Operators ask how many machines they can afford. It is the wrong question, and asking it produces the fleet shapes that come apart in soft quarters.

The binding constraint is not coin. An operator with a million coins can declare a great many chassis and pay the eight percent without difficulty.

The constraint is the market for what those machines produce, and it tightens with every machine you add — including the ones you add yourself.

What Adding a Machine Actually Does

It commits coin at declaration, permanently fixes a capability set, and adds supply to a band. The first two are obvious and the third is the one operators skip.

Your own machines compete with each other. A fleet of eight chassis producing into the same band is eight sources of supply pressing on the same demand, and the eighth one's output clears at whatever the presence of the other seven has left. This is the Great Overproduction in miniature, run inside a single fleet, and it does not require four hundred operators to happen.

Sizing Against the Band, Not the Balance

The workable method is to size each output band separately and ignore the fleet total entirely.

For a given band, establish what it has cleared across two full quarters — not one, two, for the reasons the Overproduction demonstrated. Establish roughly what your existing machines in that band contribute. Then ask whether the band absorbs your marginal addition without moving.

Concretely: a band clearing 14 to 17 coins across two quarters, into which your three existing chassis deliver a steady contribution. A fourth machine declared at 26,000 costs 2,080 in creation tax and must clear its output within that band to justify itself. If your three machines are already a visible share of the band's supply, the fourth will not clear at 17. It will clear at 14, and it will pull the other three down with it.

"Count your machines by what they make, not by how many you have. Eleven chassis across five bands is a fleet. Eleven chassis across one band is a single very expensive machine with a redundancy problem." — Dara Voss, Sinter Yards

The practical ceiling for most operators is lower than they expect. In a band with several active producers, three or four chassis is frequently the point at which marginal output starts depressing the clearing price. Beyond that the correct move is not another machine in the same band but a machine in a different one — which costs the same eight percent and does not cannibalise.

The exception is a band you dominate, typically because you originated a compound and competitors have not yet paid to follow. There, additional capacity captures premium rather than diluting it — but only for the length of the interval, and the interval is set by the competitor's declaration cost rather than by anything you control.

Why Operators Oversize Anyway

Because expansion is legible and restraint is not. A fleet that grew from four to eleven is a visible accomplishment; a fleet that stayed at four because the band would not absorb more is indistinguishable, from outside, from an operator who could not afford to expand.

The second reason is that the cost arrives late. Creation tax is paid immediately and the dilution shows up cycles later, in a clearing price that fell for reasons no single settlement attributes to you. By the time the fleet is visibly too large, the eight percent is spent and the tags are fixed.

What Operators Get Wrong About Size

The first error is treating fleet size as a measure of standing. It measures declared capacity, which is not the same as earnings and is emphatically not the same as resilience. The registries are full of large fleets producing into bands that stopped clearing.

The second error is believing an oversized fleet can be trimmed. It cannot, cheaply. Excess chassis exit at three percent to destroy, sixty percent at the Annex, or a thin used-machine market where the buyer's alternative is building fresh. An operator who oversizes has made a decision they will be living with, and the cheapest of those exits still returns nothing.

Size the band and let the fleet total be whatever it turns out to be. It will look conservative in every good cycle and it will be the reason you are still declaring in the quarter after a soft one, which is the only comparison that ends up mattering.

Note: Mechadia is a work of fiction. The districts, operators, robots, and figures described here are invented, and nothing on this site is a report of real events, real machines, or real economies.