MAE / MFE — P&L
Average maximum adverse excursion of WINNING trades (how deep winners dipped before recovering). Shallow = stops can be tight. In account currency.
- Computed from
- Trades list
- Scope
- Single report
- Range
- Any real number
- Direction
- Context-dependent
MAE and MFE — Maximum Adverse Excursion and Maximum Favorable Excursion — describe what happened while a trade was open, not just how it ended. MAE is the worst point the trade reached against you; MFE is the best point it reached in your favor. Together they expose entry timing, stop placement, and exits.
How it's read
Every trade traces a path between its open and close. MAE and MFE are the two extremes of that path:
The diagnostic power comes from splitting by outcome:
| Value | Reading | Notes |
|---|---|---|
| winners, shallow MAE | Good entries | Trades barely went against you before working — stops can be tight. |
| winners, deep MAE | Style-dependent | For a tight-entry system, winners that dipped deep are fragile — a stop just inside that dip would have killed them. For mean-reversion or scale-in systems, deep-MAE winners are by design. Read against your intent. |
| losers, large MFE | Gave it back | Trades were nicely in profit, then turned into losses — exit problem. |
| losers, small MFE | Wrong from the start | Never worked — an entry/signal problem, not an exit one. |
Where to put your stop and target
The most actionable use of MAE/MFE is placement. Plot the MAE of your winners: it shows how far trades that ultimately worked typically dipped against you first. Your stop belongs just beyond that cluster — wide enough not to knife your real winners, tight enough to cut trades that never recover. A stop inside the winners' MAE cluster is killing trades that would have paid. The MFE of your winners is the mirror: it shows how far favorable moves typically run, which is where a take-profit or trailing stop belongs.
The classic picture is a scatter: per trade, plot MAE on one axis against the final result on the other. Winners cluster at shallow MAE, losers spread across the bottom — and the gap between the two clouds is exactly where your stop wants to sit.
Derived measures
Two ratios summarize exit quality across all trades. In words: Profit Captured is, on your winning trades, how much of the best-available profit (the MFE) you actually kept; Loss Taken is, on your losers, how much of the worst-case dip (the MAE) you actually ate. Both are shown 0–100% (capped at 100% — you can't keep more than the move offered).
Profit Captured (MFE capture) = mean(realized / MFE) over winners Loss Taken (MAE capture) = mean(|loss| / |MAE|) over losers
Worked example
A winning trade enters at 1.1000, dips to 1.0980 (MAE = −20 pips), peaks at 1.1070 (MFE = +70 pips), and closes at 1.1050 (+50 pips). Its Profit Captured is 50/70 ≈ 71% — about seven-tenths of the available move was kept, and the shallow 20-pip MAE (the trade barely went against you before working) says the entry was well-timed.
Pitfalls
- Needs clean M1 data, and understates the true extreme. Excursions are reconstructed from minute bars, so the real intra-minute worst (or best) tick is at least as far as the bar shows, often further — treat MAE/MFE as a conservative floor, not the exact tick. Gaps or a missing symbol make them unreliable; EquityTruth flags trades it can't reconstruct.
- Sub-minute trades are coarse. A trade open for seconds can't have a meaningful intra-trade path at M1 resolution.
- Averages hide bimodality. "Average MAE of winners" can blur two very different trade clusters. Look at the scatter, not just the mean.
Why it matters
Most performance metrics only see the closed result. MAE/MFE is the only family that sees the journey — which is where entry edges, stop discipline, and exit efficiency actually live. A system with a great profit factor but terrible Profit Captured is leaving a second strategy's worth of money on the table.