Sleep experiments, built on research

Does chamomile tea help you sleep?

Modest gains in sleep quality, none in insomnia severity — and the trials used extracts, not tea. Test the chamomile against the ritual itself.

"I will drink a warm cup 45 minutes before bed every night: chamomile for 7, plain hot water for 7."

Does glycine help you sleep?

3 g before bed improved sleep quality and time-to-deep-sleep in trials of 11–19 people, all from one manufacturer. Cheap and testable.

"I will take 3 g of glycine 30 minutes before bed for 14 nights."

How long does magnesium take to work for sleep?

The trials measured after up to eight weeks. A first-night effect isn't in the literature — here's the realistic timeline and how to test it.

"I will take magnesium every evening for 14 nights and judge it in week two."

Does magnesium help you sleep?

The pooled trial effect is ~17 minutes faster sleep onset, low certainty, in older adults. Worth a two-week test — not a leap of faith.

"I will take magnesium every evening for 14 nights."

Magnesium glycinate vs citrate: which is better for sleep?

No sleep trial compares them. Absorption is similar; the interesting difference is the glycine that comes along for the ride. Test them back to back.

"I will take magnesium glycinate for 14 nights, then citrate for 14 nights, same dose."

Magnesium isn't helping your sleep. Now what?

Not low to begin with, misread label, wrong problem, or never measured. A 14-night stop-test tells you which — and what to try instead.

"I will stop magnesium for 14 nights and compare against the nights I was on it."

Does melatonin actually help you sleep?

Seven minutes faster sleep onset, on average. It's a clock signal, not a sedative — timing matters more than dose. Test it properly.

"I will take my usual melatonin at the same time, 1 hour before bed, for 14 nights — after 5 nights without."

Melatonin not working? Here's how to find out why

Unreliable content, wrong timing, too much, or the wrong problem. Four testable reasons — and an on/off protocol that separates them.

"I will take a low dose, 1 hour before bed, with a fixed wake time, for 14 nights."

Does tart cherry juice help you sleep?

Three trials of 8–20 people, modest gains in sleep time — and ~50 g of sugar a day at the trial dose. A tidy two-week test.

"I will drink tart cherry juice morning and evening for 14 days."

Does sleeping in total darkness improve sleep?

A dim 100 lux during sleep raised heart rate and morning insulin resistance in one night. An eye mask improved next-day alertness. Two weeks of darkness is an easy test.

"I will sleep in complete darkness — blackout blinds or an eye mask — for 14 nights."

Do you sleep better in a cold room?

Not hot matters more than cold: efficiency peaked at 20–25°C in 11,000 logged nights. Pick a number and hold it for two weeks.

"I will keep my bedroom at 19°C for 14 nights."

Do earplugs improve sleep?

Noise fragments sleep even when it doesn't wake you. Earplugs restored REM and melatonin in a simulated ICU. A cheap, side-sleeper-tested experiment.

"I will sleep with earplugs for 14 nights."

Does mouth taping help you sleep?

Halved apnoea events in 20 mouth-breathers with mild apnoea; a 2025 review calls the wider evidence thin and warns of real risks. Nose-breathing check first.

"After passing the nose-breathing check, I will use porous mouth tape for 14 nights."

Does a no-phone-in-the-bedroom rule improve sleep?

Phone use after lights-out tracked later bedtimes, slower sleep onset and more fatigue in 844 adults. One rule, one charger in the hallway, 14 nights.

"My phone will charge outside the bedroom for 14 nights."

Do socks in bed help you sleep?

Warm feet predict fast sleep onset. In a small trial, socks cut time-to-sleep by 7.6 minutes and added 32 minutes of sleep. Cheapest experiment on the list.

"I will wear socks to bed for 14 nights."

Do weighted blankets help you sleep?

60% vs 5% response in one strong trial — in people with insomnia and a psychiatric diagnosis. For everyone else: a melatonin bump and a two-week test.

"I will sleep under a weighted blanket (about 10% of my body weight) for 14 nights."

Does white noise help you sleep?

38 studies, very-low-quality evidence, results in both directions. If anything, it helps people in noisy places fall asleep. Run 7 nights on, 7 off.

"I will play white noise at low volume all night for 7 nights, then none for 7."

When should you have your last coffee?

Caffeine cuts total sleep by ~45 minutes on average — mostly without you noticing.

"I will have my last caffeine by 13:00."

Does alcohol ruin your sleep?

It helps you fall asleep — then disrupts the second half of the night. Test your Friday glass.

"I will have zero drinks for 14 days."

Does morning light improve your sleep?

Light is the main lever for your body clock — and morning light shifts it earlier.

"I will get 10 minutes of outdoor light within 30 minutes of waking."

Does an afternoon nap ruin your night's sleep?

Short and early rarely costs the night; long and late does. Seven days with a 20-minute nap before 3 PM, seven without.

"I will nap 20 minutes before 3 PM for 7 days, then not at all for 7."

