Field note
The 10-3-2-1-0 sleep rule, traced back to whoever made it up
The 10-3-2-1-0 rule is a bedtime countdown: no caffeine 10 hours before bed, no food or alcohol 3 hours before, no work 2 hours before, no screens 1 hour before, and zero hits of the snooze button. It was written by Craig Ballantyne, a fitness and productivity author, and published in HuffPost on 13 May 2016 as part of his "Perfect Day Formula." It is not a clinical protocol, it has never been tested as a package, and no sleep-medicine body issues it. Three of the five steps are well supported by research in the general direction they point. The specific hour counts mostly are not: the 10-hour caffeine rule is the clearest case, going four hours further than the trial usually cited for it.
- Origin: a productivity article, not a study. HuffPost, May 2016.
- Never tested as a whole. No trial of the protocol exists.
- The 10 is the weakest number. The evidence supports 6 hours.
- The 3 is the strongest. Alcohol timing holds up well.
It is all over TikTok, it gets described as something sleep doctors recommend, and it sounds like it came out of a lab. It did not. It came from a fitness and productivity author in 2016, and it has never been tested as a protocol. Here is each of the five numbers checked against the research it is usually credited to, including the one that overshoots its own study by four hours.
Third-year medical student at Manipal Tata Medical College, writing about sleep research. A student, not yet a licensed clinician, so everything here is sourced, and none of it is medical advice. Why you should or should not trust me →
Published 2026-08-31 · Last reviewed 2026-08-31 by Chaitanya Banerjee
Researched against primary sources, drafted with AI assistance, then reviewed and approved by the named author. How we write · Corrections log
What the rule actually says
Quoted as it was originally published, before years of paraphrase:
| Step | The instruction | Does the number trace to research? |
|---|---|---|
| 10 hours before bed | No more caffeine | No, the trial supports 6 hours |
| 3 hours before bed | No more food or alcohol | Yes, for alcohol |
| 2 hours before bed | No more work | Direction supported, number arbitrary |
| 1 hour before bed | No more screen time | Effect real, number not from the study |
| 0 | Times you hit snooze in the morning | Reasonable, evidence thin |
It is a good mnemonic, and I do not want to be sniffy about that. It is genuinely part of why it spread: a descending countdown is far easier to remember than five separate pieces of sleep-hygiene advice, and adherence to anything you cannot remember is zero. The problem is not the format. The problem is what happened to its authority on the way to your feed.
Where it came from
The trail ends at a named person on a specific date, which is more than I can say for most viral sleep advice. Usually I hit a wall of blogs citing blogs. Here I did not.
Craig Ballantyne published "The 10-3-2-1-0 Formula" in HuffPost on 13 May 2016. Ballantyne is a fitness and productivity author, best known for The Perfect Day Formula. The article's framing is explicit about its purpose: it is a routine for structuring your day so that you wake "well rested and ready for battle." The stated goal is getting more done. It is filed under productivity, not medicine.
From there it moved through Inc. and Medium, was picked up on health blogs, and eventually reached TikTok, where it acquired a second life around 2021 and started being attributed to physicians and "sleep experts" generally. That attribution drift is the whole story of this page. Nothing about the advice changed. What changed is that a productivity formula started being presented as clinical guidance.
This is not a takedown of Craig Ballantyne, who did nothing wrong. He wrote a memorable routine, labelled it as a productivity formula, and published it under his own name. The failure happened downstream, in every article since that has described it as evidence-based without checking. If you are going to repeat a number, you should be able to say where it came from.
10. No caffeine ten hours before bed
This is the number that fails hardest, and it fails in a direction I did not expect: it is too cautious, not too lax.
The study almost always cited in support is Drake, Roehrs, Shambroom and Roth (2013), published in the Journal of Clinical Sleep Medicine and run at the Sleep Disorders & Research Center at Henry Ford Hospital with Wayne State University in Detroit. It gave participants a fixed 400 mg dose of caffeine, roughly two to three cups of coffee, at 0, 3 and 6 hours before their habitual bedtime, against placebo, and measured sleep at home.
The finding was that caffeine still significantly disrupted sleep at the six-hour mark. That was the surprise: six hours out, and it was already too late. The authors' conclusion is that this supports refraining from substantial caffeine for a minimum of six hours before bedtime.
