On Wednesday, the Hindenburg Omen, named after the Hindenburg tragedy in 1937 where the hydrogen-filled dirigible exploded in a ball of fire.
Essentially, the Omen is a signal which purports to foretell the likelihood of impending market turmoil.
Following this signal's trigger on Wednesday, the S&P 500 declined on Thursday by a relatively modest 1.16 points to finish Thursday at 1540.08. The SPX remains well above its 200 and 100 day averages; however, let us put this turmoil indicator to the historical test.
This is our Market Information Machine (XMIM) analysis of the literal interpretation of the Hindenburg Omen. The criteria that need to be met are the following:
The daily number of NYSE new 52 week highs and the daily number of new 52-week lows are both more than 2.2% of the total number of equities traded on the NYSE.
There are at least 79 new 52-week highs and lows.
The average value of the NYSE Composite index (NYA) over the last 10 weeks moving average is higher than it was last week.
The McClellan Oscillator is negative on the day.
The number of stocks making new 52-week highs can not be more than twice the number of stocks making new 52-week lows.
In summary, the Hindenburg Omen is an indicator of increased market divergence which is not typical of a slow and steady uptrend market. With this increased divergence in the market, let us turn to our analysis.
It is also important to note that this is an unconfirmed Omen signal because it is the first occurrence of the event in the last 36 days. In order to be a confirmed Omen, this same event needs to occur in the next 36 days.
Despite this, the historical analysis of this event is consistent, such that each of the event dates given below are also unconfirmed signals.
Q: How has the SPX performed in the past when it does not lose more than -1% one day after the Hindenburg Omen is triggered?
A: According to the 12 previous occurrences of this event, omitting repeat occurrences within 10 trading days, SPX has shown a somewhat bearish edge that peaks 36 trading days after the event. Thus, the projected date for the peak of the bearish edge relative to the most recent occurrence of the event (Thursday Oct. 18, 2007) is Monday Dec. 10, 2007.
SPX declines in 83% of the cases (10 of 12) by an average of -4.0% relative to the close on the event date. The average of the two rallies is 1.3%. The overall return of the 12 cases is -3.1%, which, based on the close of SPX on the event date (1540.08), provides a target price of 1492.34.
Despite the fact that at the peak, SPX is only lower in 83% of the cases, there has been no prior occurrence of this event in which SPX did not move lower relative than the close on the event date over the duration of the historically rooted edge. This can be easily viewed by using our new Eventedge charts as at some point, each of the green lines (representing a bullish move at the peak) move negative.
It is also interesting to note that the last time this event occurred was mere days before the market high experienced in July. In that instance, SPX had lost -4.8% at the 36 day peak.
If you would like to see more details of this historical edge, go to www.markethistory.com
Ryan Soudan and Mickey Schoenhals are analysts with Markethistory.com.
