I felt fine until my watch disagreed.

I was halfway through breakfast, feeling reasonably ordinary, when my watch informed me that the night had gone badly. I put down my spoon and considered whether I had been tired all along without noticing.

There was a graph. Several colours. A number sufficiently specific to seem like someone had been in the room taking notes. I opened yesterday’s result for comparison. My cereal became soft.

I tried to remember being asleep.

This proved difficult. I remembered getting into bed, waking once and hearing the alarm. The app suggested a much more eventful production. I could neither confirm nor dispute most of it.

I liked the watch for running routes and knowing roughly how my week looked. Somewhere along the way I had started treating its morning summary as an instruction about what kind of day I was permitted to have.

I wrote down a different question.

Was I regularly struggling to sleep or stay awake during the day? That would be worth discussing with a clinician. A consumer wellness score by itself was not a diagnosis. The sleep-medicine guidance I read distinguished a gadget’s data from an actual medical evaluation.

I turned off the morning score notification. I could still look at patterns later, if they were useful. I did not need the number beside my bowl before I had even noticed the weather.

The spoon was where I left it.

This was not the first time I had allowed a timer to make an ordinary activity weird. Owen still occasionally asked about my cold-plunge stopwatch. I chose not to provide him with new material.

I finished the cereal and went about my day. The next morning I checked whether there was milk before checking anything on my wrist. One of those facts could immediately improve breakfast.