Why Mechanical Watches Still Exist When Quartz Is More Accurate

The question "why do we have mechanical watches if quartz is just as accurate" contains a premise I need to clear before I can answer it. Quartz isn't just as accurate as mechanical. Quartz is dramatically more accurate.

A Casio F-91W — £10 at any petrol station — keeps time to within roughly thirty seconds a month. A COSC-certified chronometer, the gold standard for high-end mechanical, runs to specification at minus four to plus six seconds a day. Roll that out over a month and the £10 Casio is at least twice as accurate as the £8,000 Submariner. An atomic-synced quartz wristwatch is accurate to within a second of true time, always.

I sell mechanical watches for a living. The first thing I tell anyone walking into the question is that quartz wins on accuracy. That's not the argument.

The actual question is why mechanical watches still exist when quartz has been winning at their stated job for over fifty years. The honest answer is that mechanical watches stopped being timekeepers in December 1969.

That was the month Seiko launched the Astron, the first commercial quartz wristwatch. It cost about the same as a Toyota Corolla and was accurate to five seconds a month — three times better than the best mechanical chronometer of the era, for a fraction of the part count. Within a decade the Swiss industry collapsed. Employment in Swiss watchmaking dropped from around 90,000 in 1970 to under 30,000 by 1988. Brands that didn't pivot or die survived only by quietly making the same case: the mechanical watch had to stop competing with quartz on accuracy because it was losing.

What it became instead is the thing I'll call The Useless Heart — the mechanical movement, which is functionally inferior to quartz in every measurable way and which is, for that exact reason, the entire reason mechanical watches still exist.

The Useless Heart is a category of object. Vinyl records have one. Manual gearboxes have one. Hand-bound books have one. Fountain pens have one. The category is: a piece of older technology that survived being beaten on function by reframing what its function was.

The mechanical movement in a watch has somewhere between 130 and 400 individual parts. It runs at 4Hz, which means the balance wheel oscillates eight times a second, and you can see it doing this through the caseback if the watch has one. The energy stored in a single wind of the mainspring is roughly enough to lift a feather a metre. From that the movement runs for between 40 and 80 hours, depending on the calibre, by metering the release through a lever escapement of a design Thomas Mudge first cut in 1755.

This is — and there's no kinder way to put it — a stupid way to keep time. A quartz crystal under voltage oscillates at 32,768Hz with effectively no mechanical loss, and you can pack a movement that does this into something the size of a peanut for under a pound in parts. The mechanical solution is harder, more expensive, less accurate, less reliable in extremes of temperature and shock, requires servicing every five to seven years at a cost frequently equal to half the watch, and has been outclassed since the Nixon administration.

Which is the whole point.

The clearest place to see this argument made on the spec sheet is the Grand Seiko Spring Drive. Spring Drive is a hybrid: a mechanical movement, wound by the wearer's wrist or by a crown, with a mainspring and a gear train and a glide wheel — but the glide wheel is regulated by a quartz oscillator. The watch is mechanical where mechanical produces a visible artefact (the seconds hand glides instead of ticking, smooth as a clock in a film), and quartz where quartz produces accuracy (±1 second a day, better than any purely mechanical movement at any price).

Spring Drive is the watch where Seiko quietly admitted the answer. The mechanical parts are the product. The quartz is doing the timekeeping that the mechanical parts do worse, because the timekeeping was never what you were paying for.

This is also why a £40,000 Patek Philippe and a £40 Casio can't really be compared in the way the original question implies, any more than you could compare a Steinway and a Yamaha Clavinova by which one is harder to tune. They are not in the same category. They share a function-name and almost nothing else.

The closest civilian parallel — and the easiest way to see what's actually happening — is the vinyl record.

Vinyl is a worse delivery format for music than digital streaming. It has more noise, more distortion, less dynamic range, less convenience, costs more, takes up more space, degrades with use, and requires hardware to extract the signal. Every measurable axis of fidelity is won by digital. By every reasonable measure, vinyl should be extinct.

In 2022, US vinyl revenue overtook CD revenue for the first time in over thirty years.

