Finding Price Extremes with a Reversal-Threshold Algorithm
Summary
The article develops an approach for detecting swing highs and lows using a chosen price-variation threshold. It contrasts this with fractals, fixed-range methods, and moving-average-based searches, arguing that those alternatives can produce insignificant turns in flat markets or miss closely spaced extremes during volatile moves. The proposed method treats price as alternating between peaks and bottoms, uses the required reversal distance to define significant turns, and searches backward from the latest data so recent movements have greater influence on the result.
An additional fraction of the main threshold is introduced to identify the first, most recent extreme with less delay. The article applies detected extremes to chart-pattern and trend logic and reports example EA optimization outcomes, including trade counts and profitable-trade percentages for selected parameter settings. Those figures illustrate sensitivity to the variation and strategy parameters; they are not enough to establish general profitability. Results depend on parameter selection, and the excerpt does not establish out-of-sample robustness across markets or periods.
Key ideas
- A price-variation threshold filters out turns smaller than a chosen significance level.
- Alternating peak and bottom identification helps separate nearby extremes.
- Searching backward from the latest data makes detected extremes more responsive to recent price action.
- A fractional threshold for the first extreme is intended to reduce detection delay.
- Strategy results vary with the extreme-detection and trading parameters, so the reported examples do not prove general robustness.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.