When monitoring the SLI, the "MODEL" calculations for the beam's profile and energy spread should be used. The method for analyzing it was modified slightly, to make certain the background level was appropriately taken into account and not over-subtracted. The "raw (visual) data" result does not take the background level into account at all, and so will differ significantly with the finite background level present now.
Also, the SLI is an 'upper limit' on the energy spread at this point,
since the optics have been changed since it was last calibrated. If we
(Hall A) give the beam-diagnostics team an hour, they have a set
procedure to run through to calibrate the SLI to give a "real"
deconvoluted energy spread, apparently removing the effect of the finite
beam-size.