You notice that your worst days for mental clarity are the days after you sleep badly. It shows up clearly. Nine times out of eleven.
It is tempting — and very common — to conclude that poor sleep causes your fog. That conclusion may be right. The data you have cannot establish it, and understanding why is what separates useful self-tracking from a hobby that generates confident wrong answers.
What the observation actually supports
"On days following less than six hours of sleep, my reported clarity was below my usual range on nine of eleven occasions."
That is the claim. It is a real finding, it is worth having, and it is worth taking to a clinician. It is not the same as "poor sleep causes my fog."
The four alternatives that are always live
Reverse direction. Whatever is causing the fog may also be disturbing your sleep. Many things do — pain, anxiety, depression, thyroid dysfunction, sleep apnoea. In that case the fog is not caused by the poor sleep; both are symptoms of the same thing, and treating sleep hygiene addresses neither.
A common cause. Something else drives both. A demanding week produces late nights and fog. Alcohol produces fragmented sleep and next-day dulling. The correlation is real and the causal arrow goes through a third thing.
Coincidence, helped by selective attention. Both variables have patterns of their own. Bad sleep clusters at the end of the week; so do deadlines. Once you have a hypothesis you notice confirming instances more readily and forget the exceptions — which is not a character flaw, it is how memory works, and it is precisely what a written record is for.
A measurement artefact. How you rate your clarity is affected by how you feel about your night. Knowing you slept badly changes the number you write down, independently of your actual function.
Why this matters practically
Because acting on the wrong arrow wastes months.
Someone whose fog and poor sleep are both being produced by untreated sleep apnoea can spend a year on sleep hygiene and get nowhere, because sleep hygiene does not treat apnoea. Someone whose fog is a depressive symptom can spend a year optimising a bedtime routine while the depression goes untreated.
The correlation was real in both cases. The inference was wrong, and the wrong inference is what prevented the useful question from being asked.
What would be needed to do better
The honest answer for a single person's data is: more than you can usually get. But some things help.
Timing. If the fog reliably follows the poor sleep rather than preceding it, that is weak support for one direction. Weak, not conclusive — but worth noting, and only visible if you record both with dates.
Look for the exceptions deliberately. The days you slept badly and were fine, and the days you slept well and were foggy. These are the most informative observations you have and the easiest to overlook. If there are many of them, the relationship is weaker than it feels.
Record the obvious third variables. Alcohol, workload, illness, medication. If fog tracks workload better than it tracks sleep, that is findable — and only if workload is written down.
Change one thing deliberately. The closest you can get to an experiment. Pick a period, change one variable, keep everything else as constant as life allows, and see. Two or three weeks, not three days. Imperfect, confounded, and still much better than observation alone.
Notice when you cannot separate two things. If your bad sleep and your busy weeks always coincide, your data cannot distinguish them. Saying that plainly is more useful than picking the explanation you prefer.
What to bring to a clinician
The association, not the conclusion.
Over the last two months, my clarity was below my usual range on nine of the eleven days following nights under six hours. There were also two days it was low after sleeping well. Alcohol was involved on four of the eleven. I have been waking unrefreshed even after long nights, which is newer.
That paragraph invites investigation. A clinician reading it will think about apnoea, about alcohol, about mood, about iron — a set of possibilities you did not have to generate.
Compare with: "I have worked out that my brain fog is caused by not sleeping enough." That is a conclusion, it may be wrong, and it tends to close the conversation rather than open it. It also invites disagreement, which is a worse use of an appointment than shared curiosity.
The general rule
A pattern in your own record answers the question what happened together. It does not answer why.
That is not a weakness of self-tracking; it is the correct reading of what it produces. The value is real and bounded: you arrive with a specific, dated, honest observation instead of an impression, and somebody qualified takes it from there.
Holding the boundary is what keeps the record trustworthy — including to yourself, six months later, when you are trying to work out whether something has actually changed.