Fractal High and Low Patterns for Directional Trading

Summary

This strategy identifies local highs and lows in candlestick data using a configurable period count. A peak high requires the candidate bar to exceed the highs of preceding bars and satisfy comparisons with following bars; a peak low applies the mirrored test to lows. The method then enters long on the detected high pattern and short on the detected low pattern, as described in the document. The listed period setting is 2, and published backtest settings specify BTC/USDT futures on hourly bars with 15-minute base data during January 2024.

Because each pattern checks bars that follow the candidate, it can only be confirmed after those bars have formed; plotting the marker back on the candidate bar does not make the signal available then. The text itself notes that pattern identification can be inaccurate and that the approach lacks explicit stop-loss logic or other contextual filters. It suggests adding stops, volume or volatility measures, and multiple timeframes. No backtest performance figures are supplied, and the source’s long/short mapping differs from the usual interpretation of swing highs and lows, so the signal logic merits scrutiny.

Key ideas

• The method marks local highs and lows by comparing a candidate candle with neighboring candles.
• The strategy description maps detected high patterns to long entries and low patterns to short entries.
• The listed lookaround period is 2, and confirmation depends on later candles.
• The document recommends adding stop logic and considering volume, volatility, or other timeframes.
• No performance figures are given, and the source’s entry mapping should be checked before use.

Tags

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