How long before bed should you stop drinking alcohol?

Alcohol at 6 PM, zero on the breath at bedtime, still disrupted the night. Timing helps less than dose — test a 4-hour cutoff for two weeks.

"I will finish my last drink at least 4 hours before bed for 14 nights."

Is exercise before bed bad for sleep?

Meta-analyses: fine unless vigorous within an hour of bed. Wearable data from 4 million nights: within four hours costs sleep. Find your own cutoff.

"I will finish any workout at least 4 hours before bed for 14 days."

Does eating late affect your sleep?

Meals within 3 hours of bed went with more night wakings; evening fat with lower efficiency on PSG. Move dinner earlier for two weeks and see.

"I will finish dinner at least 3 hours before bed for 14 nights."

How long does morning sunlight take to improve sleep?

Two days of natural light moved the clock 1.4 hours; a week, 2.5. Expect earlier sleepiness within days — and track it for 14.

"I will get 15 minutes of outdoor light within 30 minutes of waking for 14 days."

Is sleeping in on weekends bad for you?

Catch-up sleep offsets some risk of short weeks — but not metabolic damage, and irregular timing predicts mortality better than duration. Cap the lie-in for two weekends.

"I will wake within 1 hour of my weekday time on weekends for 14 days."

Should you wake up at the same time every day?

Regularity beat duration as a mortality predictor in 61,000 people. One fixed wake time, weekends included, is the strongest lever on this site.

"I will get up at 6:45 every day, including weekends, for 14 days."

Does a warm shower before bed help you sleep?

Counterintuitive but well-documented: warming up 1–2 hours before bed helps your core cool down.

"I will take a 10-minute warm shower 1 hour before bed."

Does 4-7-8 breathing help you sleep?

Slow breathing reliably shifts the nervous system toward calm; the sleep-specific trials are small but positive. Ten minutes at lights-out, 14 nights.

"I will do 10 minutes of slow breathing (4-7-8 or six breaths a minute) at lights-out for 14 nights."

Should you use your bed only for sleeping?

The bed-is-for-sleep rule is stimulus control, a core component of insomnia treatment with guideline backing. Run the whole protocol, not just the slogan.

"For 14 nights: bed only for sleep; if awake for ~20 minutes, I get up until sleepy; same wake time every day."

Is blue light before bed a myth?

The light is real; the orange filters mostly aren't. Night Shift didn't beat an unfiltered phone in 167 adults. Test filter vs no phone, week against week.

"Week 1: phone with night mode in the last hour. Week 2: no screen in the last hour."

Does journaling before bed help you sleep?

Five minutes writing tomorrow's to-do list cut sleep onset by ~9 minutes on PSG. The more specific, the faster. Cheapest experiment with a lab result behind it.

"I will write tomorrow's to-do list for 5 minutes before lights-out for 14 nights."

Does meditation before bed help you sleep?

Mindfulness beat a sleep-hygiene course in a randomized trial (effect size 0.89); 18 trials pooled show modest, lasting gains. Ten minutes, 14 nights.

"I will do a 10-minute guided mindfulness practice at lights-out for 14 nights."

Does reading before bed help you sleep?

42% of assigned readers reported better sleep after a week vs 28% of non-readers, in a 991-person randomised trial. Paper, not a screen. Seven nights each way.

"I will read a paper book for 20 minutes before lights-out for 7 nights, then not read for 7."

Does screen time before bed affect sleep quality?

Bright screens before bed cut melatonin 55% and pushed the clock 1.5 hours later in the lab. One screen-free hour, 14 nights.

"I will put all screens away 60 minutes before bed for 14 nights."

Why can't you fall asleep even when you're tired?

Sleep pressure high, alerting system higher. Caffeine, light, a late clock or a bed that means 'awake' — rank the suspects, test one.

Sleep hygiene isn't working. What now?

The AASM recommends against sleep hygiene as a treatment. The checklist is an average; the fix is one lever, one baseline, 14 nights.

How accurate are sleep trackers?

Trackers detect sleep well, misjudge wake, and guess at stages. Some users sleep worse chasing the score. Rate your night before you read it — 14 mornings.

"For 14 mornings I will rate my sleep before looking at my tracker."

Why do you wake up at 3 AM and can't fall back asleep?

The second half of the night is where alcohol, late meals, heat and worry do their work. Rank your suspects, run one experiment, 14 nights.

How long until stretching makes you flexible?

Two to six weeks — and the effective dose is only ~10 minutes a week.

"I will stretch for 5 minutes after brushing my teeth."

Is zone 2 cardio worth the hype?

Fitness strongly predicts longevity; the "magic zone" part is hype. Test the honest version.

"I will do 30 minutes of easy cardio 3 times a week."

Do cold showers actually improve your mood?

The evidence is modest and delayed — which makes it a perfect self-experiment.

"I will end my morning shower with 60 seconds of cold."