Yeah but Plex, I believe, also maintains it's own database for the media files. Why can't I just "tell it" the metadata in those cases and leave the filenames as is?
It has an edit button in the UI for entering in the metadata to its database manually.
It has the option to prefer metadata already stored locally inside the container when populating its database.
It has the "match" button for interactively discovering and retrieving the metadata from an agent like IMDb or tvdb regardless of filename.
By default it tries to spare you having to do any of these things, but they put a lot of work into facilitating several alternatives when you'd rather do the work yourself.
In my own use case, I rip blurays, and these don't have the metadata populated. Files won't even appear in the UI if the filename isn't right, or it guesses completely wrongly based on just the filename. There is tucked away, really deep, some options for "unmatched" files. I had moved on to using filebot before trying that.
