Thesis
Fast and slow trend regimes must agree before opening — long or short.
Fits Trend following setups — works best when the underlying market behavior matches the thesis, and breaks down when it doesn't. Pair with the backtester's walk-forward mode before committing real capital.
Trend follower with patience. Long flat periods when the two timeframes disagree, then steady runs when the slow regime turns and the fast one confirms. Drawdowns are deeper because exits wait for chop, not a clean flip.
YAML rule-set
Paste this into the Strategy Designer's Text tab, or use the dashboard button above to load it as a template.
name: multi_trend_confluence
weight: 1.0
long_only: false
indicators:
trend20: {kind: trend, period: 20, slope_lookback: 3}
trend100: {kind: trend, period: 100, slope_lookback: 8}
entry:
combine: OR
rules:
- combine: AND
direction: long
rules:
- {lhs: {indicator: trend20}, op: eq, rhs: {value: 1}}
- {lhs: {indicator: trend100}, op: eq, rhs: {value: 1}}
- combine: AND
direction: short
rules:
- {lhs: {indicator: trend20}, op: eq, rhs: {value: -1}}
- {lhs: {indicator: trend100}, op: eq, rhs: {value: -1}}
exit:
combine: OR
rules:
- {lhs: {indicator: trend100}, op: eq, rhs: {value: 0}}
risk:
trailing_stop_pct: 0.07
cooldown_bars: 8
Indicators
trend20Trend regime (period 20, slope 3)Categorical regime indicator: +1 when price is above EMA(20) AND the EMA is rising over the last 3 bars, -1 in the mirror down-regime, 0 in chop.
trend100Trend regime (period 100, slope 8)Same indicator on a much slower window — captures the broad market regime. Acts as the bigger-picture filter.
Entry conditions
The bot opens a position the bar after every condition below has fired.
- Long when BOTH trend20 = +1 AND trend100 = +1.
- Short when BOTH trend20 = -1 AND trend100 = -1.
- 8-bar cooldown after exit.
Exit conditions
Any of the conditions below will close the position.
- Exit immediately when the slower trend100 regime flips to 0 (chop).
- 7% trailing stop on the runner.
Expected behavior
Trend follower with patience. Long flat periods when the two timeframes disagree, then steady runs when the slow regime turns and the fast one confirms. Drawdowns are deeper because exits wait for chop, not a clean flip.
Equity-curve character, not a return forecast. Backtests are not promises about live performance.
Try it on your own data
Open the template in the dashboard to backtest it on BTC, ETH, or any CCXT symbol — or copy the YAML into the Strategy Designer to edit it first.
Past behavior on backtests is not a guarantee of future performance. Markets change; rule sets need re-validation. Trade at your own risk.