Mechanism: what percentage drawdown measures
Percentage drawdown is a risk metric that compares an account’s value after losses to a prior high-water mark. Conceptually, it answers: “By what percent did the account decline from its peak?” Typically, it uses a sequence of account equity or balance values over time, identifies the highest value up to each point, and expresses the difference as a percentage.
Key mechanics to understand:
- High-water mark choice: Whether the “peak” resets on a calendar basis, since account start, or since the last peak affects the computed percentage.
- Value definition: Drawdown may be computed from equity (including open positions) or from closed results. Using equity can make drawdown respond to unrealized price moves.
- Timing and granularity: The metric depends on sampled time points (e.g., every tick, every hour, daily close). Different sampling can report different maximum drawdowns.
Because the metric is purely descriptive of observed values, the main risks are not “the drawdown number itself,” but what can go wrong when people rely on it for decisions or comparisons.
Risks: market, operational, and counterparty factors
Percentage drawdown is sensitive to conditions that vary over time.
Market-related risks
- Volatility amplification: In more volatile periods, price swings can move equity sharply, increasing percentage drawdown even if the underlying strategy’s intent is unchanged.
- Gaps and fast moves: If the account value changes quickly between measurement points, reported drawdown may understate or overstate the true experienced decline.
- Cost effects: Trading costs and financing effects can widen losses. Even if directional moves are similar, net results after costs can drive larger equity dips.
Operational risks
- Execution and liquidity limitations: Slippage, partial fills, or delayed execution can turn a “paper” loss into a larger real loss, making actual percentage drawdown worse than what similar backtests implied.
- System and data issues: Incorrect pricing inputs, stale valuations, or broken data feeds can distort the equity curve, and therefore the computed drawdown.
- Position and risk controls: If orders are not updated as intended (for example, risk limits are not applied promptly), drawdowns can extend beyond expectations.
Counterparty and platform risks
- Settlement and availability frictions: If a platform becomes unavailable or withdrawals/deposits are delayed, the practical experience of drawdown can differ from the metric as measured under normal conditions.
- Valuation dependence: Some drawdown calculations rely on the platform’s mark-to-market prices. If those marks lag real tradable prices, the equity curve—and thus percentage drawdown—can be temporarily misleading.
Interpretation risks: limitations, failure modes, and verification
Even when calculated correctly, percentage drawdown can be misleading if interpreted as a complete statement about risk.
Material limitations and failure modes
- Peak selection bias: A metric tied to a chosen high-water mark can look “better” or “worse” depending on when the peak is defined.
- Equity-based confusion: When using equity (including unrealized P&L), drawdown can peak and then recover without realizing losses. That makes it behave differently from a drawdown based only on realized results.
- One-number oversimplification: Two accounts can have the same maximum percentage drawdown but different recovery times or loss trajectories. Maximum drawdown alone does not capture duration.
- Outcome uncertainty: Historical relationships do not establish future results. A pattern where drawdowns “typically” recover can fail when volatility, correlations, or liquidity conditions change.
A realistic scenario and possible consequence Imagine an account that is marked to market every few hours. During a fast market move, equity drops substantially and then partially recovers before the next sampling time. The reported maximum percentage drawdown may not reflect the deepest point experienced in between samples. A trader who focuses on the reported maximum might underestimate the severity of intraperiod losses.
Control point for verification To verify claims about percentage drawdown, check at least:
- The equity definition used (equity vs balance; unrealized vs realized).
- The high-water mark rule (start date, rolling peak, reset policy).
- The sampling frequency used to compute the metric.
- Whether costs and financing are included in the account value that feeds the drawdown calculation.
Limitations and next questions to keep risks grounded
Percentage drawdown is useful for describing drawdown magnitude, but it does not remove uncertainty. Market conditions, costs, execution quality, data accuracy, and how the metric is computed can all change the meaning of the reported number.