The Hindenburg Omen Triggered; Will there be a US Stock Market Crash?

Although I began my analytical work on the financial markets as a “chartist”, I eventually moved on. 

There has been recent buzz about the reemergence of the so-called  “Hindenburg Omen”, which supposedly scared US stock markets last week.

The Hindenburg Omen, a technical indicator, if triggered supposedly portends of a stock market crash, thus was named after the “Hindenburg disaster” (see image above)

Writing at the stockcharts.com Chip Anderson says that the conditions of these crash indicator has been triggered: 

It happened in mid-April and it happend again on the last day of May.  The ominous sounding "Hindenburg Omen" signal has been given.  Here's the chart:

Here's the definition from our ChartSchool Glossary page:

"Hindenburg Omen: Created by James Miekka, the Hindenburg Omen warns of potential weakness in the stock market. There are three criteria to activate the omen. First, NYSE new highs and new lows must both be more than 2.8% of advances plus declines. Second, the NY Composite is above the level it was 50 days ago. Third, the number of new highs cannot be more than double the number of new lows. The activation period is good for 30 days. Once active, a sell signal is triggered when the McClellan Oscillator moves below zero and negated when the McClellan Oscillator moves back above zero."

So Friday's big drop triggered the Omen signal by causing $NYLOW:$NYTOT (the ratio of NYSE Lows to NYSE Total Stocks) to spike up above 2.8% (the red area graph above).

The Wikipedia.org has further conditions for such pattern to take place:

The traditional definition requires each condition to occur on the same day. Once the signal has occurred, it is valid for 30 days, and any additional signals given during the 30-day period should be ignored. During the 30 days, the signal is activated whenever the McClellan Oscillator is negative, but deactivated whenever it is positive.

Some users of the omen may choose to view the 30 day limit as "working days" and not "calendar days". This is reasonable as the global finance market works on a weekday (Monday to Friday) schedule—leaving about 100 hours where only limited sharemarket trading takes place. This only extends the omen's warning by an extra 10 days, a reasonable limit.

Having met the conditions, will the US stock market crash within 30 days?

Finally the odds for the pattern’s success:

From historical data, the probability of a move greater than 5% to the downside after a confirmed Hindenburg Omen was 77% [The Wall Street Journal 8/23/2010 article cited below states that accuracy is 25%, looking at period from 1985], and usually takes place within the next forty days. The probability of a panic sellout was 41% and the probability of a major stock market crash was 24%. Though the Omen does not have a 100% success rate, every NYSE crash since 1985 has been preceded by a Hindenburg Omen. Of the previous 25 confirmed signals only two (8%) have failed to predict at least mild (2.0% to 4.9%) declines.

Because of the specific and seemingly random nature of the Hindenburg Omen criteria, the phenomenon may be simply a case of overfitting. That is, by backtesting through a large data set with many different variables, correlations can be found that do not really have predictive significance. The Omen is at best an imperfect technical indicator that is a work in progress.

The last paragraph suggests that the accuracy of the Hindenburg Omen as crash forecasting tool may have been about data fitting. In short, this may not be reliable.

Anyway if the US stock markets should crash, I think it would more about the risks of a precipitate surge in the bond yields.

The Hindenburg Omen may function as a coincident indicator which reveals of the transition of the market’s sentiment as expressed in price trends and interpreted via specific technical indicators. 

But I wouldn't bet on a crash based merely on the Hindenburg Omen.