The most expensive watches on earth are not the best watches on earth

We scored 1,605 watches from 56 brands on four specifications their own manufacturers publish, with the weights written down before we looked. Across the entire market, paying more predicts a lower score, not a higher one. The most expensive watch in the set costs $1.7 million and finishes 1,333th.

Every “best watch” list is a taste with a price tag next to it. That is not a complaint about taste — taste is most of why anyone buys a watch. It is a complaint about the arithmetic. Nobody publishes what they weighted, and a ranking whose weights you cannot see is a ranking you cannot disagree with.

So we did it the other way round. Pick specifications that the manufacturer states in public. Write the weights down first. Apply them identically to every watch an authorised dealer will actually sell you, and let the answer be whatever it is.

The ruler

Four characteristics, chosen under one constraint that turned out to matter more than any other: you have to be able to point at it on a photograph of the watch. That rules out everything that is really a brand impression wearing a number.

Percentiles use the midrank convention, which sounds like pedantry and is not. Water resistance takes only a handful of distinct values — 50, 100, 200, 300 metres and a few outliers — so without it, two watches with byte-identical specifications would score differently depending on where they happened to sit in the file. A ranking that depends on file order is not a ranking.

Price is deliberately not an input. The whole question is what the price tag knows about the watch, and you cannot answer that with a score that has the price baked into it.

What it found

Sort all 1,605 watches into ten equal groups by price and take each group’s mean score. If price bought capability, this column would climb.

The top price decile — median $89,000 — averages 36.1. The decile around $9,550 averages 57.7. That is not a rounding artefact and it is not a small number of weird outliers; it is 161 watches against 161 watches.

The mechanism is not mysterious once you see it. Very expensive watches are disproportionately dress watches: gold cases, 30 metres of water resistance, one complication or none, because the buyer is not going diving. Gold is soft. Thirty metres means “survives rain”. The specifications go down as the price goes up because, past a certain point, the price stops being about the specifications at all.

What the ruler does not measure — and where that bites

This is the part most rankings leave out, so: the score does not measure chiming complications, hand-finishing, decoration, provenance, resale value, or accuracy in service. Those are real, and several of them are exactly what a $400,000 watch is sold on.

It bites hardest on the watch at the top of that table. A grande sonnerie is the most complicated thing in mechanical watchmaking — it chimes the hours and quarters on its own — and our source has no field for it, so the Greubel Forsey scores zero on complications. Everything the ranking says about it is true on the axes we measure. Its 30 metres of water resistance is a published fact. But it is being scored on four things, and the thing it exists for is not one of them. That is a limitation of the ruler, stated out loud, rather than a discovery about the watch.

The aggregate finding does not depend on those six watches. Drop every watch over $200,000 and ρ is still −0.136; drop everything over $100,000 and it is −0.089. It is the ninth and tenth deciles as a whole doing the work, not a handful of exceptional objects.

But here is the number that cuts the other way, and it belongs in the article rather than in a footnote: below about $50,000 the relationship disappears. Restricted to watches under that price, ρ is +0.023 — flat. So the honest claim is not “cheaper watches are better”. It is narrower and stranger: across most of the market price tells you almost nothing about the specifications, and at the very top it starts telling you the opposite.

The top of the table

The winner is the Panerai Submersible Carbotech (PAM01616), at $19,900 — 87.1 out of 100, 2.41σ above the field mean of 50. It earned it on 28.1 points of water resistance (300 m), 24.7 on case durability (Carbotech, a carbon composite harder than steel and lighter than titanium), 18.3 on complications and 16.1 on movement.

Two honest notes about that result. First, #1 and #2 are the same watch in two trims and they tie exactly at 87.1; the tiebreak is declared in advance — when scores tie, the cheaper one ranks higher. Second, and more interesting: #3 is a $3,350 Longines, 86.02, within 1.12 points of a watch costing six times as much. Seventh place is a $1,150 Tissot.

Three watches you know

None of this says the Patek is a bad object, and none of it says you should buy the Casio. It says something narrower and, we think, more useful: the price of a watch is not a summary of the watch. It is a summary of the material, the name and the demand. If you want to know what you are getting, you have to look at what you are getting.

Get the data

All 1,605 watches, scored and ranked, with every input column and every percentile, plus the methodology note: what was dropped, why case durability is graded on wear rather than on cost, and how to re-grade it yourself if you disagree.

Source: manufacturer specifications and US list prices as published by Watches of Switzerland (US), pulled 2026-08-16. 4,022 products → 1,605 distinct models after dropping any watch missing one of the four measured characteristics and collapsing colourway variants of the same model to their cheapest example.