Two holidays, same hotel, same year. One closed at 100% occupancy, the other stayed at 20%. Both were a “national holiday” on the calendar. The difference was not the rate — it was the day of the week. If you price holidays as if they had a fixed value, you leave money on the table on the good ones and chase ghosts on the bad ones.
A holiday has no fixed value
The fixed-date holidays — Ohi Day (28 October), Independence Day (25 March), May Day, Assumption Day (15 August) — fall on a different day of the week every year. And that changes completely what they bring, because guests do not travel for the holiday: they travel for the long weekend the holiday creates.
| Where it falls | What it creates | Potential |
|---|---|---|
| Friday or Monday | A true long weekend — one day off attaches to the weekend | Full |
| Saturday or Sunday | Lost inside the weekend — no extra day off | As much as a weekend |
| Tuesday or Thursday | A bridge is possible with one day of leave — a smaller audience | Reduced |
| Wednesday | Cannot be bridged — a single day in the middle of the week | Minimal |
In Greece, a holiday that falls on a weekend is not moved to a working day — it is simply lost. It is moved only as an exception, by special decision, as with May Day 2024, which fell in Holy Week.
That is why the same holiday changes value from year to year. Assumption Day in 2025 fell on a Friday — a true long weekend. In 2026 it fell on a Saturday, and in 2027 it falls on a Sunday: the holiday is lost inside the weekend. If you price it with 2025 in mind, you will expect the demand of a long weekend that the calendar does not create.
The holidays that ignore the bridge
There is one important exception: the twelve days of Christmas and Holy Week with Easter are not single holidays — they are holiday periods. People book their holidays wherever 25 December falls.
Christmas 2025 fell on a Thursday — the worst possible position for a single holiday. Take a hotel open in winter that fills an ordinary December Thursday to around 20%. If Christmas closes at 90%, the rule “Thursday means a weak day” has missed by 70 points — because it was applied to a holiday period, not to a single holiday. For the same reason, on a holiday or event night a soft booking pace is not read as weakness.
The moving event and the “last year” trap
Most tools compare a date with the same day last year — in practice “minus 364 days”, so that the day of the week matches. For the flow of bookings this is right. For holidays it is wrong.
- Easter, Clean Monday, Whit Monday: they move by weeks every year.
- Local events: the Naval Battle festival of Nafpaktos (Naumachia) is held on the weekend nearest 7 October — its date changes.
- Even fixed-date holidays: Christmas minus 364 days sends you to 26 December — a different holiday.
The right way is to compare holiday with holiday, by its name: find the same event last year and read those nights. If you don’t, your “last-year reference” may be a completely ordinary day. And when there is no last-year night at all, the gap speaks for itself — see how to read the comparison with last year correctly (in Greek).
Greek holidays 2027: when they fall and what to compare them with
Orthodox Easter 2027 falls on Sunday 2 May — three weeks later than this year (12 April). Clean Monday, Good Friday, Easter Monday and Whit Monday move with it. The full list, with the last-year holiday that matches each one:
| Holiday | 2027 | Compare with (2026) |
|---|---|---|
| New Year’s Day | Fri 1 Jan | Thu 1 Jan |
| Epiphany | Wed 6 Jan | Tue 6 Jan |
| Clean Monday | Mon 15 Mar | Mon 23 Feb |
| Independence Day | Thu 25 Mar | Wed 25 Mar |
| Good Friday | Fri 30 Apr | Fri 10 Apr |
| May Day | Sat 1 May | Fri 1 May |
| Easter | Sun 2 May | Sun 12 Apr |
| Easter Monday | Mon 3 May | Mon 13 Apr |
| Whit Monday | Mon 21 Jun | Mon 1 Jun |
| Assumption Day | Sun 15 Aug | Sat 15 Aug |
| Ohi Day | Thu 28 Oct | Wed 28 Oct |
| Christmas Day | Sat 25 Dec | Fri 25 Dec |
| Boxing Day | Sun 26 Dec | Sat 26 Dec |
The third column is the comparison you want — and for none of the thirteen does “minus 364 days” land on it. For Easter it sends you to Sunday 3 May 2026, an ordinary Sunday three weeks after the holiday. For Assumption Day, to 16 August.
