Electron Embedding Problem

I have been playing around with the 2009 upsilon and single electron embeddings and noticed a potential problem.  In a given pT region, the ADC count distributions do not align.  This problem will effect our HT trigger efficiency calculations and potentially effect the analisis of anyone who is using these data sets.

Here is a small presantation detailing the issue; any input into this situation (what caused this?  which one is right?  etc.) would be greatly appreciated.

Also attached is a compareison with the 2009 JPsi embedding as well in a representative pT bin.  From this it seems that the JPsi and the single electron embeddings agree on this issue, whereas the Upsilons disagree.