This is the same as take two with different plots. What I've now done is to define a new soft ID for each bad soft ID

goodSoftID = badSoftID +/- 20

This defines a new "good" soft ID which we can use to compare plots to. This is a soft ID which lies in the same eta bin. Now we can plot the good/bad ADC and pedestal, as well as the ADC - ped overlayed on top of each other. The attached PDF's are the new ones and are from the same runs as the last time. Plots for min bias only and for all triggers are shown here.

Note: Plots for all of the bad runs can be found here: http://www.star.bnl.gov/protected/spin/jkadkins/hotTowerQaPlots/