Drawdown Review: the definition people get wrong
Drawdown Review is a way to examine how far results fall from a prior peak during a defined period, using a consistent rule for measuring performance. A common mistake is treating “drawdown” as a single, universal number without specifying the method (for example: peak-to-trough based on equity, balance, or another metric) and the time window used.
Another misunderstanding is to assume that drawdown “severity” automatically reflects trading quality. In reality, drawdown can be driven by many factors outside the core decision-making process, such as execution timing, transaction costs, and changing market behavior. If those inputs are not clearly separated, the review can incorrectly attribute cause.
Common mistakes and what they do to your conclusions
1) Mixing mechanics with variable conditions
A frequent error is to blend a stable measurement procedure with assumptions that change. For instance, you might compute drawdown using historical prices but then interpret it as if future costs, spreads, slippage, or fill quality will match history. That breaks comparability.
What to check: state which values you used (and where they came from) and keep the measurement rule separate from “what might change later.”
2) Using unclear assumptions for any calculation
Even when the concept is simple, mistakes happen when assumptions are implicit. Examples include: which peak is used, whether compounding is included, whether profits and losses are marked at the same valuation method throughout the period, and what happens during missing or irregular data.
What to check: write down the assumptions for every example or calculation. If you cannot explain the inputs clearly, you cannot verify the result.
3) Selecting data in a way that hides the true drawdown
Drawdown Review depends heavily on the chosen start date, end date, and the treatment of discontinuities (such as resets, withdrawals/deposits, or changes in account rules). A common mistake is to review only the “most convenient” segment or to compare periods with different constraints.
What to check: compare like-for-like periods and document any account changes that affect the measured equity curve.
4) Treating historical relationships as predictive
Historical drawdown patterns do not establish future outcomes. A drawdown that appears “contained” in the past may still widen later because market dynamics and costs differ.
What to check: avoid presenting drawdown findings as expectations. Use the review to understand behavior under past conditions, not to forecast.
Evidence and neutral checks you can run
A neutral check is to reproduce the calculation with explicit steps and verify consistency across definitions. Practical checks include:
- Recompute drawdown from the same equity series using the same peak-to-trough rule.
- Confirm the time window and peak selection logic match your stated method.
- Repeat the process with alternative but clearly defined valuation choices (only if you can keep the method consistent) to see whether conclusions depend on the definition.
If the conclusions change dramatically when you clarify assumptions, that is a sign the “lesson” you drew may be more about the review setup than about performance.
Limitations and failure modes to include
At least one material limitation is easy to overlook: drawdown can be distorted by measurement choices and account events. For example, deposits/withdrawals, changes in leverage, or valuation conventions can make drawdown look better or worse without reflecting the underlying strategy decision quality.
Another failure mode is misinterpreting risk. Drawdown indicates a fall from a peak, but it does not by itself prove that the strategy is “safe,” nor does it identify the cause of losses. Outcomes also vary with jurisdiction, market conditions, costs, and execution, none of which are fixed.
Verification and your next question
To avoid common mistakes, use a clear definition, state assumptions, and verify comparability before drawing conclusions. If you want to go one step further, ask: “What exact definition, inputs, and peak-selection rule did I use, and would another reader reproduce the same drawdown from the same data?”
You can also compare your approach to a broader limitations discussion to see which parts of Drawdown Review are most sensitive to measurement choices.