Skip to content

Drawdown Percentiles · The Read

Bands read against the gate ladder.

Gates

7

Lock threshold

−40%

Adverse bands

75/90/95

How MARS uses this

Reading drawdown bands against gate state is the page's doctrine, and this distribution is the read's raw material: simulated maximum drawdowns binned over the gate zones. Live drawdown better than the bands under a compressed gate tells a different story than the same number under GREEN — the overlay of zones is what makes that visible.

How it benefits you

You get a pre-agreed definition of normal pain, decided calmly before the pain started. That single artifact prevents the two classic errors - panic de-risking inside ordinary variance, and denial when damage is genuinely abnormal. You always know which situation you are in.

GROWTHRECOVERYBUFFERFLOORDEEPLOCK−0%−10%−20%−30%−40%FREQUENCY OF MAX DRAWDOWN ACROSS 50,000 PATHS

Bands over gate zones: the same drawdown number means different things depending on which zone it lands in.

The bands already assume the brakes

Simulated drawdown is generated with gate transitions modelled in.

This is the detail that gives the comparison its teeth. The distribution is not what would happen to an unmanaged account; it is what happens to an account that compresses when the ladder tells it to. Live drawdown outside those bands therefore carries a specific meaning rather than a general one: the damage exceeded what disciplined braking predicts, which points the review at execution, exposure, and override behaviour rather than at bad luck.

Four readings

Better, inside, worse, and past the floor.

Live drawdown better than the bands suggests genuine capital control and is worth documenting rather than assuming. Inside the bands is expected behaviour and calls for holding the process. Worse than the adverse percentile bands points at over-risk, execution weakness, or a system structurally weaker than the one modelled. Worse than the lock threshold is not a reading at all, it is a state.

  • Median drawdown is not the only comparison: an account above median equity can still breach the 90th or 95th adverse band.
  • Reaching Floor, Deep-Floor, or Ground-Floor earlier or more often than the model is the over-risk signature.
  • Gate threshold breach probabilities are published per gate, so early arrival is measurable rather than impressionistic.

Evaluation timing

Gates change at cycle boundaries, never mid-cycle.

The gate is evaluated after each completed four-trade cycle, and a transition affects the next cycle rather than the one in flight. That rule removes an entire class of mid-cycle rationalisation, where a deteriorating account argues itself into a different authority band while positions are still open. It also means the drawdown that matters for authority is the one measured from equity peak high at a defined checkpoint, not the worst tick of an afternoon.

Lock is not a percentile

04

Past the floor, the model stops modelling.

A path that breaches the lock threshold stops trading and is never counted as a completion. The benchmark reports lock probability at roughly twelve percent of paths, which is close to the original anchor and is not trivial. That figure is the entire justification for gate-aware contraction: a system can be high-performance on the equity fan and still route one path in eight into a state it cannot trade out of.

Reference

The gate ladder: drawdown, brake, and what the account may still do

GateDrawdown from peakBrakeMax tier
GrowthBetter than −7%Expansion, full compoundingT7 possible
Recovery−7.1% to −13%Stability brakeT5
Buffer−13.1% to −19%Defense brakeT3 to T4
Floor−19.1% to −27%Containment brakeT2 to T3
Deep-Floor−27.1% to −34%Suppression brakeT1 to T2
Ground-Floor−34.1% to −40%Seizure brakeT1 only
System LockWorse than −40%StasisNo trading

Doctrine

Reference

Every review plots realized equity against the benchmark envelope generated from the system's own evidence. Position inside the band is classified - normal variance, outperformance, or model deviation - and only classified deviation triggers structural review. Raw P&L is never interpreted naked.

Live equity drawn inside the simulated envelope. The question is never 'am I up?' - it is 'where inside my own distribution am I running?'

ABOVE MEDIAN · NORMALTRACKING P60live equity (green) read against the simulated envelope — never in isolation

Connected inside MARS

This module doesn't work alone.

Go deeper

Operator briefs on this territory.

Every module ships in the complete MARS package.

One price. Eleven workbooks, three TradingView indicators, and the full manual library — $497.