A Quantitative Method for Testing Candlestick Pattern Performance

Summary

The document proposes a framework for checking whether familiar candlestick patterns still predict subsequent price movement. For each pattern, it counts occurrences over a chosen historical interval and timeframe, then measures whether price moves a specified distance up or down within one, two, or three candles. Results are summarized with separate probability and efficiency measures; the latter weights faster moves more heavily, using default weights of 1, 0.5, and 0.25 for the three timing categories.

An illustrative Harami Cross example uses 2,000 hourly candles and a 50-point threshold. It reports 50 pattern matches, with upward and downward moves reaching the threshold 20 and 10 times respectively; the article says the upward probability is 40% and the downward probability is 20%. The proposed interface also allows users to adjust the sample, threshold, and weights. These are descriptive historical measurements, not evidence of profitability: results may vary by instrument, timeframe, and market conditions, and the weights are subjective.

Key ideas

• Evaluate a candlestick pattern by comparing its expected direction with price movement after each occurrence.
• Count how often price reaches a defined threshold within successive candles to capture both direction and speed.
• The method reports probability relative to all pattern matches and a weighted efficiency score for successful moves.
• The example favors upward movement after the tested pattern, but the result is specific to its sample and settings.
• The author cautions that pattern behavior changes across instruments and timeframes, so signals need rechecking.

Tags

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