On 15 July I sat down and tried to buy the night of 16 January 2027 in two alpine markets. Same afternoon, same night, same entry category, same booking channels a guest would use.
At St. Moritz, of the five houses I read, four returned a price for that night on its own. One was closed for the period.
At Courchevel, of the six houses I read, five returned a price only inside a minimum stay. Three nights at one property, four at two others, five at another, seven at the last. One returned no availability at all.
Two hundred kilometres apart, in the same week of the same season, one market was selling a night and the other was selling a stay.
I have been sitting with that for six weeks, because the more I look at it the less it behaves like a curiosity and the more it behaves like a problem for anybody who has to put a number on one of those buildings.
Start with what it is not. It is not a demand signal. A house that will only take seven nights over that date may be full, or may have decided months earlier that it does not want short stays in that week, or may be holding the period for a tour operator. From a booking screen those three are indistinguishable, and I want to be exact that I cannot tell them apart. What is dated and checkable is narrower: on 15 July, at those properties, through the channel a guest would use, the night could not be bought on its own.
The consequence survives the ambiguity, which is why it is worth writing down.
Almost every method of valuing a seasonal hotel runs a rate against an occupancy, and both are expressed per night, because that is the unit the industry has always used. In the second market that unit was still on the screen. Every one of those five houses displayed a per-night figure. What was not available was the ability to buy one.
That distinction gets lost quickly. A per-night number attached to a week that only sells as a week is a number that describes an accounting convention rather than a transaction. It is not wrong. It is a different kind of fact from a rate somebody paid for a single night, and the two arrive on a page looking identical.
There is a second thing in those readings that I keep returning to. The five gated houses did not agree with each other. The minimum ranged from three nights to seven. Whatever is driving that decision is being taken inside each building, and it is not visible from outside. Read side by side, five houses in one small market gave five different answers to the same question on the same afternoon. An average across them would have reported none of that, and would have looked like a market.
That is the part that should interest anyone reading a file about a property in a place like this. Not that the number is wrong. That the number is a single value standing in for a set of decisions that did not agree.
On the record — St. Moritz and Courchevel, August 2026: Confidential Markets holds that at the top of the alpine winter these two markets sell in structurally different units, and that the minimum-stay decision is taken property by property rather than market by market. The specific prediction we will be judged on: through the October readings, on our fixed panels, we expect the Courchevel panel to keep showing a spread of at least three nights between the shortest and the longest minimum on the same peak date, while the St. Moritz panel continues to price that date as a single night at the majority of houses. We are clear about what would falsify it. If by the end of October the Courchevel panel converges on a common minimum, or St. Moritz starts gating the same dates, then what we recorded in July was a booking-window effect and our reading of it as structural is wrong, and we will record it as wrong. The panels are fixed, the dates are fixed, the readings are weekly.
None of this asks anyone to distrust a number they have been given. It asks one question of an alpine file, and it takes a sentence. For the weeks that carry the season, was this rate transacted for a night, or was it derived from something that could only be bought as a stay.
—
The next Field Note continues the series.
Confidential Markets is independent research on the structural and cyclical risk of Europe's scarce ultra-luxury hospitality micro-markets. The market, before the asset.