MAE and MFE: Use Trade Excursions to Review Stops and Exits

বাংলায় পড়ুন

A trade's final profit or loss hides the path it took. A winner may have come close to its stop before recovering; another may have offered a large unrealised gain before giving most of it back. Maximum Adverse Excursion (MAE) and Maximum Favourable Excursion (MFE) turn those paths into reviewable numbers.

Learning objective

By the end of this guide, you should be able to calculate MAE and MFE in price, money and initial-risk units; use a group of completed trades to question stop and exit rules; and avoid changing a strategy from one memorable example.

1. What MAE and MFE measure

Maximum Adverse Excursion (MAE) is the largest unrealised move against a position between entry and exit.

Maximum Favourable Excursion (MFE) is the largest unrealised move in the position's favour between entry and exit.

For a long trade:

• MAE per unit = entry price − lowest price while the trade was open.
• MFE per unit = highest price while the trade was open − entry price.

For a short trade:

• MAE per unit = highest price while open − entry price.
• MFE per unit = entry price − lowest price while open.

Record zero when price never moves against or in favour of the entry. Keep adverse and favourable excursion as non-negative magnitudes; the trade's realised return can still be positive or negative.

Price units alone are difficult to compare across assets. Convert each value into money and R, where 1R is the trade's planned initial risk:

Excursion in R = excursion per unit ÷ initial risk per unit

The initial risk must be fixed from the plan at entry. Replacing it later with a wider stop makes comparisons misleading.

2. A worked hypothetical example

Suppose a trader buys 200 shares at £25.00 with an initial stop at £24.40.

• Initial risk per share: £25.00 − £24.40 = £0.60.
• Total initial risk: 200 × £0.60 = £120, before costs.
• Lowest price while open: £24.76.
• Highest price while open: £26.20.
• Exit price: £25.72.
• Total commissions and fees: £8.

The calculations are:

Measure	Calculation	Result
MAE per share	£25.00 − £24.76	£0.24
MAE in R	£0.24 ÷ £0.60	0.40R
MAE in money	200 × £0.24	£48
MFE per share	£26.20 − £25.00	£1.20
MFE in R	£1.20 ÷ £0.60	2.00R
Gross realised gain	200 × (£25.72 − £25.00)	£144
Net realised gain	£144 − £8	£136, or about 1.13R

The trade used 0.40R of adverse room, reached 2.00R of favourable excursion and realised about 1.13R after the stated costs. Its gross exit captured 60% of MFE because £0.72 ÷ £1.20 = 60%.

That does not prove the trader should always exit at 2R or demand a 60% capture ratio. The session high is known only afterwards. A rule must be executable in real time and tested across many trades.

3. Build a consistent dataset

Add these fields to a trading journal:

1. Instrument, direction, entry and exit timestamps.
2. Entry price, position size and planned initial stop.
3. Highest and lowest tradable prices while the position was open.
4. Realised gross and net P&L, including spread, commission, slippage, funding and borrow costs where relevant.
5. MAE and MFE in price, money and R.
6. Setup, timeframe and market-regime labels.
7. Whether the stop or exit followed the written rule.

Use the same price convention throughout. A long position may be entered at the ask and exited at the bid; using chart mid-prices can understate adverse movement and trading costs. For shares, use data adjusted consistently for splits and other corporate actions. For futures, include contract multipliers; for leveraged products, do not confuse margin posted with money at risk.

Sampling frequency matters. Daily bars can miss intraday paths and cannot tell whether a stop or target occurred first inside the same candle. Our same-candle backtest guide explains that ambiguity. Use data granular enough for the strategy and document any assumptions.

4. Analyse distributions, not anecdotes

Review a sufficiently large set of comparable trades. Separate longs from shorts if their behaviour differs, and do not mix a five-minute breakout with a multi-day swing trade merely because both traded the same asset.

Useful questions include:

• What are the median, 75th and 90th percentile MAE for eventual winners?
• How many winners would a proposed tighter stop have removed?
• How much loss would that tighter stop have saved on losers?
• At what MFE do trades commonly stall, and how much is later realised?
• Do large-MFE trades share a documented condition, or are they rare outliers?
• Does exit capture deteriorate after costs or in a particular regime?

A scatter plot of MAE versus realised R can show whether winners typically need more room than the current stop allows. A plot of MFE versus realised R can reveal large give-backs. Neither plot tells you what the optimal rule is; it tells you what to test next.

5. Test a rule without hindsight

If the review suggests a change, write it precisely. For example: “After price reaches +1R, trail the stop below the previous completed bar,” rather than “protect profit sooner.” Then:

1. Apply the proposed rule to historical trades using only information available at each moment.
2. Include spread, slippage, gaps and unfilled orders.
3. Compare expectancy, win rate, average win, average loss, drawdown and trade duration—not just total profit.
4. Keep a later period or separate sample untouched for validation.
5. Paper-test the operational rule before risking capital.

MAE/MFE can identify a question, but they cannot remove selection bias, look-ahead bias or regime change. A stop chosen at the 90th percentile of past winners can still fail in the next volatility regime.

Common mistakes

• Measuring excursion from the signal price instead of the actual fill.
• Using the final, widened stop to calculate R.
• Ignoring bid–ask spread, fees, slippage, funding or contract value.
• Mixing timeframes, setups and market regimes into one average.
• Treating the maximum favourable price as if it were available for execution.
• Tightening a stop only because one winner had small MAE.
• Optimising repeatedly on the same sample until the rule fits noise.
• Comparing live trades with backtests that use less granular or cleaner data.

Practical checklist

• Define entry, exit, stop and price convention before measuring.
• Store the high and low while each trade is open.
• Calculate MAE and MFE in price, money and initial R.
• Include all material trading costs in realised results.
• Segment trades by setup, direction, timeframe and regime.
• Review medians and percentiles, not only averages or extremes.
• Translate an observation into a precise, executable rule.
• Test with chronological data and an untouched validation sample.
• Recheck results when volatility or execution conditions change.
• Keep the old rule until evidence supports the replacement.

The Money Man View

MAE and MFE are diagnostic tools, not automatic stop and target generators. Their best use is to expose a mismatch: perhaps the stop is inside normal noise, perhaps profitable trades regularly surrender too much, or perhaps a supposed problem disappears after costs and comparable trades are separated correctly. The remedy still requires a forward-testable rule and risk limits.

Educational-risk disclosure: This hypothetical example is for general education, not a recommendation or universal risk rule. Historical excursions do not guarantee future price paths, liquidity or fills. Stops may execute beyond their trigger, and trading can result in losses. The Money Man uses AI-assisted research and writing; no personal trade or regulated advice is claimed. The reused cover is a conceptual illustration.