Just sent this to a user, thought I'd post it, I think this is fairly correct, but would encourage anyone who sees it differently to chime in:

not sure if this will help or not, but lets say I have 2 lineups, and each plays 20 minutes - so I have 5 players who always sub in together - this makes the math simple - here are the settings for distro:

team A
pg - 4%
SG - 8%
SF-4%
PF - 4%
C - 20%
total on the floor equals 40% (4+8+4+4+20)

if the team shoots 26 shots in the 20 minutes team A plays, approx 6 will be not from the offense and somewhat random, putbacks, etc. - which leaves 20 shots from plays or the offense or from distro.

here is the shot attempts I would expect

pg - 2 attempts
sg - 4 att
sf - 2 att
pf - 2 att
c - 10 att

now team B could be set all 5 players to 1%, that happens independent of team A (of course all 5 don't sub in together real life, but do in this simple example)

in that case, all 5 players from team B would expect to try 4 shots in their 20 minutes of action, plus the 6 random shots could go to near anyone.

now, with randomness in the offense number generator added, the 4 shots might be zero or it might be 8, but over the course of the season, the player trends to what you would expect

hope this did not confuse you

but, there is no need for the sum to be 100%, it can be ANYTHING, what matters is what % a player is at vs the sum of the players on the floor at any given time.

In the above example, if a team B player SG set to 1% went in for a SG set to 8% on team A, the sum on the floor goes from 40% to 33%, and the Team A starting Center's 20% distro - is now not expecting half the shots, but closer to 2/3 the shots (20%/33%)

so as each substitution occurs, the shot expectancy equation changes - make sense?

Yes,...excellent explanation

That all makes sense to me OR, thanks for clarifying it.  But to take it a step further, since we all know that players dont all sub in together, how can you predict the subing patterns in order to really fine tune your distro settings.  I know I always have trouble knowing what to set the back ups to and sometimes it results in starters taking too many shots.

Posted by metsmaniac2 on 12/29/2010 1:11:00 PM (view original):
That all makes sense to me OR, thanks for clarifying it.  But to take it a step further, since we all know that players dont all sub in together, how can you predict the subing patterns in order to really fine tune your distro settings.  I know I always have trouble knowing what to set the back ups to and sometimes it results in starters taking too many shots.

mets - I tend to do a little of it by trial and error, if something is out of whack, I tweek it.  The key is to do quick math in one's head to know if something that looks wrong really is a math / setting related issue, or random number theory at work.

and at least for me the arithmetic adjustments are easier if you dont try to make the numbers add to 100 - because then if you want to increase one guys distro you need to cut someone elses

make all your numbers add to 50 at the start of the season and then adjust during the season without focusing on the total

or make all your numbers add to something in the 80-90 range and do likewise

whatever is easy for your to get your head around - at least if you are like me and do this by looking at FG% and 3FG% and TOs etc and not by running a spreadsheet