Maximum Adverse Excursion: How to Read Your Own Stop Data in 2026
Maximum adverse excursion is the least glamorous metric in trading and one of the few that changes behavior. It answers a question every trader has asked out loud after a bad session: was my stop in a stupid place, or was the trade just wrong.
Most traders answer that from memory, and memory is a poor witness. It remembers the trade that reversed one tick past the stop and forgets the four that went straight through. The result is a slow drift toward wider stops, larger losses, and a daily loss limit that gets consumed in three trades instead of six.
This guide explains what maximum adverse excursion measures, how to record it without building a research operation, what patterns in the data actually mean, and how the metric interacts with the fixed limits of a funded account. TradeFundrr operates a simulated trading environment, and the numbers below are illustrative examples chosen to show the method.
- Record the worst point of every trade, not just the outcome. Maximum adverse excursion is the distance from entry to the deepest unrealized loss, win or lose.
- Compare winners against losers separately. The useful signal is whether the two groups look different early in the trade.
- Treat a wide stop as a cost, not as safety. Extra distance you never use still enlarges every loss you take.
- Pair MAE with MFE before changing anything. Adjusting stops without looking at how far winners ran is half an analysis.
- Remember the limit is fixed. In a funded account, an oversized stop does not just cost dollars, it costs attempts.
What maximum adverse excursion measures
Maximum adverse excursion is the largest unrealized loss a trade reached before it closed. If you entered long at 100, price traded down to 97 and then closed at 104, the MAE for that trade is three points. It stays three points regardless of the fact that the trade ended profitable.
That last part is what makes the metric useful. Outcome data tells you which trades made money. Excursion data tells you what those trades put you through on the way, and the difference between those two things is where stop placement lives.
The pair: adverse and favorable
Maximum favorable excursion, or MFE, is the mirror image: the best unrealized profit a trade reached. Together, MAE and MFE describe the full shape of a trade rather than its endpoint. MAE informs where the stop belongs. MFE informs whether your targets are leaving money on the table or reaching for moves that rarely arrive.
Traders who record only outcomes have a two-value dataset: win or lose. Traders who record both excursions have a picture of what each trade actually did, and that picture is what makes a review productive instead of self-critical.
Why memory cannot do this job
Human recall of trading is systematically biased toward vivid events. One trade stopped out by a single tick before running to target leaves a stronger impression than twelve orderly stops. The trader then widens the stop to avoid a scenario that occurs rarely, and pays for that widening on every loss thereafter. Writing the numbers down is the only reliable fix, which is the same argument we made in why a trading journal is your edge.
TradeFundrr · Maximum Adverse Excursion
Your stop distance is a guess until you measure how far trades actually go against you
Maximum adverse excursion is the worst unrealized drawdown a trade reached before it closed. Recorded across enough trades, it turns stop placement from an opinion into a measurement.
Five trades, one stop distance
Each bar shows how deep the trade went against the entry. The vertical marker is where the stop sat.
Trade 1
Never close to the stop. Winner
Trade 2
Half the stop used. Winner
Trade 3
Comfortable margin. Winner
Trade 4
Blew through the stop. Loser
Trade 5
Comfortable margin. Winner
What this pattern is telling the trader
The point of the exercise: maximum adverse excursion converts "my stops keep getting hit" into a number you can act on. Inside a funded account, where the daily loss limit is fixed, a stop that is 40% wider than it needs to be costs you attempts you cannot get back.
tradefundrr.comIllustrative example built to show the method, not real trade data. Past results do not indicate future performance. Simulated environment.
How to record it without overengineering
You need one extra column in whatever you already use. For each trade, record the worst price the position traded through while you held it. That is the entire data collection requirement.
Use one unit and stay in it
Record excursions in the same unit as your stop. If you set stops in ticks, record MAE in ticks. If you set them in ATR multiples, record it that way. Mixing units is the most common reason a promising review turns into an afternoon of arithmetic and no conclusion.
