Maximum Adverse Excursion: Let Your Worst Moments Set Your Stop-Loss

Every trade has a bad moment. You buy, the stock dips, your stomach tightens, and then it either recovers or it hits your stop. Maximum adverse excursion (MAE) is simply a record of how deep that bad moment went on each trade. Collect enough of those numbers and they start telling you something most traders never find out: whether your stop-loss is sitting in the right place, or whether you picked it out of thin air.

We like MAE because it uses the one dataset nobody can argue with, which is your own trade history. Not a rule of thumb from a forum, not a fixed 2% stop someone swore by. Your entries, your exits, your setups.

What maximum adverse excursion actually measures

MAE is the furthest price moved against you while a trade was open. For a long position, it is the entry price minus the lowest price reached before you exited. For a short, it flips: the highest price reached after entry minus the entry price. Its mirror image is maximum favourable excursion (MFE), the furthest price moved in your favour while you were in the trade. TradeLabs’ lesson on MAE and MFE lays out both definitions and credits John Sweeney with popularising them.

Two details matter more than they look. First, MAE is measured only between your entry and your exit. What the stock did after you left is irrelevant here, because the whole point is to study the trade you actually held. Second, MAE can never be negative. If a long trade never traded below your entry, its MAE is zero.

Sweeney used this idea to choose stops in his own system testing. In a January 1992 article in Technical Analysis of Stocks & Commodities, he wrote that the stops in his test were selected by MAE analysis to cut off the big losers while giving winners enough room to work. The article’s listing on the Traders.com store carries that passage. That is the core use case, and it has not changed in three decades.

Measure it in R, not in rupees

A ₹15 dip means one thing on a ₹150 stock and something quite different on a ₹3,000 stock. So instead of recording MAE in rupees, record it as a multiple of your initial risk. One R is the distance between your entry and your initial stop. If you have not used R-multiples before, our explainer on the R-multiple covers the basics.

The formula is short:

• MAE in R (long) = (entry price minus lowest price during the trade) divided by (entry price minus initial stop)
• MFE in R (long) = (highest price during the trade minus entry price) divided by (entry price minus initial stop)

For a short trade, swap the direction of each subtraction. Once everything is in R, a swing trade in a mid-cap and an intraday trade in a large-cap sit on the same scale, and you can compare them honestly.

Worked example 1: a long trade

All numbers in these examples are hypothetical and chosen to make the arithmetic easy to follow.

Say you buy a stock at ₹500 with a stop at ₹480. Your 1R is ₹20. Over the next few sessions the stock dips to ₹491, then runs up to ₹538, and you sell at ₹530.

• MAE = ₹500 minus ₹491 = ₹9, which is 9 divided by 20, or 0.45R
• MFE = ₹538 minus ₹500 = ₹38, which is 1.9R
• Result = ₹530 minus ₹500 = ₹30, which is +1.5R

Three numbers, one trade, and already a story. The trade never came within half of your risk distance. And you captured 1.5R of a move that offered 1.9R at its best point.

Worked example 2: a short trade

Now a short. You sell a stock at ₹1,200 with a stop at ₹1,236, so 1R is ₹36. It rallies to ₹1,218 first, then falls to ₹1,146, and you cover at ₹1,164.

• MAE = ₹1,218 minus ₹1,200 = ₹18, which is 0.5R
• MFE = ₹1,200 minus ₹1,146 = ₹54, which is 1.5R
• Result = ₹1,200 minus ₹1,164 = ₹36, which is +1.0R

Same idea, opposite direction. The heat you took was half your risk, and you gave back a third of the best open profit before covering.

One trade’s MAE is an anecdote. Fifty trades’ MAE, split into winners and losers, is evidence about where your stop belongs.

Reading MAE across many trades

A single number tells you very little. The useful move is to log MAE for a batch of trades from one setup and then split the list into winners and losers. TradeLabs suggests a simple reading of that split:

• If winning trades rarely show MAE beyond about 0.5R while losers mostly travel the full 1R to the stop, your stop may be wider than it needs to be.
• If plenty of eventual winners first dip to 0.9R or more, your stop is about right or even on the tight side, and squeezing it further would turn some winners into losers.

The logic is intuitive once you see it. Good trades in many setups tend to work quickly, without much pain. Losing trades tend to keep going against you. If there is a clean gap between the two groups, that gap is where a tighter stop could live.

Worked example 3: testing a tighter stop on a journal

Here is a hypothetical journal of 40 swing trades from a single breakout setup, all with stops at 1R:

• 16 winners with total results of +32R. Fourteen of them had MAE below 0.5R. Two had deeper dips, of 0.8R and 0.9R, and finished at +1.5R and +2.5R.
• 18 losers that hit the stop for −1R each, a total of −18R.
• 6 losers closed early on a time exit at an average of −0.4R, with MAE below 0.7R, a total of −2.4R.

Net result: 32 minus 18 minus 2.4 = +11.6R across 40 trades, or 0.29R per trade.

