Two Nights Is One Day: How to Count the Days You Actually Get

The sum nobody does before they book

You book two nights in a city. You think you have two days there.

You arrive at four in the afternoon. You leave at nine the next morning but one. Between those two moments you have one day.

Everyone knows this when it is pointed out. Almost nobody applies it at the planning stage, which is the only stage where it can be fixed.

Nights to usable days

Three nights is the minimum for any city you care about. Two nights is for a place you are passing through anyway. One night is a hotel with a view.

The whole-trip formula

Usable days = days away − 1 outbound − 1 return − 1 jet lag (long-haul only) − 0.5 per city change

That is it.

Worked: 14 days, four cities

14 − 1 − 1 − 1 − 1.5 = 9.5 usable days

Fourteen on the leave form. Nine and a half in reality. Across four cities that is a shade over two days each, which is why a four-stop fortnight feels tight even when it looks generous.

Worked: 10 days, three cities

10 − 1 − 1 − 1 − 1 = 7 usable days

Seven days across three cities. That works. Four cities would be 6.5 across four, which does not.

Worked: 17 days, four cities

17 − 1 − 1 − 1 − 1.5 = 12.5 usable days

Four cities at roughly three days each. This is the ratio you are aiming for, and it is why the bracketed seventeen-day trip in the PTO guide is worth building the calendar around.

Why the half-day per city change

Check-out is ten. Check-in is three. You have bags. You eat lunch somewhere that is neither where you were nor where you are going.

Even a two-hour journey costs you an afternoon, because the journey is not the cost — the day around the journey is.

Arrival time changes everything

The calendar says you arrive on day one. The clock says whether day one exists.

Two real examples from the guides in this set.

Lisbon. TP210 leaves JFK at 22:05 and lands at 10:15. UA64 leaves Newark at 20:25 and lands at 08:20. Both give you a genuine Friday. A 05:40 arrival into Lisbon on TP218 from Boston does not — you land before your hotel will let you in.

Lisbon, going home. TP209 departs Lisbon at 17:05 on Sunday and lands at JFK at 20:05. That is a Sunday you actually get. American's only Philadelphia return departs at 11:05, which deletes it.

When you compare two flights on price, compare what each one does to your day count first. A cheaper flight that lands at 22:00 has cost you a day, and a day is worth more than the fare difference on most trips.

Trains and planes cost different amounts of day

This is where the calculation goes wrong most often.

A two-and-a-half hour flight and a two-and-a-half hour train are not the same journey.

The train: walk into a city-centre station, sit down, arrive in another city centre. Add fifteen minutes at each end. Total cost: about three hours.

The flight: get to the airport, arrive two hours early for international, board, fly, wait for bags, get into the city. Total cost: five to seven hours, and psychologically a whole day.

The rule of thumb: under about four hours by rail, take the train. Over about six, take the plane. In between, look at where the stations and airports actually are.

Two examples from the routes in this set:

Kyoto to Osaka. JR Special Rapid, about 30 minutes, ¥580, city centre to city centre. This costs you a morning at most. It is the reason a Kyoto–Osaka split works on a fourteen-day trip when a Kyoto–Fukuoka split would not.

Hanoi to Da Nang. One hour twenty in the air, roughly thirty flights a day. But door to door with the airport transfers at both ends, it is most of a travel day. The Vietnam itinerary gives that leg its own day rather than pretending it is a morning.

The exception: a night that does two jobs

An overnight train or an overnight flight can cost you zero usable days, because you spend the night moving instead of sleeping in a hotel.

Hanoi to Da Nang, train SE3, departs 19:20 and arrives 10:36. You had dinner in Hanoi, you slept, you arrive in time for a Hoi An check-in. That leg is close to free in day terms.

But it only works if the arrival time is civilised. Train SE5 on the same line arrives at 01:31. That is a night in a station, not a saved day.

Check the arrival time before you decide an overnight service is clever. Half of them are.

The city-change trap

Every city you add costs you four things: a check-out, a journey, a check-in, and half a day of working out where you are.

Going from four cities to six costs you a full extra day of logistics, and it takes that day from the four places you were most excited about.

When the nights do not add up, remove a city. Do not shave a night off each one. Shaving takes every stop below the three-night floor and produces a trip where you saw nothing properly. Removing a city gives its nights to the survivors and removes a whole journey at the same time.

The 15% transit test

Add up every journey in your itinerary. Every one — the airport transfers, the trains, the internal flights.

If the total is more than 15% of your trip, the route is wrong. Not the transport. The route.

On a fourteen-day trip, 15% is about 50 hours of a 336-hour trip, or roughly two full days of moving. Past that, remove a city.

A worked itinerary, honestly counted

Here is the fifteen-day Vietnam route, counted properly.

Nine usable days out of fifteen away, plus the day on the bay. That is an honest fifteen-day trip and it is a good one. It is not fifteen days of Vietnam, and any itinerary that presents it that way is lying to you about the arithmetic.

Sample calendars

Long weekend, 1 leave day. Thursday: work, fly 22:00. Friday: land 10:15, low intensity. Saturday: full day. Sunday: morning, fly 17:05. Days away: 3. Usable: 2.5.

10 days, 3 cities, short-haul. 10 − 1 − 1 − 0 jet lag − 1 city changes = 7 usable. Split 3/2/2.

14 days, 4 cities, long-haul. 14 − 1 − 1 − 1 − 1.5 = 9.5 usable. Split 4/2/4/3 nights.

17 days, 4 cities, long-haul, bracketed. 17 − 1 − 1 − 1 − 1.5 = 12.5 usable. Split 4/3/4/3 nights with room to spare.

Common mistakes

Planning against the number on the leave form. It is always higher than the real number.

Booking one night anywhere you want to see. It is a hotel.

Assuming a 06:00 arrival gives you the day. It gives you a nap and costs you the afternoon.

Comparing flight times instead of door-to-door times. A one-hour flight is never a one-hour journey.

Booking an overnight train without checking the arrival time. 01:31 is not a saved night.

Fixing an over-full itinerary by cutting nights instead of cities. It makes every stop worse instead of making one stop disappear.

Never totalling the transit hours. Most people look at each leg on its own and never add them up.

What needs checking before this is published

• The three thresholds are Anny's rules, presented as hers. Three nights minimum, four hours for train-versus-plane, 15% for the transit test. All are judgment calls flagged in the September pack for confirmation. Confirm you stand behind all three before publishing them as the Bella-Ruta method.

• The half-day-per-city-change coefficient in the formula is a rule of thumb, not a measured figure. Same question.

• Flight times used as examples are August 2026 schedule data from aggregators, not from the airlines. They shift at the March and October schedule changes.

• Vietnam train times (SE3, SE5) are from Seat61, updated June 2026, not from Vietnam Railways.

• Kyoto–Osaka Special Rapid ¥580 and 30 minutes is from japan-guide, updated October 2025.

• Anny's call: this article is the spine of the whole set — six of the other thirteen reference it. Worth being the one you edit most carefully.