Polymarket Trader

What This Skill Does

The Polymarket Trader skill, developed by drakec48, is an advanced algorithmic framework designed for BTC 1h Up/Down markets on Polymarket. It functions by anchoring all trading decisions to the Binance BTCUSDT resolution source, ensuring that strategy execution remains tightly coupled with the underlying market reality. By modeling the relationship between fair probability and market price, the skill allows users to design, evaluate, and tune high-precision trading strategies. It provides a robust architecture for calculating volatility (sigma), time-to-expiry, and directional guardrails, effectively mitigating common risks like churn and poor entry timing. The skill encompasses a full lifecycle, from fetching klines via Binance to performing post-trade analysis using internal log dumps.

Installation

To integrate this skill into your OpenClaw environment, execute the following command in your terminal: clawhub install openclaw/skills/skills/drakec48/polymarket-trader Ensure your workspace has the necessary dependencies for running the bundled Python scripts located in the local directory.

Use Cases

This skill is ideal for quantitative traders who need to:

1. Develop and test mispricing models based on the discrepancy between fair probability and market-quoted prices.
2. Implement complex regime filters that distinguish between trending markets and ranging environments.
3. Conduct retrospective forensic analysis on unsuccessful trades by syncing event logs with historical Binance data.
4. Run rapid parameter sweeps to optimize threshold settings for entry and exit signals.

Example Prompts

1. "Analyze my last 5 trades in events.jsonl and tell me if the entry mode misaligned with the Binance trend slope computed by binance_regime.py."
2. "Run a parameter sweep to see how increasing my edge threshold to 0.05 affects total return for the BTC 1h Up market."
3. "Fetch the current 1m klines from Binance and calculate the fair probability for the next hourly settlement based on current sigma."

Tips & Limitations

Always check the references/strategy.md file before tuning, as it contains the critical math model parameters. The skill is highly specialized for 1h resolution markets; do not attempt to force it into non-hourly duration markets without modifying the source logic. Ensure your environment has stable network access to hit the Binance API. When diagnosing bad trades, always prioritize the explain_fills.py output to confirm if the exit logic was triggered by edge decay or a faulty mean-reversion target.