OK, I've figured out the "New Series: " issue, and we can call it a pebcak error if you like.

It seems that "New Series: " is not at the beginning of the following line, but somewhere in the middle.
New Series: Christy, a newly sober single mother, is trying...

The reason for this is that the text replacement is performed before the s/d data is removed. Therefore, the original line looked like this:
S1 E1 New Series: Christy, a newly sober single mother, is trying...

So the solution is to change the rule to look 'anywhere' for this text, as opposed to just at the beginning.

edit: it would be nice to move the text replacement logic to the end.  This would stop false positives on things like "All New" which could in all liklihood be part of a normal description.  I suppose case sensitivity helps.