Dollars work too, with one caveat: if your position size varies, dollar MAE mixes two variables and hides the pattern. Normalizing to R, meaning the distance to your stop expressed as one unit of risk, solves that cleanly. Our post on R-multiples covers the conversion.
Log the condition alongside the number
Add a second column for market condition: trending, ranging, high volatility, news session. Without it, you will end up computing an average across conditions that behave nothing alike and drawing a conclusion that applies to none of them. Two columns and a date is a complete MAE dataset.
| Metric | What it measures | What it informs | Recorded on |
|---|---|---|---|
| MAE | Deepest unrealized loss during the trade | Stop distance | Every trade, winners included |
| MFE | Highest unrealized profit during the trade | Target and trail placement | Every trade, losers included |
| R-multiple | Final result as a multiple of initial risk | Expectancy and sizing | Closed trades |
| Win rate | Share of trades that closed profitable | Almost nothing on its own | Closed trades |
Four common trade metrics and what each one is actually good for. MAE and MFE describe the path; R-multiple and win rate describe the outcome.
The one thing to check before you trust the number
Excursion data is only meaningful if your entries were consistent. If half the sample came from a setup you now consider a mistake, the MAE distribution describes a strategy you no longer trade. Before drawing conclusions, filter the sample down to trades you would take again today. A smaller honest sample beats a larger contaminated one.
This is also the step that turns the exercise into a review rather than a spreadsheet task. Deciding which past trades still count forces you to state what your strategy actually is, and a surprising number of traders discover at this point that they do not have one specific enough to filter by.
Reading the patterns honestly
Once you have enough trades across enough conditions, three patterns are worth looking for. Each one implies a different action, and only one of them implies moving the stop.
Winners rarely approach the stop
If most of your profitable trades never used more than half the stop distance, the stop is wider than the strategy needs. The extra room is not protecting anything, it is inflating every loss. Tightening it will cost you some future winners, and the honest version of the decision is a tradeoff between fewer, cheaper losses and slightly fewer wins.
Losers go deep immediately, winners do not
This is the most actionable pattern and the rarest to find cleanly. When trades that end badly move against you fast and trades that end well do not, you have a time-based exit rule available: if the position is offside by a defined amount within a defined number of bars, close it. That is a rule built from your own data rather than borrowed from a video.
Winners and losers look identical early
This is the most common result, and it is genuinely useful information even though it feels like a null finding. It means your entry timing carries no early edge, and the improvement available to you is in selection or sizing rather than in exit tuning. Fiddling with the stop in this situation is motion, not progress. Where to place your stop loss covers the structural alternatives.
Averages hide the shape
An average MAE of 12 ticks can describe two completely different traders. One has every trade clustered between 10 and 14 ticks. The other has most trades at 4 ticks and a handful at 40. The first can tighten the stop with confidence. The second would cut most of their winners and keep the tail losses, which is the worst of both outcomes.
Look at the distribution, not the mean. A simple sort of the column from smallest to largest and a glance at where the numbers cluster is enough. You are not doing statistics, you are checking whether one number can fairly stand in for the set.
MAE inside a funded account
In a personal account an oversized stop costs money. In a funded account it costs attempts, and attempts are the scarcer resource.
The attempt arithmetic
On a TradeFundrr 50K futures account the daily loss limit is $1,000. A trader risking $250 per trade gets four attempts before the day ends. If that trader's own MAE data shows winners never needing more than 60% of the stop distance, tightening the stop to reflect that reduces risk per trade to roughly $150 and produces six attempts instead of four.
Six attempts instead of four is not a small improvement. It is the difference between a strategy with a 40% win rate having a reasonable chance of finding a winner in a session and not having one. That is what MAE data buys you, and it costs one spreadsheet column.
Drawdown is the other constraint
Daily limits reset. Maximum drawdown does not, and stops that are wider than necessary erode it steadily rather than dramatically. A trader who never breaches a daily limit can still fail an account through six weeks of avoidable extra loss per trade. Investor.gov's material on managing risk makes the general case, and inside a funded program the effect is simply more visible because the buffer is published.
