Skip to main content

AdaptiveTrendChannel

AdaptiveTrendChannel is a built-in TypeScript strategy from @tradejs/strategies.

It replays candles through an adaptive channel engine, watches for bullish/bearish channel flips, and sizes entries from the distance to the channel boundary used as the stop.

Entry Logic​

  1. Builds runtime state with createAdaptiveTrendChannelEngine({ initialCandles, config }).
  2. On each candle, reads runtimeState.signal and runtimeState.snapshot.
  3. Skips until a channel flip signal exists.
  4. Selects LONG or SHORT side config from the signal direction.
  5. Applies context filters through getAdaptiveTrendChannelFilterSkipCode(...).
  6. Uses the signal floor/roof as stop-loss:
    • long stop: signal.floor
    • short stop: signal.roof
  7. Computes target from ADAPTIVE_TREND_CHANNEL_TARGET_R_MULT.
  8. Sizes quantity from MAX_LOSS_VALUE / riskDistance, with FEE_PERCENT buffer.
  9. Returns entry with channel figures and adaptiveTrendChannelContext.

Entry codes:

  • ADAPTIVE_TREND_CHANNEL_BULLISH_FLIP
  • ADAPTIVE_TREND_CHANNEL_BEARISH_FLIP

Exits​

When a position exists:

  • ADAPTIVE_TREND_CHANNEL_BREAK_EXIT when ADAPTIVE_TREND_CHANNEL_EXIT_ON_CHANNEL_BREAK=true and price breaks the active channel boundary.
  • ADAPTIVE_TREND_CHANNEL_OPPOSITE_FLIP_EXIT when ADAPTIVE_TREND_CHANNEL_EXIT_ON_OPPOSITE_FLIP=true and the engine emits the opposite flip.
  • otherwise POSITION_EXISTS.

Config Parameters​

Shared runtime:

  • ENV, INTERVAL, MAKE_ORDERS, CLOSE_OPPOSITE_POSITIONS
  • BACKTEST_PRICE_MODE
  • AI_ENABLED, AI_MODE, MIN_AI_QUALITY
  • ML_ENABLED, ML_THRESHOLD

Risk and shared indicators:

  • FEE_PERCENT, MAX_LOSS_VALUE
  • MA_FAST, MA_MEDIUM, MA_SLOW
  • OBV_SMA, ATR, ATR_PCT_SHORT, ATR_PCT_LONG
  • BB, BB_STD, MACD_FAST, MACD_SLOW, MACD_SIGNAL

Channel model:

  • ADAPTIVE_TREND_CHANNEL_REGRESSION_BARS
  • ADAPTIVE_TREND_CHANNEL_ENVELOPE_BARS
  • ADAPTIVE_TREND_CHANNEL_ATR_STRETCH
  • ADAPTIVE_TREND_CHANNEL_VOLATILITY_LOOKBACK
  • ADAPTIVE_TREND_CHANNEL_TARGET_R_MULT
  • ADAPTIVE_TREND_CHANNEL_MIN_BREAKOUT_DISTANCE_PCT
  • ADAPTIVE_TREND_CHANNEL_MAX_BREAKOUT_DISTANCE_PCT
  • ADAPTIVE_TREND_CHANNEL_MIN_CHANNEL_WIDTH_PCT
  • ADAPTIVE_TREND_CHANNEL_MAX_CHANNEL_WIDTH_PCT
  • ADAPTIVE_TREND_CHANNEL_MIN_VOLUME_REL20
  • ADAPTIVE_TREND_CHANNEL_REQUIRE_CONTEXT_ALIGNMENT
  • ADAPTIVE_TREND_CHANNEL_EXIT_ON_OPPOSITE_FLIP
  • ADAPTIVE_TREND_CHANNEL_EXIT_ON_CHANNEL_BREAK
  • ADAPTIVE_TREND_CHANNEL_MAX_FIGURE_POINTS

Side configs:

  • LONG.enable, LONG.direction, LONG.minRiskRatio
  • SHORT.enable, SHORT.direction, SHORT.minRiskRatio

Signal Payload​

The strategy stores:

  • additionalIndicators.adaptiveTrendChannelContext
  • channel figures from buildAdaptiveTrendChannelFigures(...)
  • orderPlan.qty
  • orderPlan.stopLossPrice
  • one take-profit at the computed target price

Common Skip Reasons​

  • NO_ADAPTIVE_TREND_CHANNEL_FLIP
  • DEV_TRADE_COOLDOWN
  • STRATEGY_DISABLED
  • filter-specific skip code from getAdaptiveTrendChannelFilterSkipCode(...)
  • INVALID_STOP
  • INVALID_QTY
  • RISK_RATIO:<value>

Validation Notes​

Inspect the generated channel figures before comparing metrics. This strategy depends on structural channel state, so data gaps and warmup length can change early signals.