Drawdown Metric: Calculation and Use Cases
Drawdown measures how far an account or portfolio has fallen below a previous high. Maximum drawdown is the largest such decline in a specified period, measured in currency or as a percentage. It describes the history you observed; it is not a ceiling on future losses and does not require the account to have recovered before it can be measured.
Use drawdown to compare strategy behavior, understand recovery demands, and plan exposure. On LuxAlgo, you can research a strategy with Quant, our coding agent, and inspect its drawdown alongside trades and other results on Quant Charts. The value becomes useful when you know which equity series, sampling interval, and calculation convention produced it.
Absolute vs Relative vs Maximum Drawdown
The names overlap, and platforms do not always use them identically. State both the unit and the reference point rather than treating them as three independent measures of all downside risk.
| Measure | Definition used here | Example |
|---|---|---|
| Absolute drawdown | Amount below starting equity, floored at zero if equity never falls below that starting amount | $80,000 start and $75,000 minimum: $5,000 |
| Percentage drawdown | Decline from the running equity high, divided by that high | $100,000 high to $75,000: 25% |
| Maximum percentage drawdown | Largest percentage drawdown during the selected period | The worst percentage decline across all observations |
| Maximum currency drawdown | Largest currency decline from a prior running high | The worst dollar decline, which can occur in a different episode |
For example, MetaTrader 5's testing report uses “Maximal” for the largest currency decline and “Relative” for the largest percentage decline, with separate balance and equity fields. This guide uses maximum percentage drawdown when referring to percentage MDD. Always read the definition beside a platform's result.
Calculation Methods for Drawdown
Step-by-step manual calculation
Take a chronological series of account equity values, including the starting capital. At each observation, keep the highest value seen so far: the running high-water mark. Then calculate:
- Currency drawdown = running high − current equity.
- Percentage drawdown = (running high − current equity) ÷ running high.
- Maximum drawdown = the largest drawdown in the selected series.
These formulas express losses as positive magnitudes. If you prefer a negative underwater curve, use current equity ÷ running high − 1 and take the most negative value. Do not mix the two sign conventions.
A fall from $100 to $60 is a $40 decline and a 40% drawdown. Selecting the highest and lowest values independently is not enough: the high must occur before the low. A trough that precedes a later high is not a drawdown from that later high.
Worked equity example
Assume the following equity path has no deposits or withdrawals:
| Observation | Equity | Running high | Currency drawdown | Percentage |
|---|---|---|---|---|
| Start | $80,000 | $80,000 | $0 | 0% |
| 2 | $100,000 | $100,000 | $0 | 0% |
| 3 | $75,000 | $100,000 | $25,000 | 25% |
| 4 | $105,000 | $105,000 | $0 | 0% |
| 5 | $200,000 | $200,000 | $0 | 0% |
| End | $170,000 | $200,000 | $30,000 | 15% |
The maximum percentage drawdown is 25%, but the maximum currency drawdown is $30,000. Absolute drawdown is $5,000 below the $80,000 starting equity. Current drawdown at the end is 15%, and that final episode remains unrecovered. Dividing the largest dollar loss by the largest account peak would miss the earlier 25% episode.
Spreadsheet formulas
Put chronological dates in column A and numeric equity values in B, starting at row 2. For the six observations above, use rows 2–7:
- C2, running high:
=MAX($B$2:B2) - D2, positive percentage drawdown:
=(C2-B2)/C2 - E2, currency drawdown:
=C2-B2
Fill C2:E2 down to row 7 and format D as a percentage. Then use =MAX(D2:D7) for percentage MDD, =MAX(E2:E7) for currency MDD, and =MAX(0,$B$2-MIN(B2:B7)) for absolute drawdown. Replace row 7 with your actual final data row.
For a negative curve instead, D2 would be =(B2-C2)/C2, with =MIN(D2:D7) returning −25% in this example. Investigate missing observations and zero or negative equity rather than hiding them with a formula that returns zero. Percentage interpretation requires a positive reference high; leverage can produce losses beyond 100% if equity becomes negative.
Choose the Equity Series Before Comparing Results
- Balance versus equity: a balance series based on closed trades can omit unrealized losses. Marked-to-market equity includes the value of open positions.
- Sampling: end-of-day or bar-close values can miss a deeper intraday decline. Report the observation frequency.
- Cash flows: deposits can disguise drawdowns and withdrawals can resemble losses. For investment-performance comparisons, use a properly cash-flow-adjusted or unitized series; also review actual account capital separately.
- Costs and currency: use a consistent treatment of fees, financing, slippage, and currency translation.
- Test period: a short favorable sample and a long sample containing stressed markets are not directly comparable evidence.
Reading TradingView drawdown fields
TradingView documents separate intrabar drawdown and close-to-close drawdown calculations. The former uses the strategy's position and modeled price path, while the latter uses closing equity observations. Do not expect a simple daily spreadsheet to reproduce an intrabar strategy metric.
A chart indicator tracking an asset's drawdown is also different from account drawdown: position size, cash, shorts, leverage, and costs change the account path. A performance field does not automatically create an account-wide alert or a risk-control order. Those require a defined implementation and access to the relevant account or strategy data.
