The geometry and luminostiy asymmetries that showed up on the last post were do to a bug in the code.  Fortunately while looking for it I came across a couple other mistakes and was able to correct them.  Firstly there was a binning error causing the ninth bin to not be counted for the geo. and lum. asymmetries.  Also noticed that the simulated events are generated in a circlular shape while the simulated detector is rectangular causing more than 1/3 the events to be cut off simply by being outside the rectangle.  Here's the latest version of the asymmetry plots:

This is the plot from real data used to approximate the radial dependence.  The y axis is the physics asymmetry and the x axis radial position where 1 is the innermost ring, 2 is the middle ring, and 3 is the outermost ring.