Understanding Drawdown: How Deep May the Abyss Be?

The previous letter surveyed the profit factor — the number that says whether a system wins. This letter surveys the number that says what you endure for it. It bears a name that sounds as if a poet invented it and an accountant adopted it: drawdown. The depth of the abyss.

The definition

Picture a system’s equity curve as a mountain hike: sometimes up, sometimes down, in the good case upwards over time. The drawdown answers a single question: how far below the last summit are you standing right now?

It is measured from the curve’s last peak, expressed as a percentage of that peak. If the curve reaches a high of 100 and falls to 92, the drawdown is 8 %. If it climbs to 95, the drawdown has shrunk to 5 % — it disappears only when the old high is exceeded. Until then the system is, as the jargon has it, under water.

From this running quantity derives the metric no honest history may lack: the maximum drawdown — the deepest point below the highest summit ever reached. It is the protocol of the worst moment: that place in the story where an observer with perfect mistiming would have entered, and despaired out again at the low.

The cruel mathematics of recovery

Now to the property of drawdown that almost everyone underestimates, because our intuition takes percentages to be symmetric. They are not.

Whoever loses 20 % needs not 20 % of profit to recover but 25 — for the gain is calculated from the shrunken capital. Whoever loses 50 % needs 100 %. Whoever loses 80 % needs 400 %. The abyss grows exponentially more expensive towards the bottom: the second half of the fall costs a multiple of the first.

From this follows an insight that could hardly be more sober and yet sounds like a maxim of the old schools: it is easier not to fall deep than to climb deep. Limiting risk is not caution born of timidity — it is arithmetic. A system that keeps its slumps shallow owes the future less than one that needs a resurrection after every intoxication.

The time under water

Depth is only half the survey of the abyss. The other half is duration: the time from summit until the summit is reached again — the underwater time.

It is the psychologically crueller of the two quantities. A deep, fast slump is a blow — it hurts and it is over. A long underwater time is a siege: months in which every balance sheet says the same thing — still not back. It is in this siege that most operators of systems fall over, and not at the deepest point but somewhere halfway through the recovery, when the panic has long grown quieter and exhaustion negotiates instead: maybe it simply does not work anymore. Whoever evaluates systems should therefore demand both numbers: how deep — and how long.

What depth reveals about a system

In the shop window, drawdown is treated like a disease: concealed, smoothed, cropped. Yet in the Lexicon of the Abyss the reverse rule applies — a history without visible drawdown is the warning sign, not the one with it. Real curves have dents. Curves without dents are short, selected or doctored; in all three cases they prove nothing. It is the same logic that makes loss days sacred: only the documented wound makes the chronicle credible.

Together with the profit factor, drawdown yields the most honest double portrait that two numbers can draw: one measures the ratio of light to shadow, the other the depth of the darkest place. Both work without money amounts — percentages suffice, the ratio carries the truth. That is exactly why both belong to the fixed inventory of this documentation and its FLEX principle.

In the machine’s ongoing trial, the maximum drawdown is published continuously. The trial knows a hard limit here: if a single day exceeds a defined loss depth, the trial counts as failed — no matter how good the other numbers look. That is not strictness for the sake of the gesture. It is this letter translated into a rule: a system that falls deep enough has not almost passed. It has found the abyss it claimed to know.

So the question in the title — how deep may the abyss be? — has two answers. The arithmetical one: shallow enough that recovery remains affordable. And the chronicler’s one: as deep as it actually was. It will be shown at full depth.