FXEA365 Donchian Trend Engine
- Experts
-
Kenichiro Sakamoto
I'm Kenichiro Sakamoto, the developer behind FXEA365 — a studio building transparent, rule-based Expert Advisors for MetaTrader 5. - Version: 1.26
- Updated: 10 October 2026
FREE — the complete Donchian trend engine we use in our paid EAs. No demo limit, no trial period, no crippled features.
Looking for a breakout EA built for gold? See our paid EA Gold Viper Breakout (XAUUSD M30, free demo in the Strategy Tester): https://www.mql5.com/en/market/product/182297
=== START HERE ===
1) Read the RISK WARNING below before anything else.
2) Attach the EA to a USDJPY H1 chart — on any other chart it will place no trades.
3) Test the free demo in the Strategy Tester first: see “5-MINUTE DEMO CHECKLIST” below.
4) Independent forward results: see “INDEPENDENT VERIFICATION (FXEA365)” below.
5) Platform: MetaTrader 5 (this page). MetaTrader 4 edition: https://www.mql5.com/en/market/product/185537
=== REQUIREMENTS ===
- Symbol / timeframe: USDJPY H1
- Account type: hedging. On a netting account the EA prints a warning at start-up and position management may not work as designed.
- Broker: any broker that offers the symbol above
- VPS: strongly recommended so the EA runs through the 24/5 trading week without interruption
MT5 EDITION — This is the MetaTrader 5 build of Donchian Engine. An MT4 edition is published separately on this account. Each build is native to its own platform. The MetaTrader 5 build additionally offers a selectable trading timeframe; the MetaTrader 4 build does not. Pick the one that matches the terminal you trade on.
IMPORTANT — attach the EA to a USDJPY H1 chart. The strategy reads that timeframe; on any other chart it will simply place no trades. If you are testing the free demo in the Strategy Tester, set the symbol and timeframe to USDJPY H1 or you will see zero trades.
We are FXEA365. We publish real-tick backtests and we report equity drawdown, not the flattering balance drawdown. This is the engine at the core of our trend products, released free so you can test it yourself instead of taking our word for it.
=== WHAT IT DOES ===
- Entry: the bar closes beyond the N-bar Donchian channel (a genuine breakout, evaluated on closed bars only — no repainting).
- Stop-loss: a hard stop on EVERY trade, placed at SL_ATR x ATR. Never removed, never widened.
- Exit: ATR trailing stop (Trail_ATR x ATR), plus a time-out after HoldBars if the move never develops.
- One position at a time. No martingale. No grid. No averaging down. No hidden lot doubling.
Because the entry and the stop are both scaled by ATR and the channel, the logic is broker-agnostic: it does not depend on a particular spread, feed or symbol suffix.
=== DEFAULT SETTINGS ===
The defaults are our USDJPY H1 configuration: N_Channel 40, SL 2.0 ATR, Trail 8.0 ATR, HoldBars 60, long-only, risk 0.5% per trade.
Attach it to a USDJPY H1 chart to reproduce our numbers. On other symbols and timeframes you must find your own settings — a breakout length that works on one market is not transferable, and we would rather say so than pretend otherwise.
=== HONEST LIMITS ===
- Trend-following. It loses money in ranges, wins in trends. Expect long flat periods and a hit rate below 50% with winners larger than losers. That is the shape of this edge.
- Donchian breakouts work on strongly trending markets (JPY crosses, crypto majors). Our own tests on real broker feeds show they do NOT work on ranging majors like EURUSD or on gold. We tell you this so you do not waste weeks discovering it.
- No live forward record is attached to this free product yet.
=== WHO THIS IS FOR / NOT FOR ===
Right fit: traders who want to verify our claims themselves before buying anything, anyone who wants a clean, readable trend engine as a baseline to measure their own ideas against, and anyone who would rather be told where a method fails than sold on where it works.
Not a fit: anyone looking for a finished product with a live track record — there is none attached to this free EA yet — anyone who will run the USDJPY defaults on EURUSD or gold, where our own tests say Donchian breakouts do not work, and anyone who needs a win rate above 50%.
=== 5-MINUTE DEMO CHECKLIST ===
1) Open the Strategy Tester. Symbol USDJPY, timeframe H1, defaults untouched: N_Channel 40, SL 2.0 ATR, Trail 8.0 ATR, HoldBars 60, long-only, risk 0.5%. On any other symbol or timeframe it places no trades.
