Tart Cherry Juice: The 84-Minute Sleep Result Came From Eight People. The Better Evidence Is for Something Else.

The sleepy girl mocktail is tart cherry juice, magnesium and something fizzy. Fifteen videos about it have collected over 1.8 million views, and it has quietly become the default answer to “what should I drink before bed”.

The number that powers it is 84 minutes — the extra sleep measured in a study, on a proper sleep lab machine rather than a questionnaire.

That study finished with eight people.

What the sleep evidence actually is

There are two studies. Not two dozen. Two.

The first randomised 11 people, then excluded three of them because they turned out to have sleep apnea rather than insomnia. The remaining eight completers slept 84 minutes longer on cherry juice than on placebo, measured by polysomnography, and their sleep efficiency improved on one questionnaire. On the other questionnaires, nothing. [1]

The second was a crossover trial in 15 older adults with chronic insomnia. It found a real reduction in one measure — minutes spent awake after falling asleep — and then, against placebo, no improvement in how quickly people fell asleep, how long they slept in total, or their sleep efficiency. The authors describe their own effect sizes as “moderate and in some cases negligible”. [2]

Those same authors then wrote the sentence that should travel with every mocktail video and never does. The effects, they say, were “considerably less than those for evidence-based treatments of insomnia: hypnotic agents and cognitive-behavioral therapies for insomnia”. [2]

Cognitive behavioural therapy for insomnia is the first-line treatment, it is not a drug, and it beats the juice comfortably. If you have real insomnia, that is the thing to ask about.

The three people who were excluded

Worth pausing on, because it is the most practically useful detail in either paper.

Of eleven people who signed up believing they had insomnia, three had sleep apnea and were referred elsewhere. [1] Not insomnia. A different condition, with a different treatment, that no beverage is going to touch.

If you have been fighting your sleep for months and self-treating with drinks and supplements, that ratio is worth knowing about.

Now the part almost nobody buys it for

Here is the inversion. The better evidence for tart cherry is not about sleep at all.

A meta-analysis pooled 14 studies on recovery from hard training. Muscle strength recovered better with an effect size of 0.78 — on the usual scale, 0.2 is small, 0.5 is moderate and 0.8 is large, so that is at the top of moderate. Power recovered better too, and a subgroup looking specifically at jump height came out at 0.82, which is genuinely large. Soreness improved slightly. [3]

A second, larger review in 2026 pooled 19 trials and broadly agreed on strength: maximal voluntary contraction recovered better at every time point measured. [4]

But it disagreed with the first review on soreness, finding no significant pooled effect there, and none for creatine kinase, two inflammatory markers, or range of motion. Its heterogeneity was severe — between 69% and 93% — meaning the studies were not really measuring the same thing. Its sensitivity analysis found that significance at several time points depended on individual studies. And it graded its own certainty as very low to moderate, calling for cautious interpretation.

So even the good news comes with a wobble. What survives both reviews is strength and power recovery. What does not survive is the soreness claim, which is the one people would actually notice.

So should you drink it?

If you are an athlete with another hard session in 48 hours, tart cherry is one of the more defensible things in the recovery aisle. Two independent reviews agree that strength comes back faster, and there is no safety concern — it is fruit juice.

If you are drinking it at 10pm hoping to sleep like the videos promise, the honest position is that two pilot studies in 23 people between them is not enough to tell you what it does, and the researchers who ran the bigger one said outright that proper insomnia treatments work considerably better.

It is also worth knowing what else is in the glass. The mocktail’s other ingredient is magnesium, and we checked that too: magnesium lowers blood pressure and does nothing for cramps, with sleep evidence thinner than the marketing, and the glycinate form is not better absorbed than the cheap one whatever the label says.

The verdict

Preliminary — for sleep. Something is happening, there is a plausible mechanism involving tryptophan, and both small trials moved at least one measure. But 23 people across two pilots, with several outcomes inside each study showing nothing, is not a basis for telling you it works.

This is exactly what Preliminary is for. Not debunked, not established — genuinely unresolved, and nobody has yet run the trial that would settle it.

The recovery evidence is better, from two meta-analyses rather than two pilots, though it too disagrees with itself on soreness. That is not what this page is rated on, because it is not why anyone is buying the juice.

Related: melatonin, and the dose nobody needs, what sleep hygiene advice actually achieves, and beta-alanine if the interest is performance rather than sleep.

If you want to know what earns each rating, we wrote down how we read a study, and everything we have checked is in one place.