Why this gives a range
Every flight time calculator I could find returns a single number, and a single number cannot be right for a route flown in both directions. London to New York is scheduled at seven hours eastbound and eight westbound. That is not rounding — it is the jet stream, a band of wind running west to east across the North Atlantic and the North Pacific at speeds that routinely exceed 200 km/h at cruising altitude.
Measured across six long east-west routes, the westbound penalty averages about ten per cent, and it is largest on the transatlantic legs. So this returns a range with the direction applied. A range is less satisfying than a number and it is the honest answer.
On north-south routes the penalty nearly vanishes, because you are crossing the wind rather than fighting it. The estimate scales the penalty by how much of the trip is actually east-west, which is why Johannesburg to Cairo comes back much tighter than London to Los Angeles.
The 63 minutes nobody adds
The other systematic error is subtler. Take a distance, divide by a cruise speed, and you have the time spent at cruising altitude — which is not the number anyone wants. It leaves out pushback, taxi-out, the climb, the descent, the approach and the taxi to the gate at the other end.
Fitted against fifteen real scheduled block times spanning 450 to 12,000 kilometres, the fixed overhead comes to about 63 minutes, and the model that results — 1.05 hours plus distance over 902 — lands within an average of thirteen minutes of the published schedule.
That overhead barely matters on a twelve-hour flight and completely dominates a short one. A calculator without it under-predicts short haul by twenty to forty-five minutes, which decides whether you make a connection.
Great-circle distance is not the distance flown
Aircraft do not fly great circles. They fly published airways, they route around restricted airspace and weather, and on oceanic crossings they follow tracks that are repositioned daily to sit in or out of the jet stream.
ICAO publishes a correction for exactly this, used in its own carbon calculator: add 50 km below 550 km, 100 km between 550 and 5,500 km, and 125 km above that. It is applied here and shown as its own line, so you can see what the raw great-circle figure was.
Things people ask
Why does the same route take longer one way?
The jet stream. It runs west to east, so westbound flights fight it and eastbound flights ride it. On the North Atlantic that is routinely 45 to 60 minutes on an otherwise identical flight.
Is this the scheduled time?
It is an estimate of it. Airlines pad schedules for reliability and the padding varies by carrier and season, so treat this as a close approximation rather than a timetable.
Why is the distance larger than other sites show?
Because the ICAO routing correction is applied. Other sites usually show the raw great-circle distance, which no aircraft actually flies.