Magic numberMagicNumber
- Default
- 523201
- Rentang yang disarankan
- —
The ID the EA stamps on its own orders so it never touches positions it did not open. Change it only if another EA on the same account happens to use the same number.
Max slippageSlippage
- Default
- 3
- Rentang yang disarankan
- —
How far (in pips) a market order may fill from the requested price before it is rejected. It does not affect the backtest.
One entry per barOncePerBar
- Default
- true
- Rentang yang disarankan
- —
The cross is evaluated once per closed H4 bar. The published record was measured this way; turning it off changes the strategy.
Tenkan period (slot 1)Ichi_Tenkan_i1
- Default
- 7
- Rentang yang disarankan
- —
The Tenkan-sen period the cross is read from. 7 is the measured setting — a faster or slower Tenkan is a different EA.
Kijun period (slot 1)Ichi_Kijun_i1
- Default
- 26
- Rentang yang disarankan
- —
The Kijun-sen period on the first indicator slot. Keep the two slots identical (see below).
Senkou period (slot 1)Ichi_Senkou_i1
- Default
- 52
- Rentang yang disarankan
- —
Part of the Ichimoku handle but not read by the rule — the cloud plays no part in the signal. Kept at the standard 52 so the handle matches the measured build.
Tenkan period (slot 2)Ichi_Tenkan_i2
- Default
- 7
- Rentang yang disarankan
- —
The build creates two Ichimoku handles with the same periods and reads Tenkan from one, Kijun from the other. Keep this equal to slot 1.
Kijun period (slot 2)Ichi_Kijun_i2
- Default
- 26
- Rentang yang disarankan
- —
The Kijun-sen period on the second indicator slot — the line the cross is read against. Keep the two slots identical.
Senkou period (slot 2)Ichi_Senkou_i2
- Default
- 52
- Rentang yang disarankan
- —
Same as slot 1 — present in the handle, not part of the rule.
Lot sizeLotSize
- Default
- 0.1
- Rentang yang disarankan
- —
The lot per position — 0.1 lots is 0.1 BTC of exposure, and every trade in the published record was filled at exactly that size. Bitcoin's price makes the notional large; scale it to your balance before going live.
Buy stop-loss (% of price)SL_PricePct_o1
- Default
- 2.9655
- Rentang yang disarankan
- —
The stop distance for buys as a percent of the entry price — not pips. It is wide because Bitcoin is volatile, and it scales with the BTC price, which is how one setting held across the $7,000-$120,000 range of the test window.
Buy take-profit (% of price)TP_PricePct_o1
- Default
- 0.0
- Rentang yang disarankan
- —
Zero means no fixed target — the exit is the trailing stop below. The published record was measured with no take-profit.
Sell stop-loss (% of price)SL_PricePct_o2
- Default
- 2.9655
- Rentang yang disarankan
- —
The stop distance for sells, identical to the buy side.
Sell take-profit (% of price)TP_PricePct_o2
- Default
- 0.0
- Rentang yang disarankan
- —
Zero, same as the buy side — no fixed target.
Body average lookbackRangeSizePeriod_rsfilt
- Default
- 14
- Rentang yang disarankan
- —
How many closed H4 bars the average candle body is taken over. 14 four-hour bars is about two and a half days of Bitcoin trading.
Body expansion multipleRangeSizeMult_rsfilt
- Default
- 1.3
- Rentang yang disarankan
- —
The signal bar's body must be at least this multiple of the 14-bar average for a cross to be taken. 1.3 is the measured setting — lower it and the EA trades quiet crosses the record never took.
Max positionsMaxPositions_g2
- Default
- 1
- Rentang yang disarankan
- —
The EA holds at most one position in total. A new signal while a position is open is ignored — no stacking, no hedging.
Trailing step (pips)TrailStep_trail1
- Default
- 1
- Rentang yang disarankan
- —
The stop is only moved when the improvement is at least this step. 1 keeps the trail tight; it is part of the measured build.
Trail distance (% of entry price)TrailPricePct_trail1
- Default
- 0.8896
- Rentang yang disarankan
- —
Once price has moved at least this far in the trade's favour, the stop follows at this distance — 0.8896% of the entry price. This is the exit of the strategy: it converts an open BTC move into a locked-in result instead of waiting for a fixed target.