Sleep debt: why you cannot feel it, and how to measure it
I go to bed late by choice and I do not feel tired. The laboratory says that is exactly what people feel while their alertness collapses.

Founder of Helix Health
Théo Sans grew up in Port-Leucate, in a modest family. Early on he understood one thing: in a world where systems dictate reality, whoever masters how the parts connect gets the big picture and takes back control.
Top of his class in Communication, he moved into electronics engineering out of necessity: to stop depending on other people's tools. He gathered up to 26,000 people around his online communities before building Helix alone, at 19.
He writes here about what he documents while building the product: molecule interactions, multi-domain scoring, health data sovereignty, and designing communities that do not monetise attention.
What he works on
More context on the mission: the About page.
I go to bed late by choice and I do not feel tired. The laboratory says that is exactly what people feel while their alertness collapses.
I will not give my HRV number. Not out of modesty: because compared with yours, it means nothing. Compared with mine from yesterday, it says a lot.
Three months of tablets, a ferritin that has not moved, and the conclusion that "iron does not work". Most of the time, it is the intake that does not work.
A level that drops with age, mice that live longer, a headline in Science. Two years later, the level does not drop. The story of a biomarker, and of what is left when it falls.
At first, nothing. Then, months later, the drawer opens faster. Bacopa is the only nootropic whose literature clearly says to wait, and for how long.
The day creatine gets noticed is the morning after a short night. Two recent trials say why, and at what dose. It is not the one on the box.
Three months without a break, without knowing. A lead-heavy calm, emotions switched off, and no desire left to build anything. What the trials say, and what they do not.
Sleep came. My mood sank week after week. Three days after stopping, it was back. What the literature says about what I saw.
The international standard requires kefir to carry at least 10,000 live yeasts per gram. Yoghurt requires none. Everything else follows from that.
Across 1,239 extra virgin olive oils, phenolic content ranges from 1 to 2,850 mg/kg. The label does not tell you where yours sits.
Sleep regularity predicts mortality better than sleep duration. Which moves the most repeated piece of advice of them all.
Twenty-five euros for 90 capsules of 125 mg. On a fifty-euro monthly supplement budget, that is not one more line, it is a trade-off.
Five years without those patches moving. Since I started taking zinc, they are filling in. It is not what I came for.
Nothing happens at the time. No drowsiness, nothing resembling a sleeping pill. The difference shows up the next morning.
HDL went up, triglycerides came down, LDL too. The trial was stopped early for lack of efficacy.
My father is 62. He was forgetting details I had just repeated to him, and he had become slow. I thought it was early dementia.
In one dream I read the text on a classmate's screen. Not the impression of text: the text. Here is what that is worth, and what it is not.
A single annual dose of 500,000 IU increased falls by 15% and fractures by 26%. The trial was looking for the opposite.
Hundreds of trials, five decades of hindsight. That is exactly why it gets made to say anything. Sorted by strength of evidence.
A «magnesium 500 mg» capsule can deliver as little as 70 mg of magnesium. And the most concentrated form is the worst absorbed.
Two methods go by the name biological age. The expensive one is not the more predictive, and the founding study says so.
Three weeks, then life takes over. We blame willpower. The data point elsewhere, and the number is bigger than you would guess.
In March 2025 a company holding millions of customers' DNA filed for bankruptcy. The right to erasure had existed all along. It had to have been used first.
A validated holistic health score already exists. It contains no HRV, no recovery score, no sleep stages. And it cuts stroke risk by two thirds.
An earlier version of this article told you to stop omega 3 before surgery. A trial on 1,516 patients says otherwise.
A question, a correction, a topic suggestion?