Advertisement IC Markets

Maximum Adverse Excursion: Using MAE to Set Better Stops

Master exponential money management, position sizing calculators, strict daily stop-loss limits, and overcoming FOMO on micro-timeframes.
Post Reply
FXS Prop Desk
Posts: 1689
Joined: Wed Oct 07, 2026 9:52 am

Maximum Adverse Excursion: Using MAE to Set Better Stops

Post by FXS Prop Desk »

MAE measures how far a trade moved against you before it closed. Tracking it for each trade reveals whether your stops are too tight or too loose.

How to use it:
  1. Record MAE (in pips) for every trade, together with the final result.
  2. Plot MAE for winning trades only.
  3. If most winners never went more than 4 pips against you, a 10-pip stop is giving away risk.
  4. If many winners went 6–7 pips against you before working, a 5-pip stop is cutting good trades.
Most platforms do not show MAE directly, but trade analytics tools, journaling software or a simple script on tick data can calculate it.

The companion metric is MFE (maximum favourable excursion), which helps evaluate profit targets.

Do you track MAE/MFE, and what did it reveal about your stops?
FXS Official Team | The Idea Lab & Strategy
Recommended broker for automated trading & scalping IC Markets
Fairman
Posts: 6955
Joined: Tue Jul 21, 2026 7:11 am
Location: Abuja

Re: Maximum Adverse Excursion: Using MAE to Set Better Stops

Post by Fairman »

FXS Prop Desk wrote: Plot MAE for winning trades only.
Worth watching for one trap with this. MAE is capped by the stop you already used, so if every trade had a 5 pip stop, no winner can show more than 5 pips of MAE and the plot will always seem to confirm the stop is about right. To learn whether a wider stop would have saved trades, you need to measure what price did after the stop-outs, say how many of them went on to reach the original target within the same session. Then compare the extra winners against the extra risk per trade. Running that on stopped trades often tells you more than the winners' MAE does, since the winners are the trades the stop didn't affect.
It’s Fairman :geek:
Shadow Trader
Posts: 5594
Joined: Wed Sep 09, 2026 5:19 pm

Re: Maximum Adverse Excursion: Using MAE to Set Better Stops

Post by Shadow Trader »

Fairman's point about the stop capping MAE is the reason my first attempt at this told me nothing. I was running a fixed 8 pip stop on EURUSD, so every winner sat inside 8 pips by definition and the plot looked like proof.

Inside that cap the shape was still useful. Of my last 180 winners, 144 of them, so 80 percent, never went more than 3.5 pips against me, and only 9 went past 6 pips.
img_13652_r1.png
img_13652_r1.png (61.02 KiB) Viewed 183 times
The illustration shows that winner histogram, with the 8 pip stop as the hard ceiling. Almost everything that worked did so before the 4 pip mark.

Next I tested a 6 pip stop on paper instead of 8. The 9 deep winners would have been lost, but every losing trade would have cost 2 pips less, and over that sample the smaller loss came out ahead.

The stopped trades are the harder half. I logged 61 stop-outs and checked how many reached the original target before the session ended, and it was 7, which is too few to argue for a wider stop.

Regime is my open question. Does anyone split MAE by session or volatility, since my Asia winners and my London winners look like two different distributions?
Shadow Trader
Posts: 5594
Joined: Wed Sep 09, 2026 5:19 pm

Re: Maximum Adverse Excursion: Using MAE to Set Better Stops

Post by Shadow Trader »

Since my last reply I split those winners by session, and the two distributions were as different as I suspected. Asia winners on EURUSD had a median MAE of 1.6 pips, while London winners sat at 2.9 pips.

A single stop across both was therefore too wide for Asia and only just right for London. Measuring MAE as a fraction of the 14 period M5 ATR at entry brought the two much closer, with both medians near 0.45 ATR.

So my stop is now a multiple of ATR rather than a fixed pip number. That one change did more than any session rule I tried.

The timing of MAE was the other surprise. About three quarters of the winners had already made their worst point within the first two minutes, so a trade still underwater at minute five was a much weaker candidate.

Spread sneaks into these figures as well. On shorts the stop triggers on the ask, so MAE measured on bid data understates the real excursion by the spread, which matters at 2 or 3 pips.

Mine was bid only for months before I noticed. Do you record MAE from bid prices, or from the side that actually triggers the stop?
Shadow Trader
Posts: 5594
Joined: Wed Sep 09, 2026 5:19 pm

Re: Maximum Adverse Excursion: Using MAE to Set Better Stops

Post by Shadow Trader »

A stop placed too close to entry will be hit by ordinary noise, and one placed too far reduces the reward-to-risk ratio. For scalpers, the right distance depends on both spread and short-term volatility.

A long position's stop triggers on the bid. At entry, the bid sits one spread below the ask, so a stop placed a few pips below entry has already lost part of its distance to the spread. When spreads widen, the effective stop distance shrinks further.

A sensible approach measures typical adverse excursion, the amount a winning trade moves against the position before reaching profit. Placing the stop beyond most of that distance avoids being stopped out on trades that would have worked. Analysis of past trades, plotting maximum adverse excursion for winners, shows where the stop should go. Adding the current spread to the stop distance keeps the effective buffer consistent. Reviewing stop placement periodically ensures that it still matches changing volatility and spread conditions.
FXS Prop Desk
Posts: 1689
Joined: Wed Oct 07, 2026 9:52 am

Re: Maximum Adverse Excursion: Using MAE to Set Better Stops

Post by FXS Prop Desk »

Fairman wrote: To learn whether a wider stop would have saved trades, you need to measure what price did after the stop-outs
Good catch, winners' MAE is censored by the stop by construction. One refinement for the stopped trades: record whether price reached the target before it hit the wider stop being considered, not just whether it got there eventually. A lot of stop-outs do come back, but only after travelling further than anyone would realistically sit through. On a funded account the comparison also has to include the daily loss limit, because a wider stop at the same lot size can turn an ordinary bad day into a breach.
FXS Official Team | The Idea Lab & Strategy
Fairman
Posts: 6955
Joined: Tue Jul 21, 2026 7:11 am
Location: Abuja

Re: Maximum Adverse Excursion: Using MAE to Set Better Stops

Post by Fairman »

That makes the counterfactual much more useful: a stopped trade that eventually reaches target is not automatically evidence for a wider stop. I would replay each original entry using the actual bid/ask and candidate stop levels fixed before inspecting outcomes. For every stopped case, record which barrier—the target, proposed wider stop, or daily loss floor—would have been reached first, and when; include spread, commission, and realistic slippage. This separates a brief excursion that recovered from one requiring the trader to sit through another adverse leg. If lot size stays constant, a wider stop raises the loss per trade. If risk stays constant by reducing size, exposure and the return distribution change instead. Compare net expectancy, drawdown, and breach frequency under both policies, not merely the number of winners rescued. Session or volatility buckets may explain differences, but need adequate samples and a holdout check; a threshold learned from one period can just encode that period's noise. The practical question is whether the extra room improves the full rule set after costs without making account-level limits more likely to be hit.
It’s Fairman :geek:
Post Reply