Automatic Movements, Explained From the 1755 Lever Escapement Forward

Every automatic watch sold today relies on a mechanism that reached its modern form in 1931, when Rolex mounted the winding weight on a central axis so it could turn a full circle: a rotor that swings with the wearer's arm and winds the mainspring through a set of reduction gears. The idea is older than that. Abraham-Louis Perrelet is credited with the first self-winding mechanism around 1777, in a pocket watch driven by the motion of walking rather than the wrist, and John Harwood patented the first self-winding wristwatch in 1923. Before any of them, watches were wound by hand, and the entire selling point of "automatic" was solving one specific, mundane problem: owners forgetting to wind their watch and having it stop.

That's worth sitting with, because a lot of marketing language around automatic movements implies the self-winding rotor is a mark of horological sophistication in itself. It isn't. It is a convenience feature, elegantly solved a long time ago, that happens to require genuine mechanical engineering to execute well. The actual sophistication in a modern automatic movement lives elsewhere. It lives in the escapement, the mechanism that releases the mainspring's stored energy in tiny, regular increments, which is what makes a watch tick at a steady rate instead of unwinding all at once. That is the part with the 1755 pedigree: Thomas Mudge's detached lever escapement, still the basis of the escapement in most mechanical watches made since. It lives in the balance wheel's regulation, and in finishing quality that affects longevity and serviceability more than day-to-day accuracy.

Here's where the honest accuracy conversation happens. A well-regulated mechanical watch, at any price point above roughly $500, typically runs within -5 to +15 seconds a day, a spec that hasn't meaningfully changed in decades regardless of how much more the watch above that price costs. A $50,000 watch is not going to keep dramatically better time than a $2,000 one; both are mechanical devices subject to the same physical constraints of gravity, temperature, and the balance wheel's oscillation rate. If pure timekeeping accuracy is the actual goal, a $30 quartz watch beats every mechanical watch ever made, automatic or not, because a quartz crystal's oscillation is orders of magnitude more stable than any spring-and-gear system.

So what does the money above $2,000 actually buy? Mostly three things: finishing (hand-beveled bridges, polished screw heads, decorative engraving visible through a display caseback, none of which affects function and all of which affects how the movement looks and how much skilled labor went into it), complexity (additional functions like a chronograph, moonphase, or perpetual calendar, each of which is genuinely harder to engineer and assemble), and brand heritage, which is a real cost but not a mechanical one.

Where spending more stops buying more accuracy: almost immediately, once you're past a competently regulated movement. Where it keeps buying something real: finishing, complication count, and the accumulated expertise of houses that have been solving these specific mechanical problems for over a century. Buy for those reasons, honestly stated, rather than for a timekeeping improvement the mechanism was never going to deliver.