The reason isn't that audiophiles secretly know vinyl sounds better. It's that vinyl became a different product. It stopped competing with digital on audio fidelity and started competing on physicality, ritual, gift-suitability, scarcity, and the visible passage of time across a side. You buy vinyl for the same reason you buy mechanical watches — because the inefficiency is the point.

The inefficiency is the point.

This is the line most "soul of horology" essays get wrong by accident. They imply mechanical watches are also good at telling time, just in a more romantic way. They aren't. They are bad at telling time and that is structurally what makes them interesting. If a mechanical movement were as accurate as quartz, it would be a worse mechanical movement, because the accuracy gap is what makes the engineering legible.

Consider what the gap actually contains. Inside a good automatic calibre there is a hairspring fewer than 0.1mm thick coiled into a flat spiral, an escapement releasing energy 28,800 times an hour, a rotor swinging on a ball-bearing race to convert wrist motion into stored tension in the mainspring, and a jewel-bearing assembly that exists because metal-on-metal at those speeds for forty years would gall. All of this — the work — is in service of an output a Casio beats with a £2 chip. The output is not the point. The work is the point. The Useless Heart is the work.

The "soul" answer to the original question — the one you hear in every Hodinkee piece since 2008 — is a reach for the right answer that lands short of it. Watches don't have soul. Wearers do. What mechanical watches have is legibility of effort. You can see what the object cost in human work, in component count, in years of horological accumulation since Mudge cut his first lever. You can't see that in a quartz movement because there is nothing to see. A quartz movement is a chip. A mechanical movement is a small kinetic sculpture you can hold in your hand and watch breathe.

There's a tradition-and-craftsmanship answer to the question too, and it isn't wrong, but it's a different argument running on top of the structural one. Yes — the Swiss industry has 500 years of accumulated technique, the Japanese have 130, the British had everything until they didn't, and that institutional knowledge is irreplaceable and worth preserving. But that's the argument for funding mechanical watchmaking. It isn't the argument for wearing the watch. The wearing argument is The Useless Heart.

There's a status answer, and people get nervous about it, but it's also true and also incomplete. A mechanical watch is one of the very few culturally permissible pieces of male jewellery left in the post-industrial west. So men spend on it. So the category sustains pricing structures the function alone could never justify. Fine. But men also spend on cars, and luxury cars are no more sustainable on status alone than watches would be — what makes a Porsche worth a Porsche is not the badge, it's that the badge stands for something legibly there in the machinery. Status as an answer is downstream of the structural one. The status sticks because the engineering is real.

Two related notes.

First — none of this means quartz watches are bad. Quartz watches are excellent at the thing watches were originally for, and there is a whole separate argument about high-end quartz (Citizen Chronomaster, Grand Seiko 9F, Bulova Precisionist) being some of the most under-appreciated pieces of mechanical-adjacent engineering on the market. A 9F Grand Seiko is built to ±10 seconds a year, which is what happens when you take quartz seriously rather than treat it as the budget option. That's a real watch. The argument here isn't that mechanical is "better." The argument is that mechanical is a different category, and the original question is asking why we have apples when oranges are juicier.

Second — there are mechanical watches that are bad mechanical watches, in the same way there are bad vinyl pressings. A polished-to-death Sub with a service movement and a third-party dial doesn't earn the premium it's asking. A microbrand running a Miyota 9015 in a thoughtfully-resolved case and dial frequently earns more than its price tag, because the work is legible, the choices are visible, and the inefficiency is doing real argumentative work. The Useless Heart only earns the premium when it's actually there. The standard against which a mechanical watch should be judged isn't quartz — it's other mechanical watches, and whether the work is on display.

So: why do we have mechanical watches if quartz is just as accurate?

We don't have mechanical watches because quartz is just as accurate. We have mechanical watches because quartz is more accurate, and the gap between what a mechanical watch costs and what it does on accuracy is the entire signal. The category survived the 1970s by reframing its function from "tells time" to "tells time the long way, where you can see the work." That reframe took twenty years, killed two-thirds of the Swiss industry, and produced the modern mechanical watch as a small kinetic object with a useless heart.

The inefficiency is the point. The watches I stock at CWC are full of useless hearts. That was ALWAYS the answer.