MAE and MFE, explained with a real example

Most journal statistics describe the trade you took. MAE and MFE describe the trade you could have taken: the full range the position travelled between entry and exit. That makes them the two most useful numbers in a journal for a specific question: not "is my strategy any good", but "am I managing it badly".

The definitions

• MAE: Maximum Adverse Excursion. The furthest a trade moved against you before it closed. How much heat you took.
• MFE: Maximum Favourable Excursion. The furthest it moved in your favour before it closed. How much was on the table at the best moment.

Both are usually expressed in R, multiples of your initial risk, rather than currency, so trades of different sizes can sit in the same distribution.

A worked example

You buy at 100.00 with a stop at 99.00. Your risk is 1.00 per unit, so 1R = 1.00. The trade dips to 99.40, rallies to 102.50, then falls back and you exit at 101.00.

• MAE = 0.60, or 0.6R. The trade came within 40 cents of your stop.
• MFE = 2.50, or 2.5R. At its best the trade was worth two and a half times your risk.
• Result = 1.00, or 1R. You captured 40% of the favourable move.

MAE: heat taken

0.6R

MFE: offered

2.5R

Kept

1.0R

Capture rate

40%

One trade tells you nothing. A hundred of them, plotted, tells you a great deal.

What the distributions actually tell you

Reading MAE

Look at your winners only, and find the MAE they typically took. If your winning trades routinely dip 0.8R against you before working, and your stop sits at 1R, you are running with almost no margin. Normal noise is stopping out trades that would have won. If winners almost never go beyond 0.3R against you, your stop at 1R is wider than it needs to be, and every loss costs three times what it has to.

Reading MFE

Compare typical MFE against your typical exit. If trades regularly reach 2.5R and you regularly exit at 1R, you are leaving most of the move behind. Unlike a lot of journal findings, that one is fixable without touching your entry criteria at all. If MFE and your realised result are close together, your exits are efficient and the improvement has to come from somewhere else.

Three ways these numbers mislead

1. MFE is not achievable profit. It is the single best tick in the trade, visible only in hindsight. Nobody exits at MFE consistently, and a strategy built on assuming you can is a backtest, not a plan. Treat closing 50–60% of typical MFE as good.
2. They are sensitive to how the trade is timestamped. If your journal only stores entry and exit prices, MAE and MFE have to be reconstructed from bar data, and the resolution of those bars sets the accuracy. Minute bars will understate the true excursion of a fast move.
3. They say nothing about frequency. A setup with beautiful excursion statistics that appears four times a year is not a strategy. Always read them next to the trade count.

How to use them this week

Filter to your winners from the last hundred trades. Find the MAE that 90% of them stayed within; that is roughly where your stop wants to be, plus a margin. Then find the median MFE and compare it to your median exit. If the gap is large, your next experiment is an exit change, not an entry change. If it is small, stop tinkering with exits and look at selection.

That is the whole value of these two numbers: they tell you which end of the trade to work on, which saves you from the far more common approach of changing everything at once and learning nothing.