As of 11/20/2024   Indus: 43,408 +139.53 +0.3%     Trans: 17,002 -26.31 -0.2%     Utils: 1,055 +1.25 +0.1%     Nasdaq: 18,966 -21.33 -0.1%     S&P 500: 5,917 +0.13 +0.0%	YTD  +15.2%    +6.9%    +19.7%    +26.3%    +24.1%	Targets    Overview: 11/12/2024     46,000 or 43,000 by 12/01/2024   18,000 or 16,600 by 12/01/2024   1,075 or 1,000 by 12/01/2024   20,000 or 18,400 by 12/01/2024   6,100 or 5,800 by 12/01/2024
Chart Pattern Indicator: on 11/18/24

As of 11/20/2024   Indus: 43,408 +139.53 +0.3%     Trans: 17,002 -26.31 -0.2%     Utils: 1,055 +1.25 +0.1%     Nasdaq: 18,966 -21.33 -0.1%     S&P 500: 5,917 +0.13 +0.0%	YTD  +15.2%    +6.9%    +19.7%    +26.3%    +24.1%
Targets    Overview: 11/12/2024     46,000 or 43,000 by 12/01/2024   18,000 or 16,600 by 12/01/2024   1,075 or 1,000 by 12/01/2024   20,000 or 18,400 by 12/01/2024   6,100 or 5,800 by 12/01/2024
Chart Pattern Indicator: on 11/18/24

Bulkowski's Failure Rate Study

My book, Encyclopedia of Chart Patterns Second Edition, shown on the right, has failure rates for most of the 63 chart and event patterns discussed in the book.

If you click on the above link and then buy the book (or anything) while at Amazon.com, the referral will help support this site. Thanks.

$ $ $

Do chart patterns fail more often now than they used to? Yes, according to this study.

Failure Rate Study: Summary

A study using nearly 14,000 chart patterns covering the years from 1991 to 2008 shows that chart patterns fail between two and four times more often now than they did in the past. For example, the average failure rate of chart patterns to climb at least 10% in the 1990s bull market was 14%. During the bull market years of 2003 to 2007, the failure rate had doubled to 28%.

Downward breakouts showed a similar trend with the average 10% failure rate rising from 26% in the 1990s to 49% during the bull market years of the 2000s.

Here is a more extreme example. In 1991 the 10% failure rate was 11% but peaked at 44% in 2007. In other words, four times as many chart patterns failed to show post breakout rises of at least 10% in 2007 compared to 1991.

Failure Rate Study: Methodology

I started with 13,932 chart patterns known to be good. That means I inspected them multiple times and have used them as research material for my books. For gains, I looked at all chart patterns with upward breakouts and measured the rise from the closing price the day before the breakout to the ultimate high.. For losses, I used chart patterns with downward breakouts and measured the decline from the close the day before the breakout to the ultimate low..

I did not use all of the chart pattern types at my disposal because some worked best on the weekly scale (horns and pipes) and discarded others because the breakout price is often difficult to determine (scallops, rounding turns, and broadening patterns, for example). Here is a list of the 23 pattern types that I used in the study.

• Diamond tops and bottoms
• Double tops and bottoms (all combinations of Adam and Eve)
• Head-and-shoulder tops and bottoms (both simple and complex)
• Rectangle tops and bottoms
• Rising and falling wedges
• Triangles: ascending, descending, and symmetrical
• Triple tops and bottoms

For more information on any of these chart patterns, click the link to take you to the index.

More

Failure Rate Results: Upward Breakouts

The above chart shows three failure rates, 10%, 20%, and 40%, arbitrarily chosen. The failure rates are a count of the number of times chart patterns fail to show post breakout rises of 10%, 20%, and 40% as a percentage of the number of chart patterns for that year.

For example, in 1991, eleven percent of the 140 chart patterns with upward breakouts failed to show a post breakout rise of at least 10%. In 1992, thirteen percent of the 494 chart patterns failed that year to post gains of at least 10%.

The chart shows the upward trend of failures over time. In 1991, the 10% failure rate stood at 11%. By 2007, the failure rate peaked at 44% before dropping to 32% in 2008. The 20% failure rate climbed from 22% in 1991 to 64% in 2007 before declining slightly to 62% in 2008. Finally, the 40% failure rate increased from 64% to 88% over the 1991 to 2008 period.

If you compare the bull market years in the 1990s to the 2003 to 2007 bull market, we find the average 10% failure rate climbed from 14% to 28%. In other words, 14% of chart patterns failed to show post breakout rises of at least 10% in the 1990s but that had doubled to 28% in the 2000s.

Looking at the relative height of each line, we see how often chart patterns show post breakout rises of 10%, 20% and 40% or higher. In 2008, 88% of chart patterns will fail to show post breakout rises of at least 40%. In other words, just 12% of chart patterns showed large gains in the bear market start of 2008. This compares to the 60% rate in 2000, the year that bear market began.

More

Failure Rate Results: Downward Breakouts

Downward breakouts show a similar trend as the above chart shows. The average 10% failure rate in the bullish 1990s was 26%. This had climbed to 49% during the bull market years of 2003 to 2007. The 20% failure rate climbed from 59% to 75%, and the 40% failure rate inched from 91% to 96%.

Notice that the failure rates from downward breakouts tend to drop during bear markets, just as one would expect.

The change from bear market to bull market in 2002 to 2003 showed the 10% failure rate spike from 15% to 48%. The 20% failure rate double from 42% to 83%. Clearly, you do not want to be short when the bear market ends.

-- Thomas Bulkowski.

More

See Also

• Compare the performance of chart patterns over the decades.
• Failures. Which months show the most failures?
• Shark-32 performance. The shark-32 shows trends end 30% quicker than in the 1990s.
• Pattern rank. Compare chart pattern performance.
• Best buy days. Which day of the week is the best one to buy or sell?
• Best buy months. Can buying at the end of the worst performing month and selling at the best performing be profitable?
• Seasonality. What are the best months to buy and sell stocks?
• Holidays. Does the market rise or fall before and after holidays?
• The best performing chart patterns over 1, 2, and 3 months.
• Future results? Look at the price predictions for the coming years.

24 hours in a day. 24 beers in a case. Coincidence?