-oP8oJwJIHtYg2cVs_Jw&hl=en
Schuyler Davis
Did you enjoy this post? Why not leave a comment below and continue the conversation, or subscribe to my feed and get articles like this delivered automatically to your feed reader.
Comments
I am a freshman in college and have played and watched basketball for all my life. Obviously this is extremely subjective, but that is sort of the idea here considering the goal is to point out inaccuracies in the current methods by taking a different more indepth approach into defense. I don’t mean to be arrogant or anything, but I feel I have a lot of basketball knowledge, especially when breaking down individual plays. So you can just take this with a grain of salt I guess, but I am confident I know what I am doing. Again I don’t want to come off with a bad tone or anything, sorry if it sounded that way.
Just curious what your background is in terms of understanding the difference between good and bad plays. This system seems entirely subjective to me, and even more so if you’re just a weekend warrior/casual fan.