Methodology

Distance

Distance between airports is the great-circle distance computed with the Haversine formula on a sphere of radius 6,371 km, using each airport's latitude and longitude. This is the shortest path over the earth's surface. Real flight paths are longer because of airway structure, restricted airspace, and wind routing — typically by 2–6% on long-haul sectors.

Flight duration

Duration is estimated from the great-circle distance and a typical jet cruise profile, plus a fixed allowance for taxi, climb, and descent. It is an aircraft-agnostic estimate, not a schedule lookup: it does not know your airline, aircraft type, routing, or the jet stream on the day. Eastbound transatlantic and transpacific flights are usually faster than our estimate, westbound slower, because of prevailing winds.

Time zones

Each airport carries an IANA time zone identifier (for example Europe/Stockholm). Offsets are resolved at runtime through the browser's Intl.DateTimeFormat API rather than hard-coded, so daylight saving transitions, half-hour and 45-minute offsets, and rule changes are handled by the platform's up-to-date tz database.

Jet lag severity and recovery

We first convert the clock difference into a circadian shift. The raw offset is wrapped into the range −12…+12 hours, because the body clock always takes the shorter way around the dial. Los Angeles → Tokyo is +16 hours on the clock, but the model treats it as an 8-hour phase delay (westward), not a 16-hour advance. A shift of exactly ±12 keeps its sign, so an outbound leg and its return leg are never both labelled eastward.

Let h be the absolute circadian shift in hours. Severity is h ≤ 3 mild, 4–6 moderate, h > 6 severe. Recovery uses a direction-dependent adaptation rate:

rate = 1.5 h/day westward (phase delay), 1.0 h/day eastward (phase advance)
recoveryDays = max(1, ceil(h / rate))

Those rates follow the CDC Yellow Book, which describes the body clock as shifting roughly 1 hour per day when advancing (eastward) and about 1.5 hours per day when delaying (westward), because the free-running human clock is slightly longer than 24 hours. Eight zones therefore give ceil(8 / 1.5) = 6 days westward and ceil(8 / 1) = 8 days eastward. London → New York (5 h west) is 4 days; New York → London (5 h east) is 5 days. The adjustment curve, the day-by-day plan, the hero figure, the FAQ, the return-leg comparison and the structured data all read this one value, so they cannot disagree.

The asymmetry is a population-level approximation, not a prediction for an individual: the published spread between people is wider than the east/west difference, and chronotype, sleep debt and light exposure on the day all move the number.

Every one of these steps runs in a single engine — route → circadian direction → effective shift → adaptation rate → light windows → recovery estimate — and it is covered by automated tests for ARN↔JFK, LHR↔HND, LAX↔NRT and exact 12-hour shifts, so the outbound and return legs can never contradict each other.

The day-by-day plan places light exposure relative to your estimated circadian minimum, roughly two hours before habitual wake time. Light before that point delays your clock; light after it advances your clock. Eastward plans therefore front-load morning light and block evening light, and westward plans do the reverse.

This output is an educational estimate, not medical advice. It does not account for shift work, sleep disorders, medication, or individual chronotype.

How we grade evidence

Every jet lag recommendation on this site carries an evidence label and a link to the source it comes from, so you can see how much weight to give it. We use three levels:

Timed light exposure, strategic sleep shifting, melatonin, and living on destination time are all supported countermeasures, but current guidance — including the CDC Yellow Book — stresses that the right timing depends on travel direction, the number of zones crossed, and the traveller's own circadian phase. We therefore avoid claims like "this saves one recovery day" or "full performance returns on day three": our schedules are starting points, not predictions for an individual.

Corrections

If you find an airport with the wrong coordinates or time zone, or a calculation that looks off, we want to know. More context on the project is on the about page.