I am curious as to some more advanced basketball stats, along the same line as to what sabermetrics is to baseball.  I keep a spreadsheet which tracks all the counting stats (points, reb, etc...) per minute played, but I would like to expand this a little bit.

My spreadsheet also calculates efg% and efficiency, but I'm not sold on those (particularly efficiency).  The rudimentary calculation treats points, rebounds, assists, steals, and blocks as one in the same and I don't really trust it.  Optimally I would like to calculate a Player Efficiency Rating, but I don't think we have access to some of the information needed for that calculation.

Is there anything else out there which may be a happy medium between EFF and PER?
4/18/2013 11:57 AM
out there? not that I know of.

Have i tried to make my own? absolutely. the data just isn't there are first glance and I don't have the computer programming acumen to develop something that pulls from the Play-by-play new stats.

well, i could do it but i'd have to quit my day job.
4/18/2013 12:29 PM
There are simplified versions of the Player Efficiency Rating that you should be able to use with game log data for each player. I've recently started tinkering with the formula described here: http://bleacherreport.com/articles/113144-cracking-the-code-how-to-calculate-hollingers-per-without-all-the-mess. Haven't spent enough time with it to be determine whether any adjustments would be needed for use in WIS.
4/18/2013 12:31 PM
So I tend to like TS% (which basically takes eFG% and factors in production at the line). Only problem is that it doesn't take turnovers into account.

On the defense/rebounding side, I'm not sure where to go
4/18/2013 12:31 PM
i've manually calculated +/- for my players during the course of a game a few times. not much else to do for defense.
4/18/2013 12:35 PM
Posted by jetwildcat on 4/18/2013 12:35:00 PM (view original):
i've manually calculated +/- for my players during the course of a game a few times. not much else to do for defense.
Yea defense is the main thing I am trying to quantify a little bit.  Optimally I want to know how much defense I can sacrifice for the sake of offensive production.
4/18/2013 12:50 PM
We need a +/- like hockey has to see if a team gains or loses points with a player in the game. It wont be perfect, what with assists, def and all, but it would help over the long haul.
4/18/2013 3:20 PM
Posted by mroylanc on 4/18/2013 12:31:00 PM (view original):
There are simplified versions of the Player Efficiency Rating that you should be able to use with game log data for each player. I've recently started tinkering with the formula described here: http://bleacherreport.com/articles/113144-cracking-the-code-how-to-calculate-hollingers-per-without-all-the-mess. Haven't spent enough time with it to be determine whether any adjustments would be needed for use in WIS.
Thank you, that is basically exactly what I was looking for.  Going to play around with that formula and see how it applies to HD.
4/18/2013 3:55 PM
I'm thinking through the value of this stat (Player Efficiency Rating) and am curious what others think. If I understand the formula correctly, it's telling you who you want to have in the game (because good things happen when they are on the court). So that might be helpful in setting your depth chart and fatigue level, but it wouldn't necessarily help with distro. The player might get a high rating due to assists, rebounds, etc. and you would want them on the court but you wouldn't necessarily want them taking the shots. Is that interpretation correct?
4/18/2013 10:43 PM
Posted by mroylanc on 4/18/2013 10:43:00 PM (view original):
I'm thinking through the value of this stat (Player Efficiency Rating) and am curious what others think. If I understand the formula correctly, it's telling you who you want to have in the game (because good things happen when they are on the court). So that might be helpful in setting your depth chart and fatigue level, but it wouldn't necessarily help with distro. The player might get a high rating due to assists, rebounds, etc. and you would want them on the court but you wouldn't necessarily want them taking the shots. Is that interpretation correct?
I think you're right on.  PER would be more useful in figuring out who should be on the court most frequently, thus being helpful for setting a depth chart.  As for distribution, I would look more closely at things like eFG% of TS%.
4/18/2013 11:42 PM