Megamax Donchian Trend (USDJPY H1): Free Set Files for Each Risk Tier — and What 'Relative Drawdown 58.6%' Really Means
6 August 2026, 00:00
0
32
Three ready-to-load set files for Megamax Donchian Trend on USDJPY — Defensive, Standard and Aggressive — plus a plain explanation of why the top tier shows two different drawdown numbers.
Megamax Donchian Trend (MT5): https://www.mql5.com/en/market/product/181838
MT4 version: https://www.mql5.com/en/market/product/182526
Megamax Donchian is a pure trend follower for USDJPY on H1. A closed bar breaking the 40-bar Donchian high triggers a buy (the EA is long-only by default — the yen crosses have carried a long bias for years, and symmetric selling underperformed in our tests). Every trade opens with a hard ATR stop, the stop then trails the trend at a wider ATR distance, and a time-out closes stale trades that never develop. One position at a time, no grid, no martingale, and the same engine we released for free as the Donchian Trend Engine — this is the calibrated, risk-laddered version.
The inputs that matter
- RunMode: Defensive / Standard / Aggressive / Ultra. Identical signal on every tier; only the risk fraction scales. This is what the set files change.
- N_Channel (default 40): the Donchian lookback in H1 bars. The single most important strategy input — it defines what counts as "the range" being broken.
- SL_ATR (default 2.0): the hard initial stop, in ATR. Your worst case per trade is defined the moment the order fills.
- Trail_ATR (default 8.0): the trailing distance. It is deliberately wide — a trend system that trails tightly gives back its edge by exiting every pullback.
- HoldBars (default 60): the time-out. A breakout that has gone nowhere after this many bars is cut, freeing the position slot.
- RiskPercent (default 0.5): the base risk fraction that RunMode multiplies.
The three set files
All three change only RunMode; everything else is the shipped v1.40 default. Published figures are from the product's real-tick backtest (USDJPY H1, 2017-2026, roughly 9.4 years, $10,000, MT5 strategy tester):
- MEGAMAX_DONCHIAN_Defensive.set — risk x0.5 (effective 0.25% per trade). Backtest: +71% / profit factor 1.59 / max equity drawdown 4.8%. The quiet option.
- MEGAMAX_DONCHIAN_Standard.set — baseline 0.5% per trade. Backtest: +165% / PF 1.50 / max equity DD 7.3%.
- MEGAMAX_DONCHIAN_Aggressive.set — risk x2 (effective 1.0% per trade). Backtest: +716% / PF 1.55 / max equity DD 16.4% (19.5% relative). Our recommended growth tier.
The EA's default is Ultra: +2,155% / PF 1.55 over the same window, with an absolute max equity drawdown of only ~16% but a RELATIVE peak-to-trough drawdown of 58.6%. Those two numbers describe the same equity curve: a heavily compounding system can lose 58% of a much larger peak while the dollar drawdown measured against the starting deposit stays modest. We publish both on purpose — if the 58.6% figure is the one that worries you, that is the honest one to act on, and these set files let you step down.
Download
The files are written for the MT5 build; MT4 users can copy the same values into the input dialog by hand.
How to load a set file
In the strategy tester: press Ctrl+R in MT5, select the EA, symbol USDJPY, timeframe H1, then in the "Inputs" tab right-click the parameter list, choose "Load" and select the .set file. Real-tick data reproduces the published numbers; coarser models will drift.
On a chart: drag the EA onto a USDJPY H1 chart, open the "Inputs" tab in the properties dialog, click "Load", pick the file and confirm with OK.
What to expect
The win rate is in the mid-30s. That is not a flaw — it is the shape of breakout trend following: many small stopped-out trades, a few large winners in the trending years (2021-2022 and 2024 did most of the work in this window). Expect stair-step equity with flat stretches measured in months. If a system must win most days to feel tolerable to you, a trend follower is the wrong tool regardless of whose it is.
All backtest figures above are historical simulations, not a live track record, and not a promise of future results. Trade only with money you can afford to lose.
If the set files are useful, a review on the product page helps us more than anything else.
Our measured backtest data for every EA (profit factor, equity drawdown, trade count, year-by-year results) is published at fxea365.com/ea/ranking
All of our products and free tools are listed on our seller profile: mql5.com/en/users/app.develop.sk/seller
