Quick answer: For a fixed-size winning trade with positive MFE, divide realized price profit by maximum favorable excursion over the same entry-to-exit interval, then multiply by 100. Keep quantity, units and costs consistent. Report execution costs separately, use actual position accounting for partial exits, and never interpret the ratio as proof that the peak was available to sell.

What exit efficiency measures—and what it does not

In this guide, exit efficiency means the share of an observed favorable price excursion retained at exit. It is a backward-looking description, not a rating of whether a decision was sensible. Other journals may use different definitions, so save the formula beside the number.

The existing MAE and MFE guide explains how to measure the price path. This article addresses a different question: how to compare that path with fills, aggregate the comparison and test an exit change without inventing perfect timing.

A high ratio does not establish a profitable strategy. A series of tiny wins captured near their peaks can still be overwhelmed by larger losses. A lower ratio can be compatible with a trend-following approach that accepts pullbacks to remain in larger moves. Evaluate expectancy, losses, costs and risk alongside the ratio.

The exit-efficiency formula for unchanged position size

Gross price capture (%) = realized price profit ÷ MFE × 100

Use per-share dollars, total dollars or R in both parts—not a mixture.

For a long trade, realized price profit per share is exit minus entry, and MFE per share is the highest recorded price while open minus entry. For a short trade, realized price profit is entry minus cover price, and MFE is entry minus the lowest recorded price while open. Floor MFE at zero if the position never moved favorably.

These formulas assume the same share count from entry through exit. Use the same observation interval in numerator and denominator. A price reached after the exit is not this trade's MFE; it belongs in a separately labeled research window.

Long example: 66.7% is a measurement, not a verdict

Suppose 100 shares are bought at $50, reach $56 while open and are sold at $54. Gross realized price profit is $400, and price-path MFE is $600. The capture ratio is $400 ÷ $600 × 100 = 66.7%.

If total explicit trading costs are $10, net realized profit is $390. Net profit divided by the same gross MFE is 65%. Label that second number a net-to-gross capture ratio, rather than silently comparing it with a cost-free benchmark. Slippage already reflected in actual fills must not be subtracted a second time.

Nothing here proves that the missing $200 was avoidable. The high may have preceded any exit signal, lasted briefly or lacked sufficient liquidity. The separate question is whether a repeatable exit rule could have retained more after realistic costs.

Short example: reverse the price direction

Assume 100 shares are shorted at $80, trade as low as $72 while open and are covered at $75. Gross price profit is ($80 − $75) × 100 = $500. MFE is ($80 − $72) × 100 = $800. Gross capture is 62.5%.

Track borrow charges, dividends owed and other applicable costs separately before calculating net results. Investor.gov’s Introduction to Short Sales explains the risk of potentially unlimited losses as price rises. A favorable capture score says nothing about whether the original short risk was acceptable.

Keep three records instead of one hindsight score

  1. Actual result: cash profit or loss from recorded fills after specified costs.
  2. Rule-based alternative: a modeled result from a fully specified exit rule, with execution assumptions written down.
  3. Observed MFE: the most favorable price excursion during the defined window, not a promised fill.

A rule-based alternative is still a simulation. “Sell near the top” is not a rule. “Exit the next session after a daily close below the prior two-session low” can be tested, provided the calculation uses only information available when the order would be sent. State whether the modeled order is market, limit or another supported type.

Why a chart high is not a guaranteed exit

Investor.gov's Understanding Order Types bulletin, updated August 18, 2026, explains that market fills can differ from the last-traded price, larger orders can fill at several prices and limit orders may not execute. Those execution limits matter when turning a chart extreme into a hypothetical exit.

The same bulletin explains that a standard triggered stop becomes a market order and its stop price is not a guaranteed fill. Record the actual fill when a trailing exit gaps through its trigger. Do not award the backtest a price that an order could not reasonably obtain.

  • Use the same session coverage and bar resolution across comparable trades.
  • Do not include extended-hours extremes if the strategy could not trade then.
  • Flag wide spreads, halts and isolated prints.
  • Use intraday sequencing when a daily bar contains both entry and an extreme; do not assume which came first.
  • Record the data source and how splits or distributions were handled.

Partial exits need position accounting, not a larger multiplier

Do not multiply the eventual high by the original full position after some shares have already been sold. For scaling strategies, keep the simple fixed-size ratio separate. One useful alternative is a clearly labeled trade-equity capture ratio: final net trade P&L divided by the highest positive marked trade P&L reached while open.