The stop still comes from structure
MAE tells you what your trades have historically needed. It does not tell you to put a stop at a round number that happens to match the average, because a stop with no structural logic invites the exact fill you were trying to avoid. Use the data to constrain the range and use structure or volatility, as in ATR for stop placement, to choose within it.
What to do in the first two weeks
Do not change anything. Record the extra column, tag the condition, and keep trading exactly as you were. The temptation to start adjusting stops on day three is strong and it destroys the sample, because you end up measuring several different strategies and attributing the result to one.
Two weeks of untouched data is usually enough to see whether winners and losers separate early. If they do, you have found something specific. If they do not, you have saved yourself from a month of stop tinkering that would not have helped, which is a real result even though it does not feel like one.
Where the metric misleads people
MAE is a measurement, and measurements get misused in predictable ways.
Fitting the stop to the sample
Setting the stop exactly at the worst excursion any winner ever survived is curve-fitting with extra steps. That number came from one trade in one condition, and the next market regime is under no obligation to respect it. Use the distribution, not the extreme.
Sampling one market condition
MAE gathered entirely from a trending fortnight will recommend a tight stop that fails immediately in a choppy one. Condition-tagging every trade is not optional bookkeeping, it is what makes the conclusion valid. FINRA's material on frequent intraday trading is a useful reminder of how easily short samples flatter an active approach.
Using it to justify not having a stop
Some traders discover that trades often recover from deep excursions and conclude that stops are the problem. That conclusion survives right up until the trade that does not recover, and in a funded account it usually ends the account rather than the day. The written rules of your program are the binding version of this, and the CFTC's guidance on understanding contract obligations is worth reading on why the document, not the intention, governs.
Reviewing too often
Recalculating MAE after every session and adjusting the stop accordingly is not analysis, it is reacting to noise with extra steps. Pick a review interval measured in trades rather than days, hold the settings steady between reviews, and let the sample accumulate. A rule you change weekly was never a rule.
Expecting the data to make the decision
Here is the damaging admission. MAE will not tell you where to put your stop. It will narrow a wide, emotional argument into a narrow, factual one, and then you still have to choose. That is a smaller improvement than most metrics promise and a larger one than most metrics deliver.
Frequently Asked Questions
What is maximum adverse excursion?
Maximum adverse excursion, or MAE, is the largest unrealized loss a trade reached at any point before it was closed. It measures how far a position went against you, regardless of whether it ended as a winner or a loser.
How do you calculate maximum adverse excursion?
Take the worst price the trade traded through while you held it, measure the distance from your entry, and express it in ticks, points or dollars. Recording it for every trade in the same units as your stop is what makes the numbers comparable.
What is the difference between MAE and MFE?
MAE measures the worst point of a trade and MFE, maximum favorable excursion, measures the best point. MAE informs stop placement and MFE informs target placement, and they are most useful read together.
How many trades do I need before MAE data means anything?
Enough to cover different market conditions rather than a specific count. Thirty trades from a single trending week will tell you about that week. The same thirty spread across trending and ranging sessions will tell you about your strategy.
Can MAE data help me pass a funded evaluation?
Indirectly, and mostly by cutting wasted risk. If your data shows winners rarely use more than half your stop distance, tightening the stop lets you take more attempts within the same daily loss limit, which matters when the limit is fixed at $1,000 on a 50K account.
Does MAE tell me where to put my stop?
It narrows the range and it does not choose for you. It shows what your winners historically needed, and moving the stop inside that distance will cut some future winners short. The decision is a tradeoff you make with better information, not an answer the data hands you.
Should I widen my stop if MAE shows trades going deep before recovering?
Only if you also widen your risk accounting, which usually means reducing size. A wider stop at the same position size increases the dollar loss per trade, and inside a funded account that consumes your daily loss limit faster.
Do I need special software to track MAE?
No. A spreadsheet column for the worst price reached during each trade is sufficient. Many platforms report it automatically, but the discipline of recording it matters more than the tool.
A fixed limit rewards a measured stop
TradeFundrr publishes the daily loss limit, drawdown and position limits for every program up front, so your risk math starts from a real number.
Get Funded →