Keep this in mind if you're looking at old root files.
NOTE: Xin points out this shift may be due to different trigger types.
This must also be kept in mind. It would be FAR less confusing to just
eliminate any drift-time cut (or drift-time correct) from the current
analyzer until any trigger-dependent shifts are eliminated. (Or you will
introduce a software-induced trigger-dependent bias to the displayed data
-- this is a BAD THING.)
A copy of this log entry has been emailed to: moffit@jlab.org, khem@jlab.org, jinhuang@jlab.org, mitras@jlab.org, xqian@jlab.org