With the map from the start, 2027 reads quickly: two Thursday bridges (Independence Day, Ohi Day); Clean Monday and Whit Monday are Mondays every year — Whit Monday a holiday mainly for the public sector and banks; Assumption Day is lost on a Sunday; and Christmas, New Year and Easter belong to holiday periods, where the day of the week matters less.
How to measure whether a holiday is really worth it
The question is not “did it do better than last year?” — if both years were mediocre, the answer “the same” tells you nothing. The question is: did this night beat a normal, comparable night?
In practice: take the same day of the week, three weeks before and after, drop the ones that are holidays too, and compare. And measure with RevPAR, not ADR — because ADR hides the volume.
RevPAR = 45 × 118 ÷ 47 = €112.98
Typical Friday of the period (median): €96.00 → above all the comparable nights
Ohi Day, 28 October (Tuesday): 8 of 47 rooms, ADR €95
RevPAR = 8 × 95 ÷ 47 = €16.17
Typical Tuesday of the period (median): €13.20 → just above
Look at what the second example reveals. The 28th had a high ADR (€95, above the season’s €86) and low occupancy (17%). By ADR it would look like a good day; by RevPAR you see that you sold expensively to very few. And yet — compared with a normal October Tuesday, it was not a failure. The problem was not the holiday; it was October. You can only see this if you compare with the right base.
Did you know? In WisDom RM the national holidays are calculated automatically, Easter included, as it moves; local events you enter yourself, because the same festival fills one property and leaves another untouched. On a holiday night, last year is compared by the holiday’s name, not by “minus 364 days”, and a soft pace is not read as a reason to cut the rate.
Try it free for 14 days →Why the range is not enough
If you compare only with the “lowest and highest” of the comparable nights, one almost empty night swallows everything: almost every holiday will come out “within the range”. Look also at the rank (“above 4 of 6”) and the median. That way you can tell “within the range but at the bottom” from “within the range at the top”.
One observation is not a pattern
This is where the biggest mistake happens. A holiday goes badly one year, and the next year you price it low — locking in the weak result you were afraid of.
- One closed year: it is a fact, not a trend. You don’t change strategy.
- Two that agree: a pattern is starting to form. Worth a discussion.
- Two that disagree: noise. You change nothing.
If Whit Monday comes out below the normal nights one year and above them the next, you have two observations, pointing in opposite directions. The right move there is no move — and to wait for the third year.
Your instinct is data — if you measure it
No software knows that the town is preparing a big celebration this year, that a tournament with overnight stays is coming, or that the anniversary is a round number. You know it, usually months in advance — exactly when there is not a single booking to show it.
The point is not to trust your instinct blindly, nor to ignore it. It is to write it down before — what you expect and why — and to check it after, with the same criteria you would use to judge any algorithm. After two or three seasons you know when your local market is telling you the truth and when it is leading you astray. That is knowledge you cannot buy.
Conclusion
Holidays and local events are the dates with the biggest swings of the year — and at the same time the ones where the hotelier’s knowledge is worth the most. Three things make the difference: look at where the holiday falls in the week, compare holiday with holiday and not with the same calendar date, and measure with RevPAR against normal nights of the same period.
Start from this year’s list of holidays: write the day of the week next to each one. You will see at once which ones deserve attention and which are priced like normal days. And if you want to see how this connects with pricing for the whole season, see when you should cut the room rate (in Greek) or what WisDom RM does every night.
Mathematician with studies in hospitality management and years of night shifts in hotels. WisDom RM — and this blog — were born on the night shift.
LinkedIn ↗