I chose a window instead of a deadline: lights out between 10:45 and 11:15. The half hour mattered immediately. It made the experiment feel like a direction rather than a trap.
The first change was not deeper sleep but fewer negotiations. At 10:30, I stopped beginning things. The episode could wait. The kitchen light went off. Tomorrow no longer needed a committee meeting.
The morning became more predictable
By the second week, waking required less strategy. Energy still varied, but the variation felt connected to real causes rather than a schedule redrawn every night.
Consistency did not make every night perfect. It made each imperfect night easier to understand.
I kept Friday loose and returned to the window on Saturday. A routine that cannot survive friendship is too fragile to be useful.