Maximum Adverse Excursion: The Number That Shows How Much Heat You Actually Take

A trade log records the entry, the exit, and the result. It does not record what happened in between — how far the trade went against you before it turned around, if it turned around at all. Two winning trades with identical entries, exits and P&L can have taken wildly different amounts of heat to get there, and a journal built only from closed prices cannot tell them apart.

What MAE actually is

Maximum adverse excursion (MAE)is the worst unrealized loss a trade reaches while it's open, measured from entry to whichever point the price moved furthest against the position before the trade closed. It has nothing to do with how the trade ended. A trade that closes at +2R can still have an MAE of −1.8R if price dropped that far before reversing — the win rate and the P&L column show a clean win, and the MAE is the only number that shows how close it came to being a loss.

Expressed in R, so it's comparable across trades with different stop distances:

MAE (R) = worst unrealized loss during the trade ÷ initial risk (entry − stop)

Why a winning trade can still have a bad MAE

A trade that goes straight to target and a trade that drops to −1.9R before recovering to the same target are the same line in a P&L report and completely different trades in every other sense. The second one was one tick from being a stop-out. If that pattern repeats — wins that routinely sit deep underwater before turning around — the strategy isn't being saved by good entries, it's being saved by a stop that happens not to get hit, and that's a coin flip dressed up as a result.

This is also where survivorship creeps into a trade log unnoticed: the trades that went to −1.9R and kept going show up as losses, sitting right next to the ones that went to −1.9R and reversed, which show up as wins. Nothing in the closed P&L distinguishes a system that has a real edge from one that's repeatedly gambling on the same knife-edge recovery and only sometimes getting it.

What it reveals about stop placement

Plot MAE against the stop distance for a batch of trades and a pattern usually shows up fast: either MAE clusters well short of the stop, meaning the stop has slack it doesn't need, or it clusters right up against the stop, meaning the position is one tick of noise away from being taken out on trades that would otherwise have worked. Neither is visible from win rate or average R — both come from the same summary stats a tighter or looser stop would produce, right up until the sample is large enough to separate luck from structure.

• MAE consistently small relative to the stop — the stop is further away than the setup actually needs, which caps position size for no risk-reduction benefit.
• MAE consistently close to the stop, trade still wins— the setup is being rescued by noise tolerance it doesn't reliably have; expect this to look worse over the next hundred trades than it does over the last hundred.
• MAE past where the stop should logically sit— the stop was moved after entry, which means this trade isn't testing the strategy's real risk at all.

The number to log alongside it: MFE

Maximum favorable excursionis the mirror of MAE — the best unrealized gain a trade reached before it closed. A trade that peaks at +3R and closes at +0.8R is an exit problem, not an entry problem, and it looks identical to a trade that only ever reached +0.8R unless MFE is logged separately. Reviewing MAE without MFE tells you how much risk a trade actually carried; reviewing both tells you whether the risk was managed well going in and whether the reward was captured well going out — two different failure points that a single P&L number collapses into one.

Why this is hard to track by hand

MAE and MFE both require the price path during the trade, not just the entry and exit prices — something a broker statement or a manually kept spreadsheet doesn't carry. Getting either number means pulling tick or bar data for the exact window a position was open and finding the extremes within it, trade by trade, which is impractical to do by hand past a handful of trades. getALPHAcomputes MAE and MFE automatically from your MT5 history against the market data for each trade's actual open window, so the heat a trade took — and the profit it gave back — shows up next to the entry and exit instead of disappearing the moment the position closes.