Argus Kijun Gold
- Experts
-
Yevhen Hladchenko
I am a discretionary trader turned systems developer.
After several years of trading the markets by hand - learning price action, risk management, and the psychology of execution the hard way - I reached one clear conclusion: consistency comes from process, not emotion. - Versione: 1.31
- Attivazioni: 5
Expert Advisor for MetaTrader 5
KijunMasterTwo classic Ichimoku plays, executed with machine discipline. KijunMaster buys the pullback when a trend rejects the Kijun-sen, and fades the spike when the line goes flat. Nothing else, ever.
- Hard stop on every trade
- Closed-bar signals, zero repaint
- No martingale
- No grid
- No averaging down
The idea
One line, two regimes, one EA that knows the difference
Every Ichimoku trader on the planet watches the same line. The Kijun-sen is the midpoint of the last 26 bars: the market's short-term fair value. Price leaves it, and price comes back to it. What happens at the touch depends on the regime. In a trend, the line is support and the touch is a discount. In a range, the line is a magnet and the spike away from it is an overreaction.
Most Kijun EAs pick one behavior and quietly bleed in the other regime. KijunMaster runs both as independent modules with separate magic numbers, and a strict checklist decides which regime you are actually in before a single order is sent.
Oshime-gai: buy the dip that fails
A five-point regime checklist has to pass in full. Then the EA waits for the one bar that tells the truth: price stabs through the Kijun and gets thrown back. That rejection is the entry. Shorts are the exact mirror (Modori-uri).
- Price above the Kumoand the cloud itself is bullish
- Tenkan above Kijunthe fast line confirms
- Kijun is risingflat lines are Module B's job
- Chikou span is clearno overhead traffic 26 bars back
- Rejection barlow pierces the Kijun, close recovers above it
The flat-Kijun fade: a trade with a built-in target
When the Kijun stops moving, the market is telling you where fair value is. KijunMaster measures it: if the line has not moved for N bars and price tears away by 2.5 ATR or more, that is stretch, not strength.
- Flat detectorKijun unchanged for N bars, tolerance in points
- Spike filterclose at least 2.5 x ATR away from the line
- Reversal triggerRSI comes back out of the extreme zone
- Targetthe flat Kijun itself. The market drew your TP for you
Stop beyond the spike extreme plus an ATR buffer. Fixed SL and TP from the first second: no management, no hoping.
There is also a third, optional module for purists: the classic San-yaku Kouten breakout (a fresh Tenkan/Kijun cross outside the cloud, confirmed by price and Chikou). It ships switched off, so the default profile stays pullback + fade.
Risk toolkit
The boring features that keep accounts alive
Risk-based sizing
Fixed lot, or a percent of equity computed from the actual stop distance of each trade. Your risk stays constant even when the stop is wide.
Spread filter
If your broker widens past your limit at the signal bar, the trade is skipped. No entries into news-time spreads.
Session windows
Separate GMT trading hours for the trend and flat modules, with automatic server DST handling. Trade the fade only when the market actually ranges.
Giveback stop
Once a trade has earned N x ATR, the EA refuses to give back more than your set share of the peak profit. Winners stop turning into losers.
Partial close
Bank a configurable slice of the position at a profit milestone and let the remainder ride the trailing stop.
Trade context log
Optional CSV of every entry's features: distance to Kijun, cloud thickness, slope, RSI, spread, session hour. Build your own filters on real data.
Fine print, up front
What KijunMaster refuses to do
- It will not open a position without a stop loss. Ever.
- It will not add to a losing trade. No grid, no martingale, no "recovery zones", no hidden position stacking behind a pretty equity curve.
- It will not act on a forming bar. Every signal is computed once, on the close, so your backtest and your live chart run the same logic.
- It will not promise you a smooth ride. Both strategies take losing trades; that is the cost of trading with defined risk. Anyone who tells you otherwise is selling fiction, not software.
Quick start
From download to live in four honest steps
- Backtest it on your data. Download the free demo and run the Strategy Tester on your broker's feed, real ticks, H1 or higher. Your spread and your sessions are the only ones that matter.
- Pick your module mix. Trending symbol? Module A. Ranging hours? Module B with a session window. Or run both; they never hold more than one position at a time.
- Forward-test on demo with your final settings for a few weeks.
- Go live on a VPS. One chart per symbol, a unique magic number per chart, and let the checklist do its job.
Under the hood
Key settings, in plain English
| Parameter | What it controls |
|---|---|
| Modules | |
| Module A / B / C | Switch each strategy on or off independently. Defaults: A and B on, C off. |
| Volume and risk | |
| Fixed lot / Risk % | Trade a fixed size, or risk a set percent of equity per trade based on the stop distance. |
| Max spread | Hard ceiling in points; entries are skipped above it. |
| Module A: trend pullback | |
| SL buffer, ATR | How far beyond the swing low/high the initial stop hides. |
| Trailing, ATR | Distance of the trailing stop behind the Kijun. Tightens only, never loosens. |
| Module B: flat fade | |
| Flat bars / tolerance | How long and how strictly the Kijun must sit still to count as flat. |
| Spike, ATR | Minimum stretch away from the line before a fade is even considered. Default 2.5. |
| RSI levels | The extreme zones whose exit triggers the entry. Defaults 70 / 30. |
| Protection | |
| Sessions, GMT | Separate trading-hour windows for trend and flat modules; server DST resolved automatically. |
| Giveback stop | Max percent of peak floating profit the trade may return before it is closed. |
| Partial close | Profit milestone (in ATR) and the share of the position to bank there. |
Questions traders actually ask
FAQ
Does it repaint, or peek at forming bars?No. Every signal is evaluated exactly once, on the closed bar. There is no repainting and no look-ahead, which also means the Strategy Tester runs the same logic your live chart will.
Which symbols and timeframes?The logic is symbol-agnostic and reads everything through ATR, so it adapts its distances to the instrument. It was designed around liquid, trending markets on H1 and above. Always validate on your own broker's data first: spreads, sessions and tick quality differ between brokers, and the free demo in the Strategy Tester costs you nothing.
Is it prop-firm friendly?The mechanics tick the usual boxes: a hard stop on every trade, one position at a time, no grid, no martingale, and risk sizing as a percent of equity. Check your firm's specific rules on EAs before running it in a challenge.
Do I need a VPS?Recommended. Entries happen on bar close, but the trailing stop, giveback stop and partial close work tick by tick, so you want the terminal running around the clock.
Can I run it on several symbols?Yes. Attach it to one chart per symbol and give each chart its own magic number. Each instance manages only its own positions.
Run it before you buy it
The free demo works in your Strategy Tester today. Judge the checklist on your broker's data and your symbols, not on anyone's screenshots.
Available on the MQL5 MarketTrading leveraged products carries a substantial risk of loss and is not suitable for every investor. Backtested performance does not guarantee future results. Test on demo before trading live funds.