2) Modelling: "Every tick based on real ticks". Report equity drawdown, not balance drawdown — the flattering one is why we say so explicitly.
3) Run several years, not several months. A trend engine judged on a short window tells you about that window and nothing else.
4) Now do the thing the free version exists for: run the identical settings on EURUSD and on gold. They should perform badly. We say Donchian breakouts fail on ranging majors and on gold, and this is how you confirm that for yourself rather than taking our word for it.
5) Check the stop behaviour in visual mode: a hard stop on every trade from the moment it opens, never removed, never widened, one position at a time. If you find otherwise, tell us.
6) Then run it on a demo account for a month. It costs nothing, and it is the only way to see the long flat periods that the equity curve compresses into a straight line.
This is the engine at the core of our paid trend products, released free so you can test it instead of taking our word for it.
=== REAL-TICK BACKTEST (this exact free build) ===
USDJPY H1, 2017-01 to 2026-06 (9.4 years), Dukascopy real ticks, every-tick model, $10,000, default settings, 0.5% risk:
- Profit factor: 1.50
- Net profit: +$16,522 (account x2.65)
- Maximum equity drawdown: 10.3%
- Trades: 631 Win rate: 34.5%
A 34.5% win rate is normal for breakout trend following: most trades are small losses, a few large winners carry the result. Run the same test yourself before you trust any of it.
=== WHAT THE PAID EAs ADD ===
This engine is the honest core. Our paid products add the parts that took the longest to build:
- Calibrated settings per market, validated on real-tick data across multiple brokers.
- A risk ladder (defensive / standard / aggressive) with measured drawdown at each level.
- An economic-calendar filter, an account-level safety guard, and a chart dashboard.
- Multi-pair and multi-asset portfolio versions.
Trend following works best when the higher timeframe agrees and every trade has a defined exit. These are the paid EAs where we apply that discipline to gold, with the test conditions published on each page:
- Gold Viper Breakout (XAUUSD M30, MT5): https://www.mql5.com/en/market/product/182297
Trades gold only while volatility is expanding and the EMA50/200 trend agrees; every basket is capped by a group stop that the EA executes (keep it running on a VPS). No martingale. Backtest on Exness tick history (history quality 8%, not a 100% real-tick test), 2019.01-2026.09, $10,000: Defensive mode profit factor 1.74, equity drawdown 21.3%, 2,819 trades; Standard (the default) profit factor 1.51, equity drawdown 30.8%.
- Gold Neuron (XAUUSD, MetaTrader 5 only): https://www.mql5.com/en/market/product/187329
Four rule-based decision engines in one EA: a trend core, a momentum breakout, a gradient-boosting entry gate and an M15 Donchian breakout. No martingale, but two engines add same-size positions against the market, closed by the EA's group stop. Backtest of the current v3.15 build (XM real ticks, Ultra = default, 2022.01-2026.08): profit factor 1.22, maximum equity drawdown 27.68%, 611 trades. Long window 2004.01-2026.07 (100% real ticks, two core engines only): Standard mode profit factor 1.34, maximum equity drawdown 29.49%.
Every EA above has a free demo. Run it in the Strategy Tester with the settings listed at the top of its page before you decide. These are backtest figures, not live trading results, and they do not guarantee future performance.
All of them on one page: https://www.mql5.com/en/users/app.develop.sk/seller
=== INDEPENDENT VERIFICATION (FXEA365) ===
We publish this EA's full verification record on our own site - the same real-tick
backtest, the out-of-sample split, and equity drawdown (not the flattering balance
drawdown). Check it before you buy, not after:
https://fxea365.com/en/ea/donchian-engine-free
=== FREE UTILITIES (also 100% free) (FXEA365) ===
These are free, work with any EA including the one on this page, and are the same risk plumbing we run ourselves.
- Falcon Trailing Stop Manager: https://www.mql5.com/en/market/product/182633
- Aegis Account Protector: https://www.mql5.com/en/market/product/182632
- Spread Guard Live Spread: https://www.mql5.com/en/market/product/185748
- Sentinel News Filter: https://www.mql5.com/en/market/product/182634
Why we released it, and the 9.4-year real-tick numbers: https://www.mql5.com/en/blogs/post/772521
If you have run this EA for a while, a short review would help, good or bad. Mentioning the symbol, period and broker you used makes it far more useful to other traders.
