MAE and MFE: Journaling Trade Excursions to Cut Losses

Your entry and exit tell you what a trade did. The path in between tells you what your stops and targets should have been. Most traders throw that path away and wonder why their exits never improve.

MAE and MFE are how you keep the path. They turn every trade — winner or loser — into feedback about where you’re placing your stops too wide and your targets too close. For prop-firm traders working inside tight drawdown limits, that feedback is the difference between an edge that survives and one that bleeds out on execution.

What MAE and MFE actually measure

Two numbers, both measured against your entry, both best expressed in R so they’re comparable across trades and accounts:

MAE — Maximum Adverse Excursion. The furthest the trade went against you before you exited. A trade you closed at +1R that first dipped to −0.8R has an MAE of 0.8R. That’s how close you came to a full stop-out.
MFE — Maximum Favorable Excursion. The furthest the trade went in your favor before you exited. A trade you closed at +0.5R that first ran to +2.3R has an MFE of 2.3R — you left 1.8R on the table.

The insight lives in the gap between these excursions and your actual fills. MAE exposes stops that are wider than they need to be. MFE exposes targets that are tighter than the market was offering. Neither is visible from entry and exit alone.

If you’re rusty on expressing these as R, the R-multiple explainer covers the unit; everything below assumes MAE and MFE are logged in R.

How to record excursion without watching every tick

The objection is always the same: “I can’t sit and mark the low and high of every trade.” You don’t have to, and you shouldn’t try — manual tick-watching is both unreliable and exhausting.

Three sane ways to capture it:

From the chart, after the close. When you journal the trade, glance at the candles between entry and exit and note the extreme in each direction. Thirty seconds, good enough to see patterns.
From your platform’s trade report. Some platforms expose per-position excursion or let you reconstruct it from the price range during the hold.
Automatically. Because Shibiki ingests fills and price data directly from the broker, it can compute MAE and MFE per trade without you marking anything — the excursion is captured as part of auto-journaling, in R, ready to review.

Precision to the tick doesn’t matter here. You’re hunting for a distribution, and the pattern shows up long before rounding errors do.

Reading MAE to tighten stops that are too wide

Sort your winning trades by MAE and ask a blunt question: how much heat did my winners actually take before working?

If your winners rarely go beyond, say, 0.5R against you before turning, but your stop sits at 1R, you’re risking twice what the trade needs. Tightening the stop toward your winners’ typical MAE cuts your loss size without costing you those winners.
If winners routinely dig to 0.9R against you before running, your stop is about right — pull it in and you’ll get chopped out of good trades.

This is the highest-value use of MAE, because stop size directly sets your −1R — the denominator of everything. On a prop evaluation with a hard drawdown line, shaving unnecessary stop width is one of the few changes that improves survival and expectancy at once. Confirm any new stop still clears your minimum reward-to-risk with a risk/reward calculator before you commit to it.

Reading MFE to spot targets you’re leaving on the table

Now do the mirror analysis on MFE, across all trades:

If trades you exited at +0.5R routinely showed an MFE of +2R or more, the market was handing you profit you refused. Your targets — or your hands — are too quick.
If your exits sit close to your MFE, you’re capturing most of what’s available. Reaching for more will just convert winners into round-trips.

The trap MFE exposes is the small-win / big-loss trader who feels disciplined (“I always take profit!”) but is quietly capping the exact right-tail winners that make a distribution profitable. You can’t fix that instinct without seeing, in the data, how much you gave back.

A practical response is a partial-target-plus-runner exit, tested against your MFE distribution rather than chosen by feel.

Building an excursion scatter from journal data

Individual trades are noise. The pattern emerges when you plot every trade at once. Build a simple scatter:

X-axis: MAE (in R) · Y-axis: MFE (in R) · one dot per trade, winners and losers in different colors.

Now the whole edge is on one chart:

A vertical band of dots at high MAE that never developed MFE → trades that took a lot of heat and never paid. Candidates for a tighter stop or a filter that keeps you out of them entirely.
A cluster with high MFE but low realized exit → money you’re consistently leaving on the table; a targeting problem.
Losers with tiny MAE → you were stopped out on noise before the trade could breathe; the stop may be too tight after all.

Read together, MAE and MFE let you tune both ends of the trade from evidence instead of superstition. Dedicated journals such as Edgewonk popularized the excursion scatter, and it’s genuinely one of the best tools in trade review. The friction has always been getting clean excursion data in without hand-marking every trade — which is exactly what auto-capturing MAE/MFE from the broker fill removes, so the scatter is populated for you and folds straight into each strategy’s live edge-health picture.