The gap
Ten hours was not tested. I went looking for the study behind the ten-hour rule and found one that stopped at six, and six was the point at which the effect was still detectable, so the honest reading is "at least six, and we do not know where it stops." Somebody converted that into ten, and ten now circulates as though a trial produced it.
There is a reasonable defence of ten: if six hours still disrupts sleep, a larger margin is prudent, and caffeine's half-life varies enormously between people. But that is a judgement call, and it should be labelled as one. Presenting it as a research finding is the part that is wrong.
What the study also does not tell you: it used one large fixed dose in one group of people. It says very little about a single morning cup in a fast metaboliser, which is the situation most people are actually in.
3. No food or alcohol three hours before bed
This is the strongest number in the rule, at least on the alcohol half.
The controlled dosing literature, synthesised most recently in a 2024 systematic review and meta-analysis in Sleep Medicine Reviews, shows a clear dose-dependent pattern. Alcohol suppresses REM sleep in the first half of the night, then produces rebound and fragmentation in the second half as it clears. Disruption appears at doses as low as roughly two standard drinks, and low doses taken within about three hours of bed delay REM onset and reduce total REM.
A moderate dose clears over roughly four to five hours, which maps neatly onto when the second half of an eight-hour night starts breaking apart. Three hours is a defensible floor drawn from that clearance window rather than a number chosen for symmetry.
The gap
Small, and mostly on the food half. The alcohol timing holds up. The "no food" component rests on much thinner ground: the evidence on late meals and sleep quality is real but considerably messier, confounded by meal size and composition, and it does not produce a clean three-hour threshold. Bundling food and alcohol into one step gives the weaker claim a free ride on the stronger one.
If I had to keep one step from the whole rule, it would be this one, and I would keep it for the alcohol rather than the food.
2. No work two hours before bed
Here the honest answer is that the direction is sensible and the number is invented. I could not find anything behind the two.
Cognitive and emotional arousal before bed genuinely does lengthen sleep onset, and the mechanism is not controversial: the arousal systems that keep you engaged with a difficult problem are the same ones that have to quiet down for you to fall asleep. Rumination about work is one of the better-documented contributors to sleep-onset insomnia.
The gap
No study I could find establishes two hours as the threshold, and there is no experiment comparing a one-hour work buffer against a two-hour one and finding the second superior. Two is a plausible-sounding round number sitting between the three and the one, and the countdown structure requires something there.
That does not make it bad advice. It makes it advice, rather than a finding, and the difference matters if you are the sort of person who feels they have failed at sleep hygiene because they answered an email 90 minutes before bed.
1. No screens one hour before bed
The underlying effect is one of the best-documented things in this entire rule. The number attached to it is not from the study.
The reference point is Chang, Aeschbach, Duffy and Czeisler (2015), published in PNAS, from the Division of Sleep and Circadian Disorders at Brigham and Women's Hospital and Harvard Medical School. Participants read on a light-emitting eReader in the evening, compared against reading a printed book. The light-emitting condition suppressed evening melatonin by about 55%, pushed dim-light melatonin onset more than 1.5 hours later, lengthened sleep onset, reduced and delayed REM, and left people measurably less alert the next morning.
The gap
Participants read for four hours before bed, in a laboratory, on a device held at reading distance in an otherwise controlled light environment. That is a deliberately strong exposure, designed to produce a measurable signal. It is not the same experiment as "put your phone down an hour before bed."
Nothing in the paper identifies one hour as a threshold below which the effect vanishes. I think one hour is a sensible practical compromise, advice to stop four hours before bed would simply be ignored, but it should not be dressed up as the study's result.
0. No hitting snooze
The odd one out, because it is the only step that happens in the morning, and the only one aimed at consistency rather than at removing a disruptor.
The reasoning is that fragmenting the last stretch of sleep with repeated short awakenings gives you neither useful sleep nor a clean wake-up, and that a fixed wake time is one of the more reliable stabilisers of circadian timing. Regularity of wake time is well supported as a sleep-health principle.
The gap
The specific claim that snoozing is harmful is less settled than the rule implies, and some more recent work has questioned whether a short snooze period does measurable damage for people who are not sleep-deprived to begin with. It is reasonable advice with a thinner evidence base than its confident phrasing suggests.
