V4 and V5 background rates

[samanthalwong]

V4 and V5 background rates are ~3000% higher than V6 rates after bkg generation, but we expect lower rates in V4 and V5 (and see this in our data).

[steob92]

Is this referring to the bkg files?

If so, then the magnitude likely isn't too important since it's either re-scaled or used as a relative acceptance when calculating alpha. The important thing is the spatial and spectral shape. If we want to "help out" the fitter we could re-normalize to the observed counts. So each energy bin is normalized by the total number of counts in the observed FoV. This way bkg(E) / obs(E) ~ 1.

Probably not a good solution for highly extended sources as it will bias the background to the source flux... Maybe normalize to the most similar run used during the background generation?