PolymarketMeanReversionStrategy

"""Polymarket Mean-Reversion Strategy — fade overreactions in prediction markets.

Trades event contracts by betting against short-term overreactions:
- Buy when probability drops significantly below its rolling mean (oversold)
- Sell when probability reverts to mean or overshoots above it

Designed for contracts where sharp moves are driven by noise/overreaction
rather than fundamental shifts (e.g., speculative events, sentiment spikes).
"""

from datetime import datetime

import pandas as pd
from freqtrade.strategy import IStrategy

from alpha.PolymarketFactors import PolymarketAlpha


class PolymarketMeanReversionStrategy(IStrategy):
    INTERFACE_VERSION = 3

    can_short: bool = False
    minimal_roi = {"0": 0.15}  # Take profit at 15% gain
    stoploss = -0.30
    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: int = 30

    def populate_indicators(self, dataframe: pd.DataFrame, metadata: dict) -> pd.DataFrame:
        dataframe = PolymarketAlpha(dataframe, metadata).process()
        return dataframe

    def populate_entry_trend(self, dataframe: pd.DataFrame, metadata: dict) -> pd.DataFrame:
        dataframe.loc[
            (
                # Z-score strongly negative (price well below mean)
                (dataframe["prob_zscore"] < -1.5)
                # Mean reversion signal confirms (below rolling mean)
                & (dataframe["mean_reversion_signal"] < -0.03)
                # Volume surge suggests reactionary move, not fundamental
                & (dataframe["volume_surge"] > 1.5)
                # Contract still has room to move (not near resolution)
                & (dataframe["resolution_proximity"] > 0.10)
                # Price in tradeable range
                & (dataframe["close"] > 0.10)
                & (dataframe["close"] < 0.90)
            ),
            "enter_long",
        ] = 1

        return dataframe

    def populate_exit_trend(self, dataframe: pd.DataFrame, metadata: dict) -> pd.DataFrame:
        dataframe.loc[
            (
                # Price reverted above mean
                (dataframe["prob_zscore"] > 0.5)
                # OR momentum shifted positive (reversion complete)
                | (
                    (dataframe["mean_reversion_signal"] > 0.02)
                    & (dataframe["prob_momentum"] > 0)
                )
            ),
            "exit_long",
        ] = 1

        return dataframe

    def confirm_trade_entry(
        self,
        pair: str,
        order_type: str,
        amount: float,
        rate: float,
        time_in_force: str,
        current_time: datetime,
        entry_tag: str | None,
        side: str,
        **kwargs,
    ) -> bool:
        if rate < 0.05 or rate > 0.95:
            return False
        return True