Backtesting
Maximum Drawdown in Backtesting
A practical guide to reading drawdown depth, duration, recovery behavior, and comparison workflows before trusting a crypto strategy backtest.
Traseq··7 min read
A practical guide to reading drawdown depth, duration, recovery behavior, and comparison workflows before trusting a crypto strategy backtest.
Start with a no-code crypto spot strategy, lock the version, run the backtest, and keep the result traceable for comparison.
Maximum drawdown is the worst peak-to-trough drop in a backtest equity curve. It is often the fastest way to see whether a strategy's return came with a path you could actually tolerate.
For crypto spot strategy research, review drawdown before headline return.
In Traseq, that review belongs inside a reproducible workflow: build the strategy, finalize the version, run a bar-based backtest, inspect the equity and drawdown behavior, then compare alternatives before making any live decision.
Traseq is a research workspace, not a live trading or exchange execution platform. Backtests help you evaluate historical behavior; they do not guarantee future results.
Maximum drawdown answers one urgent question: how far did the strategy fall from a previous equity high before recovering?
Use it as a first-pass risk screen:
Maximum drawdown measures the largest decline from a previous equity peak to a later equity trough before a new high is reached.
The simple formula is:
Maximum drawdown = (trough value - peak value) / peak value
If a backtest equity curve rises to $10,000, falls to $7,500, and later recovers, the drawdown for that decline is -25%. If that is the deepest decline in the test period, the maximum drawdown is 25%.
That number matters because it turns an abstract losing period into a concrete survival question:
Two backtests can end with similar total return but carry very different risk:
This is why maximum drawdown should be treated as a primary backtest risk metric, not a secondary statistic. A strategy that looks attractive on ending return can still be unsuitable if the path requires a drawdown the researcher cannot tolerate or fund.
Maximum drawdown answers only one question: how deep was the worst decline? A serious review should also inspect how the drawdown behaved.
The key point: a 15% maximum drawdown with fast recovery may be easier to evaluate than a 10% drawdown that keeps the strategy underwater for most of the test. The percentage is important, but it is not the whole story.
When you review many strategy variants, drawdown helps you discard fragile candidates before spending time on fine tuning.
A practical drawdown screen can look like this:
This workflow keeps the question grounded: not "Which backtest had the highest return?" but "Which result had the best tradeoff between return, downside risk, and repeatability?"
The most common mistake is treating the historical maximum drawdown as a future limit. It is not. It is the worst decline observed in one historical sample under one set of assumptions.
Other mistakes are just as costly:
Backtesting is most useful when it creates evidence you can challenge. Drawdown is one of the first places to challenge it.
Traseq keeps drawdown analysis tied to the broader research workflow instead of leaving it as an isolated number.
In a typical Traseq review, you can:
Current Traseq backtests are bar-based research simulations. Conditions are evaluated on bar close, and signal-driven entries and exits fill at the next bar open. Fee and slippage controls help model research assumptions, but the tool should still be used as research evidence, not execution proof.
For setup details, read the first backtest guide. For result interpretation, review Reading Backtest Results. To compare variants, use Comparing Multiple Backtests.
Before you advance a backtest result, ask:
If the answer is unclear, the next step is not more optimization. The next step is a cleaner comparison.
See the Traseq backtest workflow to run a version-linked drawdown review on a supported crypto spot pair.
Maximum drawdown is the largest peak-to-trough decline in the backtest equity curve. It shows the deepest historical loss period inside the tested range.
Lower drawdown is usually easier to tolerate, but it should be reviewed with return, trade count, recovery behavior, fees, slippage, and the strategy's intended timeframe. A low drawdown number can still be misleading if the test is too short or overfit.
Drawdown depth measures how far equity fell from a prior high. Drawdown duration measures how long equity stayed below that prior high. A serious review should consider both because a shallow but very long drawdown can still be difficult to trade.
Yes. A backtest reflects one historical sample and one set of assumptions. Future market behavior, trading costs, liquidity, timing, and strategy drift can produce worse drawdowns than the historical maximum.
Start with one finalized strategy version, run a backtest with explicit fee and slippage assumptions, inspect maximum drawdown and the drawdown chart, then compare variants under the same symbol, timeframe, range, and execution settings.
| Lens | Question | Why it matters |
|---|
| Depth | How bad was the worst loss? | Sets the downside reference point |
| Duration | How long was equity underwater? | Shows whether capital was stuck |
| Recovery | Did it recover cleanly? | Separates resilience from luck |
| Frequency | Did drawdowns repeat? | Exposes recurring stress |
| Return tradeoff | Was the return worth the loss? | Keeps review risk-adjusted |