FearGreedStrategy

"""
FearGreedStrategy — pure F&G contrarian play, no price-action overlays.

Thesis: Crypto fear/greed sentiment is a leading indicator. Buy when the herd
is in extreme fear (F&G <= 25), exit when the herd is in extreme greed
(F&G >= 75). No EMA, no RSI, no ADX — sentiment is the entire signal.

Uses alternative.me historical F&G data saved at the configurable local data
path below (user_data/data/fng_history.feather). Daily values are merged onto
each pair's 1h candles by UTC date — no lookahead because each day's F&G is
published shortly after midnight UTC.

Drafted as the kill-gate test for the "sentiment / leading indicator" question
raised after v5+v4+grid all failed.
"""
from pathlib import Path

import pandas as pd
from pandas import DataFrame
from freqtrade.strategy import IStrategy

# Configurable local data path (relative to repo root). Point this at your own
# alternative.me F&G feather dump (see scripts/fng_profile.py).
FNG_PATH = Path("user_data/data/fng_history.feather")
ENTER_FNG_MAX = 25
EXIT_FNG_MIN = 75


_FNG_CACHE: DataFrame | None = None


def _load_fng() -> DataFrame:
    global _FNG_CACHE
    if _FNG_CACHE is None:
        df = pd.read_feather(FNG_PATH)
        df["date_only"] = df["date"].dt.normalize()
        _FNG_CACHE = df[["date_only", "value"]].rename(columns={"value": "fng"})
    return _FNG_CACHE


class FearGreedStrategy(IStrategy):
    INTERFACE_VERSION = 3
    timeframe = "1h"

    minimal_roi = {"0": 100}
    stoploss = -0.99  # effectively disabled — let F&G greed exit handle all closes

    trailing_stop = False
    process_only_new_candles = True
    use_exit_signal = True
    exit_profit_only = False
    ignore_roi_if_entry_signal = False

    startup_candle_count = 1

    def populate_indicators(self, dataframe: DataFrame, metadata: dict) -> DataFrame:
        fng = _load_fng()
        dataframe["date_only"] = dataframe["date"].dt.normalize()
        merged = dataframe.merge(fng, on="date_only", how="left")
        merged["fng"] = merged["fng"].ffill()
        merged = merged.drop(columns=["date_only"])
        return merged

    def populate_entry_trend(self, dataframe: DataFrame, metadata: dict) -> DataFrame:
        dataframe.loc[dataframe["fng"] <= ENTER_FNG_MAX, "enter_long"] = 1
        return dataframe

    def populate_exit_trend(self, dataframe: DataFrame, metadata: dict) -> DataFrame:
        dataframe.loc[dataframe["fng"] >= EXIT_FNG_MIN, "exit_long"] = 1
        return dataframe