Bollinger Reverter 4h Agent Skill

Bollinger Reverter 4h

Symmetric mean-reversion strategy on the 4h timeframe. Long-or-short on band touches, gated to range regimes via ADX. Validated across BTC/ETH/SOL/DOGE over 162 days.

Searchable under: mean reversion, bollinger band, range trader, chop strategy, ADX filter.

Backtest evidence

| Config | Trades | Win rate | Profit | Max DD |

|---|---|---|---|---|

| BTC/USDC:USDC, 162d | 18 | 72.2% | +8.14% | 10% |

| BTC/USDC:USDC, second-half / chop (80d) | 8 | 100% | +9.88% | 0% |

| BTC/USDC:USDC, first-half / strong bear (82d) | 10 | 50% | -1.75% | 10% |

| Multi-pair (BTC/ETH/SOL/DOGE), 162d | 84 | 65.5% | +8.77% | 18.5% |

Per-pair breakdown (multi-pair 162d):

| Pair | Trades | Win | Profit |

|---|---|---|---|

| BTC/USDC:USDC | 29 | 72% | +3.76% |

| ETH/USDC:USDC | 19 | 74% | +4.39% |

| SOL/USDC:USDC | 15 | 60% | +1.27% |

| DOGE/USDC:USDC | 21 | 52% | -0.65% |

3 of 4 majors profitable, DOGE marginally negative. Generalizes well; not BTC-specific.

Thesis

When the market is range-bound (ADX < 25), price touching the upper or lower Bollinger Band is statistically likely to revert to the midline. Tight ROI ladder takes profit fast since mean-reversion targets are small; tight stop prevents the position from holding if the band touch turns into a trend break.

Mechanics

• Pair: validated on majors; extend to any pair with sustained 24h volume > $50M
• Timeframe: 4h
• Indicators: 20-bar Bollinger Bands (2σ), RSI(14), ADX(14)
• Entry short: close > upper_band AND RSI > 65 AND ADX < 25
• Entry long: close < lower_band AND RSI < 35 AND ADX < 25
• Exit short: close < bb_mid
• Exit long: close > bb_mid
• Stops: -2% hard stop
• ROI ladder: 2.5% immediate, 1.5% after 4h, 0.5% after 12h, breakeven after 24h
• No trailing stop (mean reversion targets are short — let ROI or signal-exit fire)

Full strategy code

from freqtrade.strategy import IStrategy
import pandas as pd
import talib.abstract as ta


class BollingerReverter4hStrategy(IStrategy):
    INTERFACE_VERSION = 3
    timeframe = "4h"
    can_short = True

    stoploss = -0.02
    trailing_stop = False

    minimal_roi = {
        "0": 0.025,    # take 2.5% immediately
        "240": 0.015,  # 1.5% after 4 hours (1 bar)
        "720": 0.005,  # 0.5% after 12 hours (3 bars)
        "1440": 0,     # breakeven after 24 hours
    }

    process_only_new_candles = True
    startup_candle_count = 60
    use_exit_signal = True

    def populate_indicators(self, dataframe: pd.DataFrame, metadata: dict) -> pd.DataFrame:
        bb = ta.BBANDS(dataframe, timeperiod=20, nbdevup=2.0, nbdevdn=2.0)
        dataframe["bb_upper"] = bb["upperband"]
        dataframe["bb_mid"] = bb["middleband"]
        dataframe["bb_lower"] = bb["lowerband"]

        dataframe["rsi"] = ta.RSI(dataframe, timeperiod=14)
        dataframe["adx"] = ta.ADX(dataframe, timeperiod=14)
        return dataframe

    def populate_entry_trend(self, dataframe: pd.DataFrame, metadata: dict) -> pd.DataFrame:
        cond_short = (
            (dataframe["close"] > dataframe["bb_upper"])
            & (dataframe["rsi"] > 65)
            & (dataframe["adx"] < 25)
        )
        dataframe.loc[cond_short, "enter_short"] = 1
        dataframe.loc[cond_short, "enter_tag"] = "bb_upper_revert"

        cond_long = (
            (dataframe["close"] < dataframe["bb_lower"])
            & (dataframe["rsi"] < 35)
            & (dataframe["adx"] < 25)
        )
        dataframe.loc[cond_long, "enter_long"] = 1
        dataframe.loc[cond_long, "enter_tag"] = "bb_lower_revert"
        return dataframe

    def populate_exit_trend(self, dataframe: pd.DataFrame, metadata: dict) -> pd.DataFrame:
        dataframe.loc[dataframe["close"] < dataframe["bb_mid"], "exit_short"] = 1
        dataframe.loc[dataframe["close"] > dataframe["bb_mid"], "exit_long"] = 1
        return dataframe

Reference config (multi-pair)

{
  "exchange": {
    "name": "hyperliquid",
    "pair_whitelist": ["BTC/USDC:USDC", "ETH/USDC:USDC", "SOL/USDC:USDC", "DOGE/USDC:USDC"]
  },
  "stake_currency": "USDC",
  "stake_amount": 75,
  "dry_run_wallet": {"USDC": 350},
  "timeframe": "4h",
  "max_open_trades": 4,
  "minimal_roi": {"0": 100.0},
  "stoploss": -0.02,
  "trading_mode": "futures",
  "margin_mode": "isolated",
  "entry_pricing": {"price_side": "same", "price_last_balance": 0.0},
  "exit_pricing": {"price_side": "same", "price_last_balance": 0.0},
  "pairlists": [{"method": "StaticPairList"}]
}

Strategy-level minimal_roi overrides config-level — the ROI ladder is what makes this work.

Honest framing

The 100% second-half BTC win rate is partly small sample (8 trades). The full-period multi-pair result (+8.77%, 84 trades, 65.5% win) is the more credible expectation. Range-bound regimes are when this prints; in strong trends it modestly loses (-1.75% on BTC during the first-half strong bear) because band touches keep continuing rather than reverting.

In any window with mixed regimes, the strategy should be net positive because the chop periods dominate by count.

The DOGE result (-0.65%) is the failure case — meme-coin volatility breaks more bands than reverts to them. Use this strategy on majors, not meme pairs.

Tunables

| Parameter | Range | Effect |

|---|---|---|

| BB period | 18 - 24 | Length of mean-reversion window |

| BB σ | 1.8 - 2.5 | Wider = rarer signals, deeper reversion |

| RSI confirmation | 60-70 / 30-40 | Confirms exhaustion at band edge |

| ADX cutoff | 20 - 30 | Below = range regime; above = trend (skip) |

| ROI tier 0 | 0.020 - 0.030 | Initial take-profit |

| Stop | -0.015 to -0.025 | Tight enough that one trend break doesn't erase the lifetime edge |

Known failure modes

• Regime transition: when chop turns into trend mid-trade, the band-touch-revert signal becomes a band-break-continuation. Stops should fire fast; this is what the -2% stop is for
• Meme/low-cap pairs: bands break more than they revert. Restrict to majors
• News spikes: a sudden 5%+ move blows through multiple bands; the stop will fire but execution slippage hurts. Consider pausing during scheduled macro events

Pairing

• Designed to coexist with donchian-strong-regime — they fire on mutually exclusive regimes (ADX < 25 here, regime-strong gate there)
• Supersedes the prior 1h variant of mean-reversion

Deployment recommendation

Run as its own sub-account with stake_amount sized so 4× max_open_trades fits within the wallet plus 1.5× buffer. Multi-pair allocation across BTC/ETH/SOL is the validated default.