Home
About
Blog
Blog Archives
Book Reviews
My Books
Favorite Books
Contact
FAQ
Glossary
Links
Privacy/Disclaimer
Search
Site Map
Visual Index
What’s New

Clicking on my books below takes you to Amazon.com. At no cost to you, they pay for the referral and that helps support this site.

Bulkowski’s Time vs Pattern Width

ThePatternSite.com logo

Candles Chart Patterns Elliott Wave Event Patterns Indicator Market Review Pattern Rank Quizzes Patternz Software Research Scoring Patterns Trading Help Trading Setups Volume Patterns

Written and copyright © 2008 by Thomas N. Bulkowski. All rights reserved.

This page examines the relationship between time to the ultimate high or low and pattern width.

Summary, Methodology, and Results

I used 40 different chart patterns (such as a falling wedge, double bottom, and so on) and over a thousand stocks, starting from July 1991 to June 2008. I logged the pattern width and the time it took to reach the ultimate high or low. The ultimate high or low is the highest high or lowest low, respectively, before price changes trend by at least 20%. As an example for the ultimate high, that means a decline of at least 20%. I uncovered 23,342 chart patterns that qualified. I used them all and wish I hadn’t. The spreadsheet was about 25 megabytes in size and when you changed one variable, it took between 20 and 40 minutes to recalculate it.

Downward Breakouts

A frequency distribution of the ratio between the time to the ultimate low vs pattern width

The chart shows a frequency distribution of the ratio of the time from the end of the pattern to the ultimate low divided by the pattern width. Values above 1 mean it takes longer to reach the ultimate low than the pattern is wide. I show 1.00 circled in red. The bar is off the chart because 55% of the patterns take less than the pattern’s width to reach the ultimate low.

The two red bars include the Fibonacci values of 38% and 62% (the first column includes all ratios up to 1. The next column includes the ratios from over 1 up to and including 1.05. Thus the 38% and 62% values will appear in the 1.40 and 1.65 columns). The right-most column is a catch-all because it includes all ratios over 2.40.

I do not see anything significant here at all. I had hoped to see peaks at the Fibonacci extension levels, but they are not there.

Top

Upward Breakouts

A frequency distribution of the ratio between the time to the ultimate high vs pattern width

What about upward breakouts? A similar chart appears and I show that directly above. I used the same method as before only I compared the time to the ultimate high versus the pattern width. The red bars include the Fibonacci values of 38% and 62%, just as before. This time, however, fewer patterns flame out before exceeding the pattern width. Just 39% fall into the first category (every ratio up to 1.00). That means 61% of the patterns take longer to reach the ultimate high than the pattern’s width.

Top

Copyright © 2008 by Thomas N. Bulkowski. All rights reserved.