How this instrument works
This calculator is built on one specific, peer-reviewed study: Yetish, Kaplan, Gurven, et al., 'Natural Sleep and Its Seasonal Variations in Three Pre-industrial Societies' (Current Biology, 2015). Researchers fitted continuous sleep-tracking devices on members of the Hadza (Tanzania), San (Namibia), and Tsimane (Bolivia) — three societies living without electric light or fixed work schedules — and measured, night after night, exactly when sleep actually began relative to local sunset, and how long it lasted.
The finding: average sleep onset across all three groups fell between 2.5 and 4.4 hours after sunset, with a mean of 3.3 hours, and average total sleep time ran between 5.7 and 7.1 hours, with a mean of 6.4 hours. This calculator takes a sunset time and reports that same window — earliest, latest, and recommended (mean) bedtime — plus a natural wake time derived by adding the study's own mean sleep duration to the recommended bedtime. Every number on this page traces back to those two reported figures; nothing here is extrapolated beyond what the paper measured.
This calculator deliberately does not layer a 90-minute-sleep-cycle count on top of these results, even though some other sunset-based sleep tools do. The Yetish study reports total sleep duration, not a count of discrete 90-minute cycles — combining the two would imply the cited paper established something it didn't. For cycle-based bedtime or wake-up planning anchored to a clock time instead of sunset, this site's separate sleep-cycle tools cover that ground without mixing citations.
Treat this as descriptive science, not a personal prescription. It describes averages across three specific pre-industrial populations studied over a limited number of nights each — individual variation was real even within those groups, and 'what three hunter-gatherer and forager-horticulturalist societies did on average' isn't automatically 'what's healthiest for a given person living a modern, electrically lit schedule.' It's a genuinely interesting natural baseline, not a diagnosis or a treatment plan.
- Look up local sunset time for the relevant date and location — a weather app or a quick search for 'sunset time [city]' works.
- Enter that time into Local sunset time (24h clock) — e.g. 19.5 for 7:30 pm.
- Read Earliest natural bedtime — sunset plus 2.5 hours, the low end of the study's observed range.
- Read Recommended (mean) natural bedtime — sunset plus 3.3 hours, the study's own average across all three groups.
- Read Latest natural bedtime — sunset plus 4.4 hours, the high end of the range.
- Read Natural wake time — the recommended bedtime plus 6.4 hours, the study's own mean sleep duration.
Worked example — sunset at 7:30 pm
Sunset at 19:30 (7:30 pm): the earliest natural bedtime is sunset plus 2.5 hours, 22:00 (10:00 pm); the recommended (mean) bedtime is sunset plus 3.3 hours, 22:48 (10:48 pm); and the latest natural bedtime is sunset plus 4.4 hours, 23:54 (11:54 pm) — the full window the study observed across the Hadza, San, and Tsimane. Natural wake time adds the study's mean 6.4-hour sleep duration to the recommended bedtime, landing at 05:12 (5:12 am).
Push sunset later, to 21:00 (9:00 pm, a long summer evening), and the same arithmetic carries the recommended bedtime to 00:18 (12:18 am) and wake time to 06:42 (6:42 am), crossing into the next calendar day — the calculator wraps this automatically, but it's worth noticing which side of midnight each result actually lands on.
Questions
Does this mean my 'ideal' bedtime is exactly sunset plus 3.3 hours?
No — it means that's the average across three specific pre-industrial societies studied in one 2015 paper, not a personalized measurement of any individual's biology. It's a genuine, well-sourced data point about natural sleep timing without electric light, useful as context, not a prescription that overrides how a specific person actually feels or functions on a given schedule.
Why anchor bedtime to sunset instead of a fixed clock time?
Because that's what the underlying research actually measured and found: across the Hadza, San, and Tsimane, sleep onset tracked the local light-dark cycle rather than a fixed hour on a clock, since none of these groups used clocks or electric lighting to structure their days. A fixed-clock-time sleep calculator answers a different question — 'what time did I say I'd go to bed' — than this one, which asks 'how did sleep timing relate to sunset when nothing but daylight structured the day.'
Does this calculator also count 90-minute sleep cycles, like some other sunset-based sleep tools?
No, deliberately not. The Yetish et al. 2015 study this calculator is built on reports total sleep duration in hours, not a count of discrete 90-minute cycles — layering a cycle count on top would imply the cited research established something it didn't measure. For 90-minute-cycle bedtime or wake-up planning anchored to a clock time instead of sunset, this site's separate sleep-cycle calculators cover that ground on their own citation.
Did these pre-industrial societies really sleep much more than people today?
Surprisingly, no — that's one of the study's more notable findings. Average total sleep time across the three groups was 5.7 to 7.1 hours, broadly similar to sleep durations reported in many modern industrialized populations, which pushed back against a common assumption that pre-industrial or 'natural' sleep was dramatically longer than modern sleep.
What role does modern electric light play, if this study didn't use any?
That's exactly the gap this study was designed to probe: without electric light, these three societies' sleep timing tracked sunset fairly consistently across seasons, with only modest seasonal shifts despite large changes in day length. Modern electric lighting and fixed work schedules routinely push sleep timing later and can compress sleep duration relative to this natural baseline — this calculator shows what that baseline was, not a claim that matching it is necessary for everyone.
Is sleeping on this 'natural' schedule automatically healthier?
Not necessarily, and the study doesn't claim that. It's descriptive science about how three specific populations without electric light actually slept, not a clinical recommendation. Individual sleep needs, chronotypes, and life circumstances vary widely, and a schedule that matches this natural baseline isn't guaranteed to suit every person's health, work, or family situation better than their current one.
References
- Yetish et al. — Natural Sleep and Its Seasonal Variations, 3 Pre-industrial Societies (Curr Biol, 2015)
- NIGMS — Circadian Rhythms Fact Sheet (NIH)
Read this first: This instrument computes a screening figure from population formulas — it is not a diagnosis, and it cannot see the whole picture a clinician can. Use it to inform a conversation, not to replace one.