Haha, glad you are having fun. Did you use the highlighting buttons etc to help you along the way? You have no idea how long it took me to get that working but when figuring out xpath expressions I found it invaluable.
So long as you don't hit Finish on the Wizard you can safely experiment without using a test library, after all it is only if you actually specifically click the buttons to create empty books and click Finish that this plugin will actually make any changes to your library.
The series is indeed a problem in this situation. Of course this plugin didn't originally have series as an option, it was only when I added FF as a source and looked at the "scraping" possibilities that it came in. My thoughts were always that you could just do a metadata download to get series information populated like you would with any other import of books into your library, so in your situation I would just leave the series/series idx columns blank.
*If* the strip field was allowed to be applied at the field level (allowing different regexes for title, author, series etc) then you would have a solution to both you series columns above, allowing you to strip what you don't want out in both circumstances and would be more flexible. I'm not overly against changing the plugin to support this (it always felt a bit filthy applying the same "strip" regex to both title and author) but it is a question of how/where to configure this on that screen. Maybe it would have to be a popup dialog - so where it says strip it just tells you what fields have been given regexes for stripping, and a button pops up showing a grid where for each field you can edit the regex for it... how does that sound?
I can also appreciate how in the situation of scraping from goodreads (or from FF for that matter) that it would be darned useful to scrape the goodreads id as an identifier. So then when you do a metadata download you are getting data for exactly that same book edition. Hmmm...