Testing Bitcoin On-Chain Metrics as Trading Signals

Summary

This research article evaluates Bitcoin trading rules based on institutional and on-chain metrics, including NUPL, Puell Multiple, and MVRV-Z. It describes a common backtest setup with starting cash, trading fees, full-portfolio entries, and versions with and without a fixed stop-loss. The rules generally use low metric readings to enter and high readings to exit; the article compares results with buy-and-hold and moving-average crossover baselines using return and risk measures.

The case studies show that outcomes vary by metric and that stop-loss rules can reduce returns, especially when a strategy exits during a market decline before a recovery. The article reports strong historical results for some metrics, but the backtests cover one asset and a specific historical period, use simple all-in sizing, and do not establish live-trading performance. It also identifies tuning thresholds and adding position sizing as possible next steps, while noting the risk of improving historical fit at the expense of generalization.

Key ideas

• The article tests Bitcoin entry and exit rules based on on-chain metrics and compares them with basic trading baselines.
• Its backtests use trading fees, full-portfolio positions, and separate versions with and without a stop-loss.
• Some metrics produce strong historical results, but performance varies and stop-loss exits can reduce returns.
• The study uses simple sizing and historical data, so its reported results do not demonstrate live-trading effectiveness.
• Threshold tuning and position sizing are proposed as areas for further investigation.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.