MAE/MFE: what your trades did while you weren't looking.

The key idea

The two shadows

Adverse excursion audits stops. Favorable excursion audits exits.

Maximum Adverse Excursion is the deepest a trade traveled against entry before resolving; Maximum Favorable Excursion is the furthest it ran in your favor. Winners that routinely survive deep MAE say the stop is wider than the trade needs. Exits that consistently capture a small fraction of MFE say the market offered more than the management style collected. Neither shows up in P&L — both are alpha sitting on the table. Both shadows are recorded automatically at close, because a figure that depends on the operator remembering to note it is a figure that will be missing exactly when it matters.

Efficiency, quantified

Captured R over offered R is an honesty ratio.

The lab expresses execution efficiency as the share of available excursion actually monetized, per branch. This is where management styles get judged by evidence: Trend Partial's early monetization versus Trend No-Partial's full-exposure patience stop being philosophy and become measured trade-offs. A branch capturing a thin slice of what it is offered has a management problem, not a selection problem — and those need different repairs. The ratio is diagnostic rather than a target: pushing capture toward one hundred percent would require exits nobody can time.

Giveback

The distance between peak profit and realised profit is a managed number.

Giveback is MFE minus the R actually booked — the open profit a trade showed and then surrendered before exit. Some giveback is unavoidable and even correct: a trailing method that never gives anything back is a method exiting far too early. The question the lab makes answerable is how much giveback each branch is paying and what it is buying. A branch surrendering a large share of its peak to catch occasional outsized runners may be making an excellent trade; a branch surrendering the same share and catching no runners is simply exiting late.

Excursion is branch-specific evidence

A stop that is wrong for one doctrine is right for another.

Aggregating excursion across a whole book produces an average that describes none of it. Normal-branch trades resolve quickly and their MAE distribution is tight; trend branches survive deeper adverse excursion by design, because giving the trade room is the entire premise. Pooling them yields a stop recommendation too wide for one and too tight for the other. The lab therefore reports per branch, and the discipline that follows is to resist the tidy single answer — four branches will produce four different correct stops, and that is the finding, not a failure to converge.

Entry quality and exit quality fail separately

Excursion data separates two problems that P&L merges.

A disappointing trade has at least two distinct possible causes, and the closing P&L cannot tell them apart. If adverse excursion was shallow and favourable excursion was large, the entry was good and the exit collected too little — a management problem. If adverse excursion was deep and favourable excursion never developed, the entry was poor and no exit method would have rescued it — a selection problem. These require opposite responses, and an operator working from outcomes alone will frequently apply the wrong one, tightening entry criteria when the real leak was an early exit. Recording both shadows is what makes the distinction observable, and it is the reason the lab reports them as a pair rather than separately.

The repair loop

Excursion evidence turns tweaks into engineering.

Because MAE/MFE is recorded per trade and analyzed per branch, adjustments become testable hypotheses: widen this branch's stop by a defined ATR increment and the MAE distribution predicts the survival gain; move the partial target and the MFE capture ratio predicts the cost. The lab closes the loop between what trades did and what the rules should become.

FigureThe excursion repair loop

Each pass turns a management instinct into a testable hypothesis with a measured outcome.

Connected inside MARS

The Formula Family →

MAE/MFE Execution Efficiency Lab →

Execution Efficiency diagnostics →

Branch Logic →

Every brief documents the same shipped system.

The complete MARS package — eleven workbooks, three TradingView indicators, the full manual library — $497.

Get MARS Pro — $497 Explore the Platform