The worst nap of my life was a Sunday afternoon between two late shifts. I lay down at three, meant to sleep for twenty minutes, and woke at ten past four with no idea what day it was. I was groggy until dinner and then wide awake at midnight. This is the calculator I wish I had set the alarm with.
Tell it when you are lying down, or when you need to be alert, and pick a nap. It gives you the alarm time, and it tells you when the nap sits too close to bedtime to be worth it.
How the math works
A nap has two parts: the minutes it takes you to drop off, and the sleep itself. The calculator adds them. Lie down at 1:00 p.m. with 10 minutes to fall asleep and the power nap alarm is 1:30 p.m. The full cycle from the same start is 2:40 p.m.
The other mode counts backwards. If you have to be sharp at 3:00 p.m. and want a full cycle, it puts you on the bed at 1:20 p.m., and the alarm is your alert time itself.
The start time is set to now when the page loads, rounded up to the next whole minute, because that is usually the honest answer. Change it if you are planning ahead.
Time to fall asleep is set to 10 minutes. That is a fair middle for someone who is tired enough to want a nap. The slider goes 0 to 30, and if you are still awake after thirty minutes, get up. Lying there is not a nap.
The last input is your usual bedtime, set to 23:00. When the nap would end less than six hours before it, you get a warning. Six hours is my own rule of thumb from the spreadsheet, not a study. A late nap uses up the sleep pressure you need at night, and if you know your body handles late naps, ignore it.
Why 20 and 90, and not 45
Sleep runs in cycles of roughly 90 minutes: light sleep first, then deep sleep, then back up through REM. A 20 minute nap stays in the light stages, so the alarm wakes you before deep sleep has started and you can stand up straight away. A 90 minute nap is about one whole cycle, so you surface at the end of it, which is the same trick the sleep calculator does with a whole night.
Thirty to sixty minutes is the awkward middle. That is long enough to reach deep sleep and not long enough to come back out of it. The alarm lands in the deepest part, and the thick, slow feeling afterwards is what the research calls sleep inertia. Tassi and Muzet reviewed it in Sleep Medicine Reviews in 2000: it is worse when you are woken from deep sleep, and it can hang around for a while. That is the ten past four feeling from my Sunday.
The clearest nap comparison I know is Brooks and Lack, in Sleep, 2006. They gave people 5, 10, 20 and 30 minute afternoon naps and tested them afterwards. The 10 minute nap gave the most benefit with the least grogginess. The 30 minute nap helped too, eventually, but it came with a groggy patch first. So the power nap on this page is 20 minutes with the fall-asleep time on top, and the alarm goes before most of the groggy patch has a chance.
The 60 minute recovery nap is on the calculator anyway, because some days you are so far behind that you will take the grogginess for the sleep. Expect the groggy patch when the alarm goes. Use it when a night has gone badly wrong, not as a habit.
The coffee nap
The idea is to drink the coffee right before you lie down, not after. Caffeine takes roughly 30 to 45 minutes to peak in the blood, so it has not done much by the time you drop off, and it arrives about when a 20 minute alarm goes. You wake up as the coffee kicks in instead of waiting for it.
Reyner and Horne tested this in Psychophysiology in 1997. They gave sleepy drivers 200 mg of caffeine, roughly a large cup, followed by a nap of under 15 minutes, and measured lane drifts in a driving simulator. The caffeine plus the nap did better than the caffeine alone, and better than the short nap on its own from their earlier study the year before. That is the whole claim. It is one lab study with a simulator, and it does not work for everyone. The first time I tried it I just lay there buzzing, and it took a couple of goes before I could drop off with the coffee in me.
That is why the toggle only appears for the power nap. A coffee before a 60 or 90 minute nap means the caffeine peaks while you are trying to stay asleep, which is the opposite of the point.
When to trust it and when not to
Trust it when your problem is that you nap too long, or too late, or both. Those are timing problems and the arithmetic fixes them. The warning is a warning, not a lock: if you are working nights and 6 p.m. is your morning, the six hour rule does not apply to you the way it applies to a day shift, and you should read the bedtime input as your real bedtime, whenever that is.
Trust it less if you never feel better after a nap no matter the length, or if you need one every single day to get through the afternoon. That is a sign the nights are short, and the fix is in what time should I go to bed, not here.
If you lie awake for an hour most nights, or you snore hard enough to wake yourself, or nine hours in bed still leaves you exhausted, those may be insomnia or sleep apnea and they belong with a doctor, not a nap calculator. I am not a doctor, and I would rather say so plainly.
Here is the step for today. If it is early enough, set the alarm for a 20 minute power nap right now and see how you feel when it goes. If the page is warning you about bedtime, skip the nap, and count back from the alarm tonight instead.