Alexia Mimic Reconstruction

In order to try to pin down the differences between Alexia and my reconstruction, I've tried to mimic the conditions under which she did her reconstruction for her paper.

The binning is as follows:
#Angular Correlation
corrBins = 40.0 #should be one more than you want it to be
mincorr = 6.0 #in degrees
maxcorr = 12.0 #in degrees

#Auto Correlation
xicorrmin = 5.0
xicorrmax = 50.0
nxicorr = 40

#make redshift bins
rbinmin = 0.0 #in Mpc/h
rbinspace = 100.0 #in Mpc/h
rbinmax = 500.0 #in Mpc/h
nrbins = 5.0

Here are plots...

Correlation Functions

Distribution Functions and Data

Best Fit Reconstruction

A couple things I am noticing...
First: I still get a problem with changing the binning and the reconstruction going all weird.
Second: Even in this best fit scenario I am getting a mis-fit between the reconstruction and the data, could there be a problem with the lining up of my bins? Like perhaps something with left/right binning in the histogram?
Third: The reconstruction tends to go weird at the ends, even though both ends are contained within the redshift bins.

Some other reconstructions that aren't working as well:

The log file for this is here: ../logs/100707log.py
The run directory is here: ../pythonJess/run201077_1424