Drawdown and Recovery Are Different Percentages
After a decline of fraction d, the gain needed to regain the old high is d ÷ (1 − d), assuming no cash flows. A 25% loss requires a 33.33% gain from the trough; a 40% loss requires 66.67%; a 50% loss requires 100%. This arithmetic does not predict whether or when recovery will occur.
Record duration as well as depth. Time from peak to trough describes the decline; time from trough back to the old high describes recovery; time underwater spans the whole episode. State your convention and flag episodes still open at the end of the sample. The deepest decline need not be the longest.
Research Drawdown with LuxAlgo
Compare strategy versions with Quant
Use Quant, our coding agent, to build or revise a strategy with explicit entries, exits, and sizing. Review the generated code and run it on the intended symbol and timeframe. A change intended to reduce drawdown might also reduce returns, change turnover, or alter the trade sample.
The native backtest summary reports maximum drawdown with net profit, trade count, win rate, and profit factor. Open the full viewer to inspect performance and individual trades. Save a run to retain its script, symbol, timeframe, inputs, and simulation properties. Compare versions using the same data, costs, capital, and evaluation window.
For example, compare fixed-size and equity-based sizing, or test a precisely defined exposure-reduction rule against the original strategy. Specify when a drawdown threshold is evaluated, what happens to existing positions, and when normal sizing resumes. A rule using the completed bar's equity must not assume it acted earlier within that bar.
Review a later period that was not used to select the settings, and inspect the trades around the largest decline. Backtests remain simulations. Do not assume that a generated strategy automatically includes Monte Carlo analysis, walk-forward validation, or live drawdown alerts.
Review recorded trades in the Journal
LuxAlgo's built-in Journal provides a separate workflow for recorded fills, trades, and notes. It lives on your account rather than in a workspace, and supports manual records, imports, and supported broker connections. Use it to review actual trading records alongside your research, while checking how a displayed curve treats open positions and cash flows before comparing its drawdown with a backtest.
A low historical drawdown in a Quant backtest is a starting point for further evaluation, not assurance of future capital protection.
Drawdown Control Methods
Position size management
Reducing exposure can reduce loss amounts, but it does not repair a losing strategy. If a rule risks 2% of current equity, $100,000 implies a $2,000 planned risk budget and $90,000 implies $1,800. The adjustment only occurs if the sizing method actually recalculates it. Percentage position value is not the same as percentage risk at a stop.
As an illustration, ten consecutive losses of exactly 2% of then-current equity leave about 81.71% of the starting amount, an 18.29% drawdown before additional costs. Gaps, simultaneous positions, or different exit losses can produce a worse result. A 2% convention is not a universally suitable recommendation.
Diversification and portfolio exposure
Diversification can reduce concentration when the components behave differently, but asset-class labels do not establish independence. An illustrative 40% stock-index, 30% forex, 20% commodity, and 10% crypto capital allocation says little about risk without position direction, leverage, volatility, and correlations. It is not a recommended portfolio.
Evaluate the combined, time-aligned portfolio equity series. Do not average each component's maximum drawdown: their peaks and troughs occur at different times. Correlations can change under stress, and currency or leverage exposure can connect positions that initially appear diversified.
Volatility-based exits and review limits
ATR measures price movement, while account drawdown measures changes in equity. An ATR-based stop can be tested, but there is no universal rule requiring 2–3 ATR in calm markets and 1 ATR in high volatility. A tighter stop may increase stop-outs; a wider stop increases risk unless size is reduced.
Define the ATR period, multiplier, price anchor, update timing, and whether the stop can loosen. A portfolio review threshold also needs a planned response: pausing new entries, reducing exposure, and closing positions are different actions. Stops and review limits do not guarantee an account-loss ceiling, especially across gaps or illiquid markets.
Watch: Understanding Drawdown
LuxAlgo's drawdown explainer offers additional context. Any older product interface shown in the video is separate from the current Quant Charts and Journal workflow described above.
Use Drawdown as Part of the Risk Picture
Report the largest historical decline together with its dates, duration, equity definition, sampling, and costs. Compare it with returns, trade count, and other risk measures rather than selecting the smallest MDD in isolation. For a broader review, see five metrics for evaluating trading strategies.
FAQs
What is drawdown in trading?
Drawdown is the decline in account or portfolio value from a prior high. It can be measured in currency or as a percentage of that high. Maximum drawdown is the largest decline in the chosen period, including an ongoing decline that has not recovered. It describes observed history rather than the maximum possible future loss.
How do Absolute, Relative, and Maximum Drawdown differ?
Absolute drawdown commonly measures how far equity fell below starting capital, with zero if it never did. Percentage drawdown measures the decline relative to a prior high. Maximum drawdown selects the largest decline in the period, in currency or percentage terms. Platform labels vary; MetaTrader's Relative field is the largest percentage decline, while Maximal emphasizes the largest currency decline.
Why are drawdown metrics important?
They help evaluate historical loss severity, recovery demands, position sizing, and strategy behavior during stress. Use comparable equity series and sampling, and include duration and costs. Drawdown alone does not capture every risk, guarantee recovery, or prove that a strategy will remain within the same loss range in the future.
References
Read next