Reading survival curves without fooling yourself.

The three numbers

Completion, lock, and timing are answers to different questions.

A survival readout leads with three headline figures. Completion probability: what share of all simulated paths reached the target within the horizon. Lock rate: what share breached the terminal drawdown boundary and stopped trading — locked paths never count as completions. And hit-week percentiles: among the paths that did complete, how early or late they got there. The first two are population-wide truths about the whole distribution. The third is conditional — a statistic computed inside the winners' club. The reference benchmark run makes the shape concrete: roughly 87.4% of 50,000 paths complete, about 12.0% lock, and the completers' hit weeks land at weeks 15, 17, and 20 for the quartiles.

The conditional trap

'Median hit week 17' is a sentence about winners only.

The trap is reading the timing figure as if it applied to your account. It doesn't — it applies to the subset of futures where the account already succeeded. If a configuration shows a 6% completion probability with a flattering median hit week, the honest sentence is: 'in the rare event this works, it works by week N' — which is a very different promise from 'expect target by week N.' The Operating Standard's front-door warning exists for exactly this: hit-week percentiles must always be read beside completion probability and lock rate, never alone. The workbook's graded verdict engine enforces the same discipline, pairing every timing read with the probability that makes it meaningful.

Lock is not a drawdown

The −40% boundary is terminal, and the arithmetic explains why.

System Lock is not a bad week — it is the end of the simulated account's trading life. Paths that breach the boundary stop, permanently, and their equity is frozen wherever it fell. The severity is arithmetic, not attitude: a 40% drawdown requires roughly a 67% gain just to return to flat, and the gate ladder above the boundary exists precisely so that deployment compresses long before recovery mathematics become hopeless. When a survival readout shows a material lock rate, it is telling you what fraction of your possible futures end in that arithmetic — which is why lock rate, not median equity, is the first number a sizing decision should consult.

The reading order

A fixed sequence that resists optimism.

Read survival output in an order that front-loads the uncomfortable numbers. Lock rate first: what share of futures end the account. Completion probability second: what share reach the goal. Adverse drawdown percentiles third: how bad the surviving journeys get. Only then the timing percentiles, explicitly framed as winners-only. The order matters because the human default is the reverse — timing first, ruin last — and by the time an optimistic reader reaches the lock rate, the narrative is already set. A fixed sequence is cheap insurance against your own preferences.

• Lock rate → completion → adverse DD percentiles → hit-week timing. In that order, every time.
• Any verdict that quotes a timing statistic without its completion probability is incomplete by definition.
• A rising lock rate across scenario runs is the earliest honest signal that sizing, not edge, is the problem.

The key idea

Survival reading is the discipline of counting the paths that aren't there.

Every flattering statistic in a survival readout stands on a population of paths that quietly excludes the dead ones. The whole skill is refusing to let that exclusion happen in your head: always ask what fraction of futures never made it into the number you're admiring. Operators who internalize this read the same workbook as everyone else and reach systematically more survivable conclusions — which, compounded over years, is the entire difference the simulation layer exists to make.