Do magnesium, melatonin, the sleepy girl mocktail and mouth tape actually work for sleep?

By Amanda Monger, who fixed her own broken sleep after thirteen years on tour and tracks every night. Published September 30, 2026. Not medical advice.

Magnesium

For: a 2021 systematic review found 3 randomized trials, 151 people, all older adults, and magnesium got them to sleep about 17 minutes faster than placebo. 2024 added two more trials, one on magnesium L-threonate and one called the largest placebo-controlled magnesium and sleep trial to date, both reporting better self-rated sleep. A 2024 review of magnesium for anxiety and sleep: helpful mainly in people who were already low in magnesium.

Against: the 2021 review rated all three trials low quality and said the evidence was too thin to recommend it to a doctor's patient. Trials are small, short and mostly in older adults, not women in their 30s. If your magnesium is normal, there is no trial showing it does much of anything. The largest single trial (155 people) found a small effect.

Who it might help: someone actually low in magnesium, which is common with a low-vegetable diet, heavy sweating and some medications. A basic blood panel your doctor can add to any visit tells you. Dose and form are a pharmacist question, not a TikTok one.

The sleepy girl mocktail (tart cherry juice, magnesium, sparkling water)

For: the cherry half has real, tiny studies. 15 older adults with insomnia for two weeks, less time awake after falling asleep. 20 adults for seven days, more sleep. About 8 to 11 adults with insomnia, 84 more minutes of sleep. A 2025 review: 5 of 6 studies reported better sleep quality after tart cherry.

Against: add up every tart cherry sleep study ever run and you get fewer than 100 people; the biggest had 20. Nobody has tested the actual mocktail. The sparkling water does nothing. A 2023 meta-analysis titled "Too Sour to be True?" found the pooled effect small and the studies at high risk of bias. And 8 ounces of tart cherry juice is about 25 grams of sugar, roughly a can of soda, an hour before bed, plus a full glass of fluid right before the hours you want to stay in bed.

Who it might help: honestly, the person who needs a bedtime ritual. The ritual is the part that works, and it works for free. If you want the mocktail anyway: 2 ounces of juice in a big glass of water, 60 to 90 minutes before bed, or the magnesium in water with no juice. Same ritual, less sugar, fewer bathroom trips.

Melatonin

For: the biggest meta-analysis, 19 studies and 1,683 people, found melatonin got people to sleep 7 minutes faster and gave them 8 more minutes, and the effect held over time. Strong evidence for jet lag and for people whose body clock is shifted (shift work, night owls). Low doses, 0.5 to 1 mg, work as well as high doses in most trials.

Against: seven minutes. That is the effect. It is real and it is small. A 2023 JAMA study lab-tested 25 melatonin gummies: 22 of 25 were mislabelled, with actual content from 74 percent to 347 percent of the label, and one contained no melatonin at all. Long-term nightly use for years is simply unstudied, and the sleep medicine society recommends against it for insomnia.

If you use it anyway: a USP Verified or NSF Certified mark on the bottle, tablets or capsules and never gummies, 0.5 to 1 mg not 5 or 10, for a reason (jet lag, a shifted clock) and a season, not forever. Melatonin is a timing signal, not a sleeping pill.

Mouth tape

For: one systematic review covering 25 years: 10 studies, 213 people. 2 of the 10 found a real improvement in breathing during sleep, in people with mild sleep apnea whose noses were clear.

Against: the other 8 found no difference. Zero studies on healthy people who just want better sleep, which is everyone it is being sold to. And the safety warning is specific: if your nose is blocked (allergies, congestion, a deviated septum) and your mouth is taped, you have closed your backup airway. Up to 75 percent of women with sleep apnea do not know they have it, and you cannot tell which one you are from a video. Mine is blocked most of the year, so that is a hard no for me.

L-theanine and ashwagandha, briefly

L-theanine at the doses trials used has 13 small studies showing faster sleep onset; it is the best-supported of the group and still small. Ashwagandha shows a modest effect across 5 trials over 8 or more weeks, with rare liver injury reports, and it is not for pregnancy or thyroid disease, with no data past 3 months. Both are pharmacist conversations.

"Fix your cortisol"

Cortisol is real, and chronic stress and short sleep do shift its daily rhythm; poor sleep raises evening cortisol. But everyday stress does not push it high enough to change your face, "adrenal fatigue" does not exist (a review of 58 studies could not find it and the Endocrine Society does not recognize it), a random cortisol blood test is nearly useless because cortisol swings all day, and no supplement, powder or mocktail has ever been shown to "reset" it. Puffy in the morning and better by afternoon is salt, alcohol or sleep. Rounding that builds over weeks and stays all day, with stretch marks, bruising or rising blood pressure, is a doctor visit, and a good time to ask for thyroid and iron too.

The nightly allergy pill

Benadryl, ZzzQuil and Unisom as sleep aids: the sleep medicine society recommends against them, tolerance builds within days, and heavy long-term use of that drug class is linked to a higher dementia risk in older adults. Please do not.

What the evidence says works, for free

A fixed wake time (in 60,977 adults, regularity predicted health better than hours). The bed for sleep only, up after 20 minutes awake. Caffeine cut early: 400 mg six hours before bed cost people over an hour of sleep and they did not notice. No alcohol as a sleep aid. A warm shower an hour or two before bed. A specific to-do list before bed. That is the base of the pyramid, and it is the whole first month of my protocol.