There is also something circular about it that I cannot get past. If you need the snooze button, the problem is usually that you did not get enough sleep, which is a quantity problem the other four steps were supposed to solve.
So should you follow it?
Yes, mostly, and for a reason that has nothing to do with its scientific pedigree. Having spent a while pulling it apart, I would still use it.
Sleep hygiene advice fails at the point of memory. I know I should manage caffeine and light and alcohol and wind-down time, and then at 10pm I do none of it, because the advice sits in my head as five disconnected facts rather than one thing I can recall. A descending countdown is a good piece of design. Ballantyne's real contribution was packaging, and packaging is not nothing.
What I would not do is treat the specific numbers as thresholds you have failed to meet. Ten hours of caffeine abstinence is stricter than the evidence requires. One hour of screen-free time is looser than the study that gets cited for it. Two hours away from work is a guess. The rule is a reasonable routine with arbitrary precision bolted on, and the arbitrary precision is what makes people feel they are doing it wrong.
Use it as a shape, not as a specification. And if you want to know where any other sleep number you have heard came from, that is what I built the source index for.
References
- Ballantyne C. The 10-3-2-1-0 Formula: Win Your Mornings and Sleep Better. HuffPost, 13 May 2016. The origin of the rule, in the author's own words and framing.
- 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 the six-hour caffeine finding and the 400 mg dose.
- Chang AM, Aeschbach D, Duffy JF, Czeisler CA. Evening use of light-emitting eReaders negatively affects sleep, circadian timing, and next-morning alertness. PNAS. 2015;112(4):1232–1237. Source for the 55% melatonin suppression and the four-hour exposure.
- The effect of alcohol on subsequent sleep in healthy adults: a systematic review and meta-analysis. Sleep Medicine Reviews. 2024. Source for the dose-dependent REM suppression and the roughly three-hour window.
- Watson NF, Badr MS, Belenky G, et al. Recommended Amount of Sleep for a Healthy Adult: A Joint Consensus Statement of the American Academy of Sleep Medicine and Sleep Research Society. J Clin Sleep Med. 2015;11(6):591–592. Referenced for what an actual sleep-medicine consensus statement looks like, in contrast.
What this page does not cover
This is a provenance check on five numbers, not a treatment guide. If you consistently cannot fall asleep, wake repeatedly, or are sleepy during the day despite spending enough time in bed, no bedtime countdown is the right tool and you should talk to a doctor. Insomnia has actual treatments with actual evidence behind them, and cognitive behavioural therapy for insomnia is the first-line one.
I have also not tested this rule on myself in any structured way, and I am not going to pretend otherwise. Everything above is a reading of published sources, not a personal experiment. Since no one appears to have tested the package as a package, I am planning to run it on myself and publish whatever comes back, including if it is nothing.
Questions people actually ask
Who created the 10-3-2-1-0 sleep rule?
Craig Ballantyne, a fitness and productivity author, published it in HuffPost on 13 May 2016 as part of his "Perfect Day Formula." It is not the work of a sleep researcher and is not issued by any sleep-medicine body.
Is the 10-3-2-1-0 rule backed by science?
Not as a package. No clinical trial has tested the protocol as a whole. Individual components have research support of varying strength. The alcohol timing is well supported, the screen and work components point in a supported direction with arbitrary hour counts, and the 10-hour caffeine cutoff goes four hours beyond the study usually cited for it.
Should you really stop caffeine 10 hours before bed?
Six hours is what the evidence actually supports. Drake et al. (2013) found that 400 mg of caffeine still disrupted sleep six hours before bedtime, and recommended a minimum six-hour gap. Ten hours is a larger safety margin someone added; it is defensible as caution but was never tested.
What does the 0 in 10-3-2-1-0 mean?
Zero is the number of times you hit the snooze button in the morning. It is the only step aimed at the morning rather than the evening, and its evidence base is thinner than the rule's confident phrasing suggests.
Is the 10-3-2-1-0 rule the same as sleep hygiene?
It is a repackaging of standard sleep-hygiene principles into a memorable countdown. The principles are conventional; the format and the specific hour counts are Ballantyne's.
Put a number on it
The calculator takes your wake-up time and counts backwards. Every assumption it makes is printed on the page, including the ones it is guessing at.
Open the sleep calculator