Chaitanya Banerjee

What I am not

I am not a doctor. I am a student on the way to becoming one, which is a different thing, and the difference matters more on a health website than almost anywhere else.

I have not completed my degree, I am not licensed to practise, I have never independently managed a patient, and I have never scored a polysomnogram or run a study. I cannot tell you what is wrong with your sleep, and I will not try, no matter how specific the question in the email is. If you want a clinical opinion about your own sleep, you need a qualified clinician, and I am not one yet.

This goes first rather than in small print at the bottom, because on a health topic it is the single most decision-relevant fact about the author. Burying it would be exactly the behaviour this site exists to avoid.

What I do have

Three years of medical school, which is not a licence but is not nothing either. It means I can read a paper properly: I know what a confidence interval is, why a prospective cohort is not a randomised trial, why a finding in 573 people is weighted differently from one in 60,000, and how physiology papers are structured. Most sleep content on the internet is written by people who cannot do that, and it shows.

It also means I know exactly where the edge of my competence is, because I spend most of my week being shown it. That is the useful part. I am not confident about sleep; I am confident about which claims I have checked.

How I got here

Matthew Walker's work is what made me pay attention to sleep as a subject rather than as a thing that happens between days. I suspect that is true of a large fraction of people currently interested in sleep science, and I do not think there is anything embarrassing about it — popular science is supposed to recruit people into the primary literature. What it is not supposed to do is replace it. So when a claim from a popular book turns up in something I am writing, I go and find the study, and occasionally the study says something narrower than the book did. That habit is most of what this site is.

The thing that turned interest into a website was smaller and more personal. For a stretch I was sleeping about eight hours a night but at wildly different times — late one night, early the next, catching up at weekends. It was, on paper, enough sleep. It felt terrible. When I later settled into roughly six hours but at genuinely consistent times, going to bed and waking within about the same half-hour every day, I felt substantially better than I had on the longer, chaotic schedule.

That is one person's experience and I want to be careful about what it does and does not license. It is not evidence that six hours is enough, and I am not recommending it — the American Academy of Sleep Medicine says adults should get seven or more hours on a regular basis, that is the guidance this site uses, and my own six-hour stretch was a symptom of a medical timetable rather than a plan. Measured properly, over six weeks in early 2026, my own number came out at seven hours, with eight noticeably better. What it did do was make a finding in the literature suddenly legible to me: a UK Biobank analysis of 60,977 people found that regularity of sleep timing predicted mortality more strongly than duration did. I had felt the shape of that before I read it, and reading it changed how I think about the whole question.

It is also why the calculator on the front page ends by telling you the schedule matters less than hitting it consistently, which is an odd note for a calculator to end on.

How I work

• Primary sources over summaries. Where a secondary source is used because the primary is paywalled or highly technical, it is named as such.
• Ranges, not midpoints. If the study reported 70–120 minutes, the article says 70–120 minutes.
• Absence of evidence gets stated. "I could not find a study showing this" is a publishable sentence here, and it appears in the calculator audit.
• AI assistance, disclosed. Articles are drafted with AI assistance against sources I have selected, then reviewed and approved by me before publication. The full policy is in the editorial standards.
• Corrections in public. Dated, logged, and never a silent edit. The log is here.

Tell me I am wrong

Genuinely. The most useful message I can receive is a citation that contradicts something published here — particularly on the sleep-cycle-alignment question, where my current position is that the central claim is unproven. If a study exists that I have missed, that page changes and the change goes in the log with your correction credited.

How to reach me →