Does the Military Sleep Method Actually Work? What the Evidence Says

Short answer: partly, and not for the reason it went viral.

The mechanism underneath the military sleep method is well supported. The famous claim attached to it — asleep in two minutes, 96% success after six weeks — is not. Those are two different statements and almost every article about this technique runs them together.

Where the claim comes from

The method, and the numbers, come from Relax and Win: Championship Performance, a 1981 book by Lloyd Bud Winter, a sprint coach who had worked on relaxation training with naval aviation cadets during the second world war. The book describes the routine and reports that after six weeks of practice, 96% of pilots could fall asleep within two minutes — even after coffee, even with gunfire noise playing.

That is the entire evidentiary basis. It is a coach’s account in a trade book. There is no published randomised controlled trial of this specific six-step sequence, no replication, and no measured sleep-onset latency. Forty-five years of repetition has not turned it into data.

So when a headline says “science says you can fall asleep in two minutes”, the honest correction is: no it does not, a 1981 sports book says so.

What is actually established

Strip the branding off and the military method is progressive muscle relaxation — systematically releasing tension group by group — with a brief mental-imagery step at the end.

Progressive muscle relaxation has real support for sleep onset. It appears across more than thirty randomised trials of relaxation techniques for falling asleep, and relaxation training is one of the standard components inside CBT-I, which is the first-line treatment for chronic insomnia recommended ahead of medication.

The imagery step has its own small literature. Imagery distraction — deliberately occupying the mind with neutral visual content — has been shown to shorten sleep onset compared with general distraction or no instruction, in work going back to Allison Harvey’s group in the early 2000s. That is roughly the same mechanism as cognitive shuffling, which is a related but separate technique.

So: a well-supported relaxation protocol, plus a plausible imagery step, wearing an untested number.

What to actually expect

Not two minutes on night one. The trials that show relaxation techniques working train people over weeks. Winter’s own claim was after six weeks of practice, which the viral versions consistently drop — the number people repeat is the endpoint of a training programme, presented as a first-night result.

A wind-down, not a switch. What relaxation reliably does is lower physiological arousal. If you are lying there tense, that is exactly your problem and this helps. If you are lying there calm but thinking hard about a conversation from 2019, relaxing your jaw will not touch it, and an imagery technique is a better fit.

Diminishing returns if you keep score. Checking whether it worked is itself arousing. The moment the method becomes a test you can fail, it starts working against you — which is the whole logic behind paradoxical intention, the technique that tells you to try to stay awake instead.

Why people report it failing

Three reasons show up repeatedly, and none of them is the technique’s fault.

They ran it once. A trained skill, tested once, in a bad week.

They cut the last step. Most short write-ups list four physical steps and skip the ten seconds of clearing the mind. That step is where the imagery evidence lives. Without it you have done some stretching.

They were too awake to be in bed. If you have been lying there for forty minutes, no relaxation sequence is the right tool — getting up is. Stimulus control is the best-evidenced element in this entire field and it is the one nobody wants to do.

How to give it a fair test

Two weeks, every night, in order, including the imagery step, and without checking the clock afterwards. Give each stage fifteen to twenty seconds rather than three. If you lose your place in the dark — which nearly everyone does around the legs — use something that paces it for you rather than trying to hold a six-step list in your head.

That is the specific job Sleep Switch does with it: the method runs as a guided drill with a voice setting the pace, the screen dims as it goes and turns itself off at the end, and the app labels the technique honestly — “well studied”, for the relaxation mechanism, and it makes no two-minute claim anywhere. The full step-by-step is here if you would rather run it yourself.

When to stop experimenting

If you have had trouble sleeping most nights for a month or more, you are past the point where technique-hunting is useful. That is chronic insomnia, it has an actual treatment in CBT-I, and a doctor can point you at it. Relaxation drills are a reasonable thing to do on an ordinary bad night. They are not a treatment, and an app that tells you otherwise is selling you something.