Now suppose you replay the same trades with a stop at 0.7R of the original distance, and assume prices behave the same way. The 14 shallow-dip winners are untouched, and they contributed 32 minus 4 = +28R. The two deep-dip winners now get stopped out at −0.7R each, which is −1.4R. The 18 full losers lose −0.7R each instead of −1R, which is −12.6R. The six time-exit losers do not change, at −2.4R.

New net: 28 minus 1.4 minus 12.6 minus 2.4 = +11.6R. Exactly the same, measured in the old R.

That surprises people. The tighter stop saved 5.4R on the losers but cost 5.4R on the two winners it killed (they went from +4R combined to −1.4R). So is there any point? There can be, because of position sizing. If you keep the rupee amount you risk per trade fixed, a stop that is 0.7 times as far away lets you hold more shares. Expressed in the new, smaller R, that same +11.6 old-R becomes 11.6 divided by 0.7, about +16.6R. On a fixed risk of ₹2,000 a trade, that is the difference between roughly ₹23,200 and ₹33,140 over the 40 trades, before costs.

Before anyone gets excited, read the fine print. Bigger positions mean bigger rupee losses when a stock gaps through your stop. Brokerage, taxes and slippage rise with more shares. And the replay assumes the future looks like the past, which is the assumption that hurts traders most. This is why the next section matters.

How not to fool yourself with MAE

MAE analysis is easy to misuse, mostly because it is so easy to fit to history.

Use enough trades. TradeLabs suggests waiting for at least 30 to 50 trades of one setup before drawing conclusions. Sweeney himself flagged a test with fewer than 30 trades as too small to rely on in the 1992 article. Twelve trades from a good month prove nothing.

One setup at a time. A breakout trade and a mean-reversion trade behave very differently after entry. Mixing them blurs the gap between winners and losers until it disappears.

Change one thing. Adjust the stop or the target, not both in the same review. Otherwise you will not know which change did what.

Paper-test before adopting. Run the new stop on paper or with small size for the next batch of trades. If the gap between winners and losers was real, it should show up again.

Watch for changing volatility. Excursions in a calm market are not the same as excursions in a jumpy one. A stop tuned in a quiet stretch can get shredded when volatility picks up.

There is also an honest dissent worth keeping in mind. On the Omega TradeStation users’ mailing list in 2002, one trader who tested the approach reported that it worked less well in practice than expected: some losers that would have recovered partway before hitting the normal trailing stop were cut earlier by the tighter MAE-based stop, which lowered the system’s expectancy. You can read that exchange in the Omega list archive on PureBytes. The lesson is not that MAE fails. It is that the replay has to include every exit rule you actually use, not only the initial stop.

A tighter stop is not automatically a better stop. It only helps if the winners it would have killed were rare to begin with.

Don’t ignore the other half: MFE

MAE answers “how much heat did this trade take?” MFE answers “how much profit was on the table?” Together they fix both ends of a trade.

If your average MFE on winners is far larger than your average exit, you are probably taking profits too early, and a trailing stop or partial exits may suit you better. If a lot of your losing trades were up 1R or more before reversing into a full loss, that is a strong case for moving the stop to breakeven sooner or booking part of the position earlier. TradeLabs makes both points, and flags ignoring MFE on losing trades as a common mistake, since that column often hides the biggest leak in a system.

In worked example 1, you exited at +1.5R on a trade that reached +1.9R. One trade like that is nothing. If your journal shows winners consistently giving back half their MFE, that is a pattern worth fixing, and it may matter more than a slightly better stop.

Setting up an MAE log in a spreadsheet

You do not need special software, though many platforms and journaling tools can compute excursions from your fills automatically. A plain spreadsheet with these columns is enough:

1. Date, symbol, setup name, long or short
2. Entry price, initial stop, and 1R in rupees
3. Lowest price and highest price while the trade was open
4. MAE in R, MFE in R
5. Exit price and result in R

Fill in the lowest and highest prices from the daily (or intraday) highs and lows between your entry and exit, using the same chart timeframe every time. Then sort by result, split winners from losers, and look at the MAE column for each group. If you track the profit factor of your system too, our guide to profit factor in trading shows how to read it alongside these numbers.

Where maximum adverse excursion fits in your risk plan

MAE does not replace a stop-loss rule, a position-sizing rule or a maximum daily loss. It audits them. It tells you whether the stop you chose is in a sensible spot for the way your setup actually trades, and it gives you a measured, testable reason to change it instead of a hunch after a painful week. Our pillar guide on risk management in trading covers the wider framework that MAE plugs into.

Start small. Log the next 30 trades of your favourite setup with MAE and MFE in R. You may find your stop is fine. You may find it is generous by a third. Either way, you will know, and knowing beats guessing every time the market tests your nerve.

Disclaimer: This article is for general information and education only and is not investment advice. All trade examples are hypothetical and ignore brokerage, taxes and slippage. Trading in securities involves risk of loss; consider consulting a SEBI-registered investment adviser before making trading decisions.