Write a prioritized list. Put only features or capabilities on the list, not apps.
Include features that may seem obvious, such as "fast search" or "reorganize easily". For example, one reason for not writing things out on paper using a lovely fountain pen (or whatever) might be the complete absence of automated search. In other words, paper and pen doesn't support ⌘F for finding things; you have to manually catalog all the written words and then manually search the catalog.
Also include features like "runs on all my platforms: iOS, OS X, web", and features like "easy sync".
Make sure the list is kept prioritized. That is, if feature A is more useful than feature B, it should be higher in the list (closer to the start). If you're keeping the list in pen & paper format, write one feature per index card, and sort the cards in priority order.
Make sure the order is based on actual usefulness, not just "it's nice even though I rarely use it". If you want to know what actually gets used, keep track of that over the course of a week or so. I think you might be surprised.
Once the list starts to stabilize, you can start matching the features to the various apps. This should pretty quickly tell you exactly which apps are indispensible and which are meh, or even which are just kept around for sentimental reasons.
At any point, you can also do a "If I had to cut, where would that be?" on the prioritized list. In other words, if you had to draw a line in the list, and everything below the line was completely eliminated from your everyday tools, where would that line be?
You can maintain the list, move items up and down in priority, rematch to apps as their features change, etc.
One neat thing about the list is it's an exercise in writing, which means it uses the very tools and features being listed. Keeping the list is an exercise in evaluating the quality of the list. If you find you keep the list in a simple text editor, that alone tells you something useful.
Making a prioritized list is a well known way to determine which features are necessary when designing new software, vs. those which are merely nice to have. This can be used for decision guidance when it's time to schedule work or cut features.
It's also a good way to cull woodworking tools, or mechanics tools, or pretty much any set of tools.
There will always be tradeoffs between tools that are a more precise fit, and those that are "sloppy" but more flexible. To me, the canonical example is a set of open-end wrenches (spanners for UK readers), which have fixed sizes but are much less likely to strip a head, vs. an adjustable wrench (aka Crescent wrench) or a set of Vise-Grip pliers or just slip-joint pliers. Another example is a pipe-cutter vs. a hacksaw, or a chisel vs. a utility knife.
Wow, amazing post.
I really love your suggestion. I'll definitely do that and then come up with the results, so perhaps ideas can be discussed together. After all, my purpose here is not only to benefit myself, but to help other incipient writers—or even dedicated writers—to realize how they can do their work in the most efficient way. (As some have already said, sometimes technology can get in the way.)
By the way, are you a writer, software developer, both or none?