Synopsis.
Name
The Brittle Optimum. Your system has slack in it: a spare supplier, a buffer of stock, a margin on the schedule, a second way to do the thing if the first fails. In calm times all of it just sits there, costing money and earning nothing, so you trim it: cut the spare, run the buffer down, tighten the schedule, drop the backup, and every cut makes the system leaner and cheaper, and nothing breaks, so you cut again. The system reaches its optimum, lean and tight and admired for it ... and then a shock comes, as shocks do, and there is no give left to take it, so what a looser system would have shrugged off tears straight through, one failure hitting the next with nothing between them, and the leanest system of all is the one that shatters.
Arc
The bend is this: the slack that looks like waste is the give that absorbs a shock, so optimising it away buys efficiency now and fragility later, and the leaner the system the harder it breaks. It runs on a trade the calm hides. Slack is redundancy, buffer, spare capacity, a margin of safety: things that, on any ordinary day, do nothing but cost. Each slice of it is locally wasteful, so trimming any one slice is a real gain, visible on the books at once, and the system does not break when you cut it, because on an ordinary day the slack was not needed. So the reward for cutting is immediate and certain, and the cost of having cut is deferred and invisible, waiting on a shock that has not come. And the cutting tightens the system. Slack is not only spare, it is give: the looseness that lets one part fail without the next feeling it. Take the give out and the parts couple tight, so a jolt in one runs straight to the others with nothing to damp it, and the same optimisation that raised the efficiency raised the coupling, until the system sits at a lean, taut optimum where every part depends on every other holding. None of it is waste or folly. Cutting idle slack is the job, each cut is a true saving, the slack really was doing nothing on the day it was cut, and no one can say which shock will come or when. The optimiser is praised for every trim and never charged for the fragility, because the fragility is invisible until the day it is not, and on that day it looks like bad luck, not like the bill for the cutting. Keep the slack, hold the spare and the buffer and the margin as insurance against the shock you cannot foresee, and the system can flex instead of shatter: but the reserve costs on every ordinary day, underperforms the leaner rival through all the calm, cannot prove its worth because the shock it absorbs leaves no mark, and can itself be overdone into mere bloat, so robustness is bought with a standing inefficiency that always looks, right up until the shock, like the very waste the optimiser was right to cut.
Stakes
Whether a system can hold the give that saves it, or must always trim its way to the optimum that breaks. Every plot but the last moves the same way: toward the cut, which each time is a real saving and harms nothing, and away from the slack, which costs on every ordinary day and earns only on the rare one no one can predict. The production cannot make the slack pay in calm, because slack is precisely the capacity that does nothing until the shock, so its whole value is invisible exactly when the decision to keep or cut it is made. It can only show that no one was wasteful or wrong, that cutting idle slack is sound and each trim a true gain and the shock genuinely unforeseeable, and that the one thing that would let the system flex, a reserve of slack held through all the calm, must be paid for every day it is not needed and can never prove the shatter it prevented, so the efficiency is real and the fragility is real and the optimum is the point where the two cross, one cut past safe.
Estate
khai-misfits: the house that holds this production and answers for the whole run.