Core Metric 02 / 05
Drawdown (DD)
“Is the edge survivable?”
Drawdown is the primary survivability metric and the highest capital authority inside MARS. It is not merely a performance statistic — it is a capital-state event that directly controls how much risk the system is allowed to deploy.
The formula, derived
MARS measures drawdown from Equity Peak High (EPH) — protecting accumulated gains, not just starting capital.
How MARS reads it
Technical explanation
Drawdown from EPH routes the account into one of the capital-state gates: Growth, Recovery, Buffer, Floor, Deep-Floor, Ground-Floor, and System Lock. Each gate defines capital posture, brake tone, tier cap, and risk authority.
No lower-authority signal can override the gate. A green EV week, a beautiful setup, or a confident read cannot expand risk beyond what the drawdown state permits.
The SDE drawdown pipeline tracks current value, cumulative truth, rolling condition, stability, drift, and z-score so hidden capital pressure cannot be ignored.
Edge without survival is worthless. A positive-EV strategy that produces intolerable drawdown depth, frequency, or velocity is not deployable at that risk level.
Interpretation bands
Strong. Neutral. Weak.
Strong
Shallow, controlled, inside or better than benchmark bands, with fast recovery and low defensive-gate dwell.
Neutral
Inside the normal envelope but trending deeper or dwelling longer in Recovery/Buffer than the model expects.
Weak
Deeper than adverse percentile bands, accelerating, or approaching System Lock territory — capital defense overrides everything.
Use cases
Where it earns its place
- ▸Gate-state routing: converting drawdown depth into deterministic risk authority
- ▸Comparing live max drawdown against Monte Carlo adverse percentile bands
- ▸Detecting structurally unhealthy growth — above-median equity with worse-than-band drawdown
- ▸System Lock enforcement when drawdown breaches the hard survivability boundary
- ▸Monthly gate-dwell review: how much time capital spends in defensive states
Edge cases
Where it can mislead
- !Profitable-but-bleeding: equity can grow while drawdown expands beneath it when risk is poorly sequenced — the DD pipeline exposes this contradiction.
- !Shallow-but-constant drawdown: chronic 5–8% oscillation can be more corrosive to compounding than one clean 12% event; stability and dwell metrics catch the difference.
- !Post-peak denial: measuring from starting balance instead of EPH hides damage to accumulated gains — the exact failure EPH-anchoring prevents.
Example scenarios
The metric in the wild
Above median, worse than bands
Live equity sits at the P60 path, but max drawdown is deeper than the P25 adverse band. The benchmark doctrine is blunt: this is not clean alpha — it is probably over-risk, and the gate ladder responds accordingly.
Gate compression event
Drawdown crosses from Buffer into Floor. The tier cap drops, the pool budget contracts, and the Throttle Control Panel automatically authorizes a smaller next-cycle deployment — no willpower required.
Monte Carlo connection
The 50,000-path benchmark produces expected drawdown distributions and gate-dwell ranges. Live drawdown better than the bands suggests strong capital control; inside the bands is normal; worse than adverse bands signals over-risk, execution weakness, or structural deterioration.
Monte Carlo Lab →Live benchmark comparison
Worse-than-modeled drawdown reclassifies apparent outperformance: returns above the median with drawdown beyond the bands is flagged as risk distortion, not alpha.
7-Tier MC Benchmark →Go deeper
Operator briefs on this territory.
Deep dive — 01
Expectancy arithmetic: what a trade is worth before it happens.
The EV formula walked end to end — why per-trade worth beats realized P&L as the master metric.
Read the full brief →
Deep dive — 02
Reading the fan: Monte Carlo without self-deception.
What the percentile bands actually promise, why P10 governs sizing, and where simulation authority ends.
Read the full brief →
Deep dive — 03
MAE/MFE: what your trades did while you weren't looking.
Reading maximum adverse and favorable excursion to audit stops, targets, and exits with evidence.
Read the full brief →
DD is calculated for you — automatically.
Every reading on this page is produced, tracked, and interpreted inside the MARS workbook ecosystem.