Hoi Ric, there was a bug in the implementation of the pidkilling at the recoilNtp level, this change should affect electrons most directly. In addition, the track selection scheme is tightened to leptons in this version (it used to be GTVL in that posting). I am not sure what the best syst. error for muons is. The current variation is based on the merged 200x tables, and as evident from my table, the central value is already 5.4% higher compared to the split 2000/2001 tables (which were used in the Beta-level killing). Maybe we should consider using the variation relative to the merged tables as systematics (and I'd need to run the merged electron PidTables jobs). More to come. Cheers, --U.