Steve, expanding on my previous post:
What I realized in my writing is that complexity and convolution are not the same thing, and separating them in your mind is valuable for a writer for several reasons.
First, to separate them is to ask what you’re wanting to write. Do you want to challenge the audience with double-backs and twists or do you want them to experience richness? Or both? To separate complexity and convolution is to help you set goals.
Secondly, to separate them is to ask when is one or the other appropriate within a story. One part may need complexity, one part may need convolution. It is possible what seems to be appropriate may, at later examination, not be – a complicated murder plot may be more interesting from the viewpoint of a character who has it figured out, so you can explore their character.
Good analysis, one I should enrich further with some other things I've mulled since.
Setting goals for a story means a number of things, but more than anything else they seem to revolve around what kind of story you want to tell. Most people just want to entertain, and there's nothing wrong with that, but if you are aiming past that then you need to encompass things that allow you to shoot for such a goal. A story that is actually about something is hard to pull off if the only ways you're thinking about it are how to mess with the audience's head in a superficial way.
The second bit, about separation of concern by application, seems even more crucial. I've long felt that the best stories stood out not because they had the cleverest plots, but because they made the most compelling and thoughtful use of their material — that if they had a tangled storyline, they found the right lens through which to look at all of it. In fact, it doesn't even have to be about anything particularly complicated. Natsuo Kirino's Out had a fairly straightforward plot (wife kills no-good husband, enlists disgruntled female coworkers in cover-up), but the effects that story had on its characters was not so predictable, and so the book (which I cannot recommend highly enough) wisely focused on that aspect of its story rather than trying to invent complications that didn't need to be there.
Most authors, even many who don't think of themselves as but entertainers, want on some level to have their work thought of as profound. The bad news is that it's hard to engineer profundity into something, in the same way it's next to impossible to engineer cult status for something. It's a by-product, and it comes as a result of the way you handle other things, like complexity or simplicity.