Range Sip

A mean-reversion strategy that fades Bollinger Band extremes in well-defined ranging markets, targeting quick returns to the midline.

Visual Overview

Key Concept: In ranging markets, fade extremes (Bollinger Band touches) back to the mean when volatility is declining.

Strategy Logic

Overview

Trades mean reversion in horizontal ranges by fading touches of the 2-sigma Bollinger Bands when volatility is declining and orderbook shows exhaustion. Uses small grid entries with tight stops to capture quick moves back to the range midpoint.

Entry Conditions

• Well-defined horizontal range (flat Bollinger Band slope)
• Price tags 2-sigma Bollinger Band
• Declining volatility (Bollinger Band slope ≤ 0)
• Orderbook exhaustion signals (large imbalance at extremes)
• ADX < 25 (not trending)
• No breaking news or scheduled macro events
• No recent failed fades (avoid fighting acceleration)

Entry Logic:

• Long: Price at lower 2σ band with flat/declining volatility
• Short: Price at upper 2σ band with flat/declining volatility

Position Sizing

• Small grid entries: 1-2% per level
• Maximum 30% allocation per position
• Maximum leverage: 3x
• Avoid chasing multiple fades in accelerating tape

Exit Conditions

• Price reaches Bollinger Band midline (primary target)
• Price reaches 0.8x ATR profit from entry (secondary target)
• ADX rises above 25 (trend emergence)
• Bollinger Band slope turns positive (volatility expansion)
• Time stop reached: 90 minutes without mean reversion
• Breaking news or scheduled macro event
• Multiple failed fades: 3 consecutive stop-outs within 2 hours
• Stop loss: 1.2x ATR beyond extreme

Risk Management

Stop Loss

• Hard stop: 1.2x ATR beyond the band extreme
• Time stop: 30-90 minutes without mean reversion

Take Profit

• Primary target: Bollinger Band midline
• Secondary target: 0.8x ATR from entry

Position Management

• Small grid: 3-5 levels near band extreme
• Scale out: 50% at midline, 50% at 0.8 ATR
• Quick exit: If ADX rises or volatility expands

Execution Strategy

• Entry: Post-only maker orders (grid near band)
• Exit: Taker orders on target or invalidation

Performance Expectations

• Win Rate: 65-75%
• Average Hold Time: 30-60 minutes (minimum 45 minutes)
• Profit Factor: 1.6-2.0
• Max Drawdown: 8-12%
• Expected Switching Cost: 20 bps per trade
• Best Regimes: Range-bound

Configuration

Strategy parameters:

[strategy.range-sip]
enabled = true
max_allocation = 0.30
max_leverage = 3.0
risk_per_trade = 0.015  # 1.5% risk

# Bollinger Band parameters
bb_length = 20
bb_deviation = 2.0
bb_slope_threshold = 0.0  # Flat or declining

# Range parameters
max_adx = 25
atr_period = 14

# Risk parameters
stop_atr_multiple = 1.2
target_atr_multiple = 0.8
time_stop_minutes = 90
max_failed_fades = 3
failed_fade_window_hours = 2

Technical Parameters

Bollinger Bands

• Length: 20 periods
• Deviation: 2.0 standard deviations
• Slope: ≤ 0 (flat or declining)

Range Filters

• ADX: < 25 (not trending)
• ATR Period: 14 bars
• Volatility: Declining or stable

Risk Metrics

• Stop Loss: 1.2x ATR beyond band extreme
• Take Profit: Midline or 0.8x ATR
• Time Stop: 30-90 minutes

Regime Compatibility

Compatible Regimes

• Range-bound: Ideal for mean reversion

Avoid Regimes

• Trending: Bands can walk in one direction
• Event-risk: News can break ranges violently

Example Trade

Setup: BTC in tight range

• Range: $42,000 - $44,000 (well-defined horizontal)
• Bollinger Bands: 20-period, 2σ at $42,100 (lower) and $43,900 (upper)
• BB Slope: -0.02 (declining volatility)
• ADX: 18 (not trending)
• Price: Tags lower band at $42,120
• Orderbook: Large bids exhausted, asks building
• Entry: Long at $42,150 (small grid: $42,120, $42,150, $42,180)
• ATR: $400
• Stop Loss: $41,670 (1.2x ATR = $480 below entry)
• Target: $43,000 (midline) or $42,470 (0.8x ATR = $320)
• Outcome: Price reverted to midline in 35 minutes, exited at $42,980 for +$830 profit (+1.97%)

Monitoring

Key metrics to track:

• Bollinger Band width and slope
• ADX levels (trend strength)
• Orderbook imbalances at extremes
• Time in position
• Failed fade count
• Volatility regime changes
• Range integrity (support/resistance holding)

Risk Warnings

• Trend Risk: Ranges can break into trends, causing losses
• Acceleration Risk: Multiple failed fades indicate trend emergence
• Event Risk: News can violently break ranges
• Overtrading: Too many fades can accumulate losses
• Liquidity Risk: Thin orderbooks at extremes can cause slippage
• Correlation: Multiple range trades can correlate during market-wide breakouts

Backtest Outline

# Pseudocode for backtesting
bb_upper, bb_mid, bb_lower = calculate_bollinger_bands(length=20, dev=2.0)
bb_slope = calculate_slope(bb_upper, bb_lower)
adx = calculate_adx(period=14)

if price <= bb_lower and bb_slope <= 0 and adx < 25 and orderbook_exhaustion():
    if failed_fades < 3:
        place_maker_grid_long(levels=[bb_lower, bb_lower + 0.1%, bb_lower + 0.2%])
        set_stop_loss(bb_lower - 1.2 * atr)
        set_target(bb_mid or entry + 0.8 * atr)
        set_time_stop(90_minutes)

elif price >= bb_upper and bb_slope <= 0 and adx < 25 and orderbook_exhaustion():
    if failed_fades < 3:
        place_maker_grid_short(levels=[bb_upper, bb_upper - 0.1%, bb_upper - 0.2%])
        set_stop_loss(bb_upper + 1.2 * atr)
        set_target(bb_mid or entry - 0.8 * atr)
        set_time_stop(90_minutes)

# Exit management
if price_reaches_target():
    exit_position()
elif adx > 25:
    exit_position()  # Trend emerging
elif bb_slope > 0:
    exit_position()  # Volatility expanding
elif time_stop_reached():
    exit_position()
elif stop_loss_hit():
    exit_position()
    increment_failed_fades()

Why This Suits Small Capital + LLM Latency

• Maker-led approach: Post-only orders reduce costs
• Decisions at bar closes: Not microsecond-sensitive
• Low frequency: Only trades at band extremes
• Clear signals: Band touches are unambiguous
• Time-boxed: Quick exits prevent capital lock
• Grid entries: Can place multiple orders and wait for fills