Looking for a breakout EA built for gold? See our paid EA Gold Viper Breakout (XAUUSD M30, free demo in the Strategy Tester): https://www.mql5.com/en/market/product/182297
=== START HERE ===
1) Read the RISK WARNING below before anything else.
2) Attach the EA to a USDJPY H1 chart — on any other chart it will place no trades.
3) Test the free demo in the Strategy Tester first: see “5-MINUTE DEMO CHECKLIST” below.
4) Independent forward results: see “INDEPENDENT VERIFICATION (FXEA365)” below.
5) Platform: MetaTrader 5 (this page). MetaTrader 4 edition: https://www.mql5.com/en/market/product/185537
=== REQUIREMENTS ===
- Symbol / timeframe: USDJPY H1
- Account type: hedging. On a netting account the EA prints a warning at start-up and position management may not work as designed.
- Broker: any broker that offers the symbol above
- VPS: strongly recommended so the EA runs through the 24/5 trading week without interruption
MT5 EDITION — This is the MetaTrader 5 build of Donchian Engine. An MT4 edition is published separately on this account. Each build is native to its own platform. The MetaTrader 5 build additionally offers a selectable trading timeframe; the MetaTrader 4 build does not. Pick the one that matches the terminal you trade on.
IMPORTANT — attach the EA to a USDJPY H1 chart. The strategy reads that timeframe; on any other chart it will simply place no trades. If you are testing the free demo in the Strategy Tester, set the symbol and timeframe to USDJPY H1 or you will see zero trades.
We are FXEA365. We publish real-tick backtests and we report equity drawdown, not the flattering balance drawdown. This is the engine at the core of our trend products, released free so you can test it yourself instead of taking our word for it.
=== WHAT IT DOES ===
- Entry: the bar closes beyond the N-bar Donchian channel (a genuine breakout, evaluated on closed bars only — no repainting).
- Stop-loss: a hard stop on EVERY trade, placed at SL_ATR x ATR. Never removed, never widened.
- Exit: ATR trailing stop (Trail_ATR x ATR), plus a time-out after HoldBars if the move never develops.
- One position at a time. No martingale. No grid. No averaging down. No hidden lot doubling.
Because the entry and the stop are both scaled by ATR and the channel, the logic is broker-agnostic: it does not depend on a particular spread, feed or symbol suffix.
=== DEFAULT SETTINGS ===
The defaults are our USDJPY H1 configuration: N_Channel 40, SL 2.0 ATR, Trail 8.0 ATR, HoldBars 60, long-only, risk 0.5% per trade.
Attach it to a USDJPY H1 chart to reproduce our numbers. On other symbols and timeframes you must find your own settings — a breakout length that works on one market is not transferable, and we would rather say so than pretend otherwise.
=== HONEST LIMITS ===
- Trend-following. It loses money in ranges, wins in trends. Expect long flat periods and a hit rate below 50% with winners larger than losers. That is the shape of this edge.
- Donchian breakouts work on strongly trending markets (JPY crosses, crypto majors). Our own tests on real broker feeds show they do NOT work on ranging majors like EURUSD or on gold. We tell you this so you do not waste weeks discovering it.
- No live forward record is attached to this free product yet.
=== WHO THIS IS FOR / NOT FOR ===
Right fit: traders who want to verify our claims themselves before buying anything, anyone who wants a clean, readable trend engine as a baseline to measure their own ideas against, and anyone who would rather be told where a method fails than sold on where it works.
Not a fit: anyone looking for a finished product with a live track record — there is none attached to this free EA yet — anyone who will run the USDJPY defaults on EURUSD or gold, where our own tests say Donchian breakouts do not work, and anyone who needs a win rate above 50%.
=== 5-MINUTE DEMO CHECKLIST ===
1) Open the Strategy Tester. Symbol USDJPY, timeframe H1, defaults untouched: N_Channel 40, SL 2.0 ATR, Trail 8.0 ATR, HoldBars 60, long-only, risk 0.5%. On any other symbol or timeframe it places no trades.
