What a sham control is, and why grounding research mostly doesn't have one

If you read one thing about how to judge a wellness claim, make it this. It applies far beyond grounding.

The problem a control group solves

Give thirty people a thing and ask whether they feel better, and a good number will say yes. That's not people lying. Expectation genuinely changes what people report and, for some outcomes, what they experience.

The effect is strongest for exactly what grounding studies measure: sleep quality, pain, mood, energy. These are self-reported, subjective, and highly responsive to attention.

A control group is the comparison that tells you whether the thing did anything, or whether having a thing did.

Why "no treatment" isn't good enough

The obvious control is a group that gets nothing. It's better than no control and still weak, because the two groups now differ in more than the treatment.

The treated group knows they're being treated. They've been given equipment, they're being observed, they're paying attention to their sleep in a way they weren't before. Any of those could produce the result.

What a sham control actually is

A sham is a fake version, indistinguishable from the real one.

For a grounding sheet: identical cotton, identical stitching, identical cord and connector — with the conductive path broken somewhere invisible. It looks the same, feels the same, plugs in the same way, and does nothing.

If the grounded group improves more than the sham group, expectation is no longer a plausible explanation.

Why it's hard here

Three practical problems.

Checking the connection. Anyone with a basic meter can verify a grounding sheet is connected in seconds. In a study that's a liability — anyone motivated to check can unblind themselves.

Sensation. Some people report feeling something on contact. If real and sham differ perceptibly at all, blinding leaks.

Cost and honesty. A sham requires manufacturing a product designed not to work, and it requires telling participants they might receive it. Both are more effort than running a single-group study, which is part of why most early grounding work didn't.

Where grounding research actually stands

Most of the literature is unblinded. Participants knew.

That doesn't make the findings false. It means expectation is doing an unknown amount of the work, and there's no way to tell how much from the study itself.

The exception matters. A 2025 randomised, double-blind, placebo-controlled trial used an identical sham mat, ran 31 days with 60 participants, and measured sleep with actigraphy alongside questionnaires.

That last part is important. Actigraphy is a wearable measurement — it doesn't care what the participant believed. When a subjective measure and an objective one point the same direction in a sham-controlled trial, that's meaningfully stronger evidence than anything grounding had before.

It's the sole reason our library grades sleep Promising, and everything else lower.

What one good trial is and isn't

It is: the best evidence in this field, from a group independent of the researchers who produced most of the earlier work.

It isn't: replicated. Or long — 31 days. Or large — 60 people. Or about a fitted sheet, since it studied a mat used about six hours a day, so we're extrapolating across products.

One good trial is a reason to run more. It isn't a conclusion.

How to use this when you read anything

Next time you meet a wellness claim, ask three questions in this order:

• Was there a control group? If no, you're reading about expectation as much as about the intervention.
• Was it a sham, or just no-treatment? No-treatment controls don't separate the thing from the attention.
• Did participants know which group they were in? If yes, subjective outcomes carry much less weight.

Most wellness claims fail at question one. Grounding mostly fails at question three — with the one recent exception above.

Why we're telling you this

Because it's the strongest argument against our own product, and you'd find it eventually.

We think a well-made sheet with an inline safety resistor and ninety nights to decide is worth buying on its own terms. We don't think you should buy it because a study of twelve unblinded people reported better sleep.

Everything that weakens the case is here, collected in one place.

Related: How to read a small study without overreading it · What we don't know yet · The research library

These statements have not been evaluated by the Food and Drug Administration. This product is not intended to diagnose, treat, cure, or prevent any disease.