Why novelty costs you sleep
Every variable in your evening is a decision waiting to be made badly. Which programme next. Whether to answer that message. One more chapter or not. Each of these is trivial at noon and genuinely difficult at half past eleven, because the part of you that decides well has already gone to bed ahead of you.
Tired people do not choose the early night. They choose the thing in front of them, and novelty keeps placing things in front of them. An evening with no new inputs is an evening with nothing to say yes to, which is precisely the point.
The order matters more than the time
A fixed bedtime sounds like the goal. It is not quite. What does the work is the sequence before it: the same handful of steps, in the same order, every night. Repeated often enough, that chain becomes a signal. The body learns that these particular small events end in sleep, and it begins the descent when the chain begins, not when your head reaches the pillow.
Kettle off. Teeth. Lamp. Done. A shifting bedtime with a fixed order beats a fixed bedtime that arrives out of chaos, because the order is the mechanism and the clock time is just the label on it.
Screens, briefly
The standard warning about screens concerns light, and the light is not the main problem. The content is. A feed is an unfinished thing: it never ends, never resolves, and every item in it was selected to make stopping feel premature. Closing a book leaves a story paused at a chapter break. Closing a feed leaves nothing completed and the faint sense that you left early.
Unfinished things follow you to bed and continue in your head. So finish things, or start only things that can be finished. A film ends. A chapter ends. A feed never does, and your evening ends with it still open.
Build a sequence you could do tired
Four steps are enough. No choice inside any of them, and the same order every night, weekends included. Something like: lock up, wash, lay out tomorrow, read ten pages of something you have already read. The test of a good sequence is whether you could do it half-asleep, in the dark, with a head cold.
If any step requires judgement, it will fail on exactly the nights you need it most. Boredom is not the price of the system. Boredom is the system.
Same time, same order, same dark. Give the sequence three weeks of dull repetition before you decide whether it works. The novelty can happen in the daytime, where it belongs.