Detecting Excessive Market Moves Across Multiple Lookback Lengths

Summary

The article proposes finding unusually large price moves across a range of lookback lengths, rather than relying on a single fixed period such as the conventional RSI or Bollinger Band settings. It estimates volatility from prior chart data, measures moves across candidate lengths, normalizes them by volatility, and selects the largest standardized move. A threshold flags moves that exceed an expected-volatility multiple; the signal is displayed only after a confirmation bar fails to extend the move’s extreme.

Examples span Bitcoin, German power, and yen foreign-exchange data, with different move thresholds and confirmation delays. A reported signal-efficiency test says the average profit factor after detected bullish reversals in daily yen data exceeded one, including during a broader falling market. That is limited evidence for a particular setup, not a general guarantee: the article provides no full performance figures here and says signals should be used as context within a strategy. It suggests different length ranges for short, medium, or long horizons and emphasizes stops, position sizing, and risk management.

Key ideas

• The method searches multiple lookback lengths to detect unusually large moves.
• Price moves are normalized by volatility, and the largest standardized move is compared with a threshold.
• A confirmation bar waits for the move to stop extending before the signal appears.
• The article reports a positive profit-factor observation for bullish yen signals, but not broad proof of profitability.
• The author recommends using signals with trend context, stops, and risk controls.

Tags

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