Trade Excursion Report - MAE and MFE of your closed trades (script) - MetaTrader 5용 스크립트

A trade result tells you what happened at the end. It does not tell you how the trade got there. Two numbers fill the gap:

• MAE (maximum adverse excursion): the worst the price went against the position while it was open.
• MFE (maximum favourable excursion): the best the price went in the position's favour while it was open.

With them you can ask: how much drawdown do my winners really need (is my stop too tight)? How far did my losers go my way before they turned (did I give profit back)? How much of the best move did my winners keep?

Files

• TradeExcursionReport.mq5 - the script.
• TradeExcursion.mqh - the class CTradeExcursion (put both files in the same folder; the class can be used in your own code).

What the script shows

• A table in the Experts tab, newest trades first: ticket, symbol, side, lots, open and close time, result, MAE, MFE and stop distance in points.
• A summary on the chart: number of positions; for the winners the median and the 90th percentile of the MAE; the share of the best move the winners kept on average; for the losers the median and the 90th percentile of the MFE; and the share of the losers (with a stop loss) that were at +1R at some moment (the price went at least as far in their favour as the stop distance). A stop wider than the 90th percentile MAE of the winners would have kept nine of ten winners alive.
• A CSV file MQL5\Files\TradeExcursion.csv with every position (price, result, stop, MAE, MFE, realized points).

Inputs: days of history, one symbol (empty = all), one magic number (-1 = all, 0 = manual trades), rows in the table, CSV on/off, summary on the chart on/off.

How the numbers are made

• A position is all its deals: the first entry deal gives the open time, the exit deals give the close time; the result adds profit, commission, swap and fee. Partial closes count as one position. A position with several entry deals uses the volume weighted open price. Positions that are still open, or opened before the history window, are left out.
• MAE and MFE come from the 1-minute bars from the minute of the entry to the minute of the exit (bid prices). The first and the last minute can contain prices before the entry or after the exit, so both numbers can be a little generous (at most one minute of price range). Sell positions are measured on the bid as well, so their MAE is understated and their MFE overstated by about the spread.
• The stop loss is the stop of the opening order. A stop that you set or moved later is not known to the history.
• Points are the points of each symbol: when you mix symbols, read the summary as "points of each symbol" or filter one symbol.

Use in your own code

#include "TradeExcursion.mqh" CTradeExcursion ex; ex.Load(TimeCurrent() - 90 * 86400, TimeCurrent(), "", -1); // from, to, symbol ("" = all), magic (-1 = all) STradeEx t; for(int i = 0; i < ex.Total(); i++) if(ex.Get(i, t)) Print(t.id, " MAE ", t.mae / t.point, " MFE ", t.mfe / t.point);

Also: Count(), Stat(winners, which, percentile), LosersReachedOneR(), WinnersCapture(), StillOpen(), SkippedNoBars().

How it was tested

• Strategy Tester, EURUSD H1, two months: 57 closed positions with and without a stop loss, buys and sells, some with a partial close. The unchanged class was run on the tester history, and Python recomputed every number from the exported deals and the exported 1-minute bars (positions, open and close time, volume, prices, result, stop, MAE, MFE, realized distance, bars used): 627 comparisons, 0 differences.
• The script itself ran on a Strategy Tester history of simulated trades with different stops; the picture is a real screenshot of that run.
• MetaEditor: 0 errors, 0 warnings.