Marked trade P&L at time t = realized P&L to t + unrealized P&L on shares still held − costs accrued to t

Trade-equity capture (%) = final net P&L ÷ peak positive marked trade P&L × 100.

This convention measures giveback along the actual sizing path. It is not the same as asking what the untouched original position might have earned, and it should not be pooled with fixed-size price capture without disclosure.

Example before costs: buy 100 shares at $50. Sell 50 at $54, realizing $200. When the remaining 50 reach $58, their unrealized gain is $400, so total marked profit is $600. Sell those 50 at $55 for another $250. Final profit is $450; if $600 was the peak, trade-equity capture is 75%. Multiplying the $8 peak move by all 100 original shares would create a different, counterfactual $800 benchmark.

With additions, reversals or multiple entry lots, use timestamped fills and a consistent cost-basis method. Reconcile the final result to the broker record. Marking rules also matter: a last-trade mark and a liquidation-side bid or ask estimate are different conventions.

Zero MFE, losses and tiny denominators

When MFE is zero, the capture ratio is undefined. Record N/A, not zero percent. If MFE is positive but the trade ends at a loss, a signed ratio can describe the giveback; for example, −$100 divided by $200 MFE equals −50%. Label losses separately rather than calling them ordinary winning-trade efficiency.

Tiny MFE values make ratios unstable. A few cents of spread or fees can dominate the number. Set a documented minimum positive denominator for ratio analysis, retain all trades in the overall P&L record and disclose exclusions. Never delete unattractive trades to improve the headline score.

Average ratios and pooled capture answer different questions

Consider two fixed-size winners: trade A realizes $50 from $100 MFE, while trade B realizes $900 from $1,000 MFE. Their individual ratios are 50% and 90%.

  • Mean per-trade ratio: (50% + 90%) ÷ 2 = 70%. Each trade gets equal weight.
  • Pooled dollar capture: ($50 + $900) ÷ ($100 + $1,000) = 86.4%. Larger dollar excursions receive more weight.

Neither result is inherently the correct headline. Report the method, sample count and trade population. Do not divide winners' profit by MFE from all trades or compare a net numerator with an undisclosed gross denominator. Separate directions, setups and exit styles before drawing conclusions.

There is no universal good exit-efficiency percentage

Choosing a target such as 80% without context encourages fitting exits to known peaks. A tight target rule and a wide trailing rule pursue different outcomes. The relevant test is whether a proposed change improves the full strategy under consistent risk and execution assumptions—not whether it raises one capture ratio.

Measure net expectancy in R, drawdown, loss size, holding time and transaction costs as well. Keep the initial risk denominator fixed as explained in the position-sizing guide. An exit that captures a higher fraction but consistently cuts substantial winners short may make the overall strategy worse.

A practical test that limits hindsight bias

  1. Freeze the original strategy, sample definition, trade timestamps and current exit rule.
  2. Choose one specific alternative, such as a target-and-trail combination, before inspecting a later evaluation sample.
  3. Define signal timing, order type, costs and missing-data treatment.
  4. Simulate each rule through its own exit. If it exits later, obtain later price data rather than stopping at the original exit.
  5. Compare capture within each rule's holding window and compare net outcomes over the same candidate entries. The MFE denominators may differ when exits differ.
  6. Validate on trades not used to select the rule, and monitor subsequent results without retuning after every loss.

Use the hindsight-resistant trade-review checklist to grade whether the written exit was followed. For stalled trades, the time-stop guide provides a different management tool; it should not be added merely because a completed chart makes one exit look attractive.

Journal fields to retain

  • Setup, direction, entry/exit timestamps and session coverage.
  • Initial risk, quantity changes, actual fills and costs.
  • MFE value, units, time window and price-data resolution.
  • Gross price capture or separately labeled trade-equity capture.
  • Net result, reason for exit and whether the plan was followed.
  • Modeled alternative, its execution assumptions and any ambiguous bars.
  • Sample exclusions and the reason for each exclusion.

Bottom line

Exit efficiency is useful when its numerator, denominator and observation window are explicit. Start with realized profit divided by MFE for unchanged-size trades, use actual marked position accounting for scaling, and separate theoretical peaks from modeled executable alternatives. Improve a rule only when the broader evidence supports it—not because one trade missed its high.

Start with MAE and MFE formulas → Review the decision and execution → Browse all Insights →

Educational analysis only—not individualized investment advice, a recommendation to trade or a promise of improved returns. Examples are hypothetical and omit costs where explicitly stated. Historical prices and modeled exits do not guarantee future results or executable fills. Trading can cause loss of principal; short sales can produce losses beyond the initial proceeds.