Interpreting Maximum Drawdown in Backtest Results
Learn how to read maximum drawdown in a backtest, what it tells you about risk, and how to use it alongside other metrics for robust strategy evaluation.
Produced with automation, then checked by deterministic quality rules and an independent source-grounded review before publication.
- 01Maximum drawdown measures the largest peak‑to‑trough loss during a backtest period.
- 02It must be evaluated relative to the strategy’s time horizon and capital allocation.
- 03Comparing drawdown to a benchmark or similar strategies helps gauge relative risk.
- 04High drawdown may indicate fragile position sizing, leverage, or market‑specific exposure.
- 05Use drawdown together with win rate, profit factor, and other metrics for a balanced view.
Maximum drawdown is the greatest loss from a portfolio peak to a subsequent trough during a backtest. It quantifies the worst‑case capital erosion a strategy experienced under historical conditions. Interpreting this number correctly helps you assess whether the strategy’s risk profile matches your tolerance and capital constraints.
What Does Maximum Drawdown Actually Measure?
The metric tracks the deepest decline in equity, ignoring any subsequent recovery. It is expressed as a percentage of the peak equity value, so a 25 % drawdown means the portfolio fell 25 % from its highest point before climbing again. This figure does not predict future losses, but it reveals how the strategy behaved during adverse market moves in the sample data.
How Should I Contextualize a Drawdown Figure?
A raw percentage is only meaningful when placed in context. Consider the following factors:
- The time horizon of the backtest - a 20 % drawdown over one month is far more concerning than the same figure over five years.
- The capital base and position sizing - aggressive leverage can inflate drawdown without adding real skill.
- The market regime - periods of high volatility or structural breaks often produce larger drawdowns.
- The benchmark performance - comparing to a relevant index or peer strategy shows whether the drawdown is unusually high.
When Is a Drawdown Too Large?
There is no universal threshold, but several practical guidelines help you decide if a drawdown is acceptable:
- 01Align the drawdown with your personal risk tolerance and the amount of capital you are willing to lose before exiting the strategy.
- 02Check whether the drawdown exceeds any owner‑signed daily‑loss limits you might enforce in live deployment.
- 03Assess if the drawdown coincides with known data quality issues or gaps; missing or stale market data can artificially inflate losses.
- 04Determine if the drawdown is driven by a single event (e.g., a market crash) or by repeated small losses, which may indicate systematic flaws.
How Can I Reduce Unacceptable Drawdowns?
If the backtest reveals drawdowns beyond your comfort zone, consider the following adjustments:
- Refine position sizing rules to limit exposure during high‑volatility periods.
- Introduce stop‑loss or trailing‑stop mechanisms, remembering that backtests are read‑only and do not execute real orders.
- Add diversification across assets, timeframes, or strategy components to smooth equity curves.
- Re‑evaluate the data set for coverage, freshness, and warnings; see Why Historical Data Coverage Matters in a Backtest for guidance.
What Other Metrics Should Complement Drawdown?
Relying solely on maximum drawdown can be misleading. Pair it with metrics such as profit factor, Sharpe ratio, win rate, and average trade duration. Together they provide a fuller picture of risk‑adjusted performance. For a deeper dive into benchmarking, read Why Every Trading Backtest Needs a Benchmark. For related context, see Understanding Maximum Drawdown in Trading Backtests.
A strategy with a modest profit but a massive drawdown may be unsuitable for most investors; risk management is as important as return generation.
Frequently asked questions
Not necessarily. A larger drawdown can be acceptable if the strategy delivers proportionally higher returns and fits the investor’s risk tolerance. Context matters.
No. Market conditions change, and a single period may not capture all possible scenarios. Running multiple out‑of‑sample periods helps validate robustness.
Missing, stale, or inaccurate price data can create artificial peaks or troughs, inflating drawdown. Always verify source timestamps and warnings in the data feed.
Owner‑signed limits can be configured to cap daily loss or overall drawdown, but they must be reviewed regularly and may need adjustment as market conditions evolve.
Sources and verification
Product claims in this article were checked against these first-party references. Runtime status remains authoritative for current availability.
- Felix documentationfirst party
- Felix machine referencefirst party
Build with Felix now.
Felix infrastructure is live through MCP and the API. The full trading app launches September 17.
Partial fills are common in automated trading. This article explains how an AI agent can reliably detect them, reconcile positions, respect limits, and manage uncertainty without assuming perfect execution.
A money map guides an AI trading agent through capital handling, risk controls, market data validation, and position tracking. This article outlines each element and its practical considerations.