2) Modelling: "Every tick based on real ticks". Report equity drawdown, not balance drawdown — the flattering one is why we say so explicitly.
3) Run several years, not several months. A trend engine judged on a short window tells you about that window and nothing else.
4) Now do the thing the free version exists for: run the identical settings on EURUSD and on gold. They should perform badly. We say Donchian breakouts fail on ranging majors and on gold, and this is how you confirm that for yourself rather than taking our word for it.
5) Check the stop behaviour in visual mode: a hard stop on every trade from the moment it opens, never removed, never widened, one position at a time. If you find otherwise, tell us.
6) Then run it on a demo account for a month. It costs nothing, and it is the only way to see the long flat periods that the equity curve compresses into a straight line.
This is the engine at the core of our paid trend products, released free so you can test it instead of taking our word for it.
=== REAL-TICK BACKTEST (this exact free build) ===
USDJPY H1, 2017-01 to 2026-06 (9.4 years), Dukascopy real ticks, every-tick model, $10,000, default settings, 0.5% risk:
- Profit factor: 1.50
- Net profit: +$16,522 (account x2.65)
- Maximum equity drawdown: 10.3%
- Trades: 631 Win rate: 34.5%
A 34.5% win rate is normal for breakout trend following: most trades are small losses, a few large winners carry the result. Run the same test yourself before you trust any of it.
=== WHAT THE PAID EAs ADD ===
This engine is the honest core. Our paid products add the parts that took the longest to build:
- Calibrated settings per market, validated on real-tick data across multiple brokers.
- A risk ladder (defensive / standard / aggressive) with measured drawdown at each level.
- An economic-calendar filter, an account-level safety guard, and a chart dashboard.
- Multi-pair and multi-asset portfolio versions.
Trend following works best when the higher timeframe agrees and every trade has a defined exit. These are the paid EAs where we apply that discipline to gold, with the test conditions published on each page:
- Gold Viper Breakout (XAUUSD M30, MT5): https://www.mql5.com/en/market/product/182297
Trades gold only while volatility is expanding and the EMA50/200 trend agrees; every basket is capped by a group stop that the EA executes (keep it running on a VPS). No martingale. Backtest on Exness tick history (history quality 8%, not a 100% real-tick test), 2019.01-2026.09, $10,000: Defensive mode profit factor 1.74, equity drawdown 21.3%, 2,819 trades; Standard (the default) profit factor 1.51, equity drawdown 30.8%.
- Gold Neuron (XAUUSD, MetaTrader 5 only): https://www.mql5.com/en/market/product/187329
Four rule-based decision engines in one EA: a trend core, a momentum breakout, a gradient-boosting entry gate and an M15 Donchian breakout. No martingale, but two engines add same-size positions against the market, closed by the EA's group stop. Backtest of the current v3.15 build (XM real ticks, Ultra = default, 2022.01-2026.08): profit factor 1.22, maximum equity drawdown 27.68%, 611 trades. Long window 2004.01-2026.07 (100% real ticks, two core engines only): Standard mode profit factor 1.34, maximum equity drawdown 29.49%.
Every EA above has a free demo. Run it in the Strategy Tester with the settings listed at the top of its page before you decide. These are backtest figures, not live trading results, and they do not guarantee future performance.
All of them on one page: https://www.mql5.com/en/users/app.develop.sk/seller
=== INDEPENDENT VERIFICATION (FXEA365) ===
We publish this EA's full verification record on our own site - the same real-tick
backtest, the out-of-sample split, and equity drawdown (not the flattering balance
drawdown). Check it before you buy, not after:
https://fxea365.com/en/ea/donchian-engine-free
=== FREE UTILITIES (also 100% free) (FXEA365) ===
These are free, work with any EA including the one on this page, and are the same risk plumbing we run ourselves.
- Falcon Trailing Stop Manager: https://www.mql5.com/en/market/product/182633
- Aegis Account Protector: https://www.mql5.com/en/market/product/182632
- Spread Guard Live Spread: https://www.mql5.com/en/market/product/185748
- Sentinel News Filter: https://www.mql5.com/en/market/product/182634
Why we released it, and the 9.4-year real-tick numbers: https://www.mql5.com/en/blogs/post/772521
If you have run this EA for a while, a short review would help, good or bad. Mentioning the symbol, period and broker you used makes it far more useful to other traders.
