How to actually fall asleep for a nap, with my own numbers

What normal actually looks like

The figure repeated everywhere is that a healthy adult falls asleep in ten to twenty minutes. That is a range presented as though it were tight, and my own data says it is not.

I kept a nightly diary across the 28 nights of February 2026, from the 1st to the 28th. Lights-out recorded at the time, an estimate the next morning of how long I had taken, no nights dropped.

One person, self-estimated rather than measured, and self-reported latency is known to skew long. Take it as illustrative rather than authoritative. The full method and caveats are here.

But look at the shape. My median sits inside the textbook range and my spread runs well outside it in both directions. One night in ten took me over half an hour. If I judged each night against "ten to twenty minutes" I would conclude something was wrong with me roughly a third of the time, and nothing was wrong on any of them.

That is the first practical thing to take from this page: you are probably not bad at falling asleep. You are probably comparing a distribution against a point estimate.

Why trying harder fails

Falling asleep is not something you do. It is something that happens when you stop doing other things, and effort is one of the things.

The mechanism is not mysterious. Sleep onset requires the arousal systems that keep you engaged with the world to quieten down. Monitoring whether you have fallen asleep yet is engagement with the world. Checking the clock is engagement with the world, and worse, it adds a deadline. Every minute you spend evaluating your progress towards sleep is a minute of the exact activity that prevents it.

This is why a nap you are anxious about is harder than one you are indifferent to, and why the advice to "just relax" is useless: it names the goal rather than a route to it. The routes below are all versions of one idea, which is to give your attention somewhere harmless to sit.

What moved my own number

I logged five things alongside the estimate. The gaps between subgroups were larger than I expected, and they were not the things nap articles usually talk about.

Now the caveats, because these are easy to misread as five independent effects and they are nothing of the kind. A high-stress night was also more likely to be a late-caffeine night, a screen night, and a night I had not exercised. Twenty-eight nights split five ways leaves few observations per cell. And it is one person.

What survives all that is a hypothesis rather than a finding: the two biggest gaps were stress and late caffeine, and those are what I would look at first if my own falling-asleep time started drifting. Notably neither appears on the standard list of nap tips, which tends to be about pillows and blackout curtains.

If your naps are hard and you are drinking coffee at 4pm, the coffee is a more promising lead than the curtains. The coffee nap piece covers caffeine timing in more detail, including the trial finding that 400 mg still disrupted sleep six hours before bed.

What to do before you lie down

Ordered by how much difference I think each one actually makes, rather than by how often it appears in listicles.

• Nap in the dip. Early afternoon, when your body is already heading that way. Fighting your circadian rhythm is a bigger obstacle than any amount of room setup. More on the window here.
• Empty your head onto paper first. A three to five item list of what you are supposed to be doing. This is the intervention I would keep if I could keep only one, because the thing keeping most people awake is rehearsal of unfinished business.
• Set the alarm and then stop tracking time. Turn the clock away. The alarm exists precisely so you do not have to monitor, and monitoring is the problem.
• Slow breathing or a body scan for two or three minutes. Not because either is magic, but because both give attention a boring job. That is the whole function.
• Cool, dark, quiet. Genuinely helpful and genuinely last on this list. Worth doing, unlikely to be your bottleneck.

One practical note that gets missed: count your falling-asleep time into the alarm. If you set a 20-minute timer and take my median 12 minutes, you get an 8-minute nap. The nap calculator adds it for you.

The release valve: you do not have to sleep

This is the part I would put first if the article were written for me rather than for search.

You do not have to fall asleep for the twenty minutes to have been worth taking. Lying still with your eyes closed, sometimes called wakeful rest, is not nothing. It is used in memory research precisely because it does something measurable, and at minimum it removes the demands that were tiring you.

The reason this matters is not consolation, it is mechanism. Requiring yourself to fall asleep creates the performance anxiety that stops you falling asleep. Removing the requirement removes the anxiety. It is the one piece of advice on this page that works by making you care less.

So the instruction I actually use: lie down for twenty minutes with no obligation to sleep. If sleep happens, good. If it does not, you rested, which was the point.

If you can never nap

Some people cannot nap at all, and that is a real and reasonably common thing rather than a failure of technique. If you sleep well at night, being unable to nap is not a problem to solve, it usually means your sleep pressure is well managed and there is not much to discharge in the afternoon.

Where it is worth attention is the opposite pattern: needing naps badly, every day, and still feeling unrested. That is not a napping question. Persistent daytime sleepiness despite adequate time in bed has clinical causes worth investigating, and no amount of breathing technique addresses them. I am a medical student and not a licensed clinician, and that is a conversation for a doctor.

Questions people actually ask

References

• Self-tracking diary, 1 to 28 February 2026. 28 consecutive nights, lights-out recorded at the time, latency estimated the following morning, with caffeine, alcohol, screens, exercise and stress logged alongside. No nights excluded. Full distribution and limitations on the nap calculator.
• Drake C, Roehrs T, Shambroom J, Roth T. Caffeine Effects on Sleep Taken 0, 3, or 6 Hours before Going to Bed. J Clin Sleep Med. 2013;9(11):1195–1200. Source for caffeine still disrupting sleep six hours before bed.
• General sleep-hygiene guidance on environment, wind-down routines and wakeful rest, as summarised by the Sleep Foundation and similar sources. None of these are strong individual findings and they are presented here in that spirit.

Related reading

• Nap calculator, which adds your falling-asleep time rather than assuming you drop off instantly.
• The best time of day to nap, the dip, and why fighting it is the hardest way to nap.
• The coffee nap, and what late caffeine costs you at bedtime.
• Does napping wreck your sleep at night?, the other side of the same trade.