The return of the Hindenburg Omen signal
While everyone was jumping up and down on Alibaba the internals of the stock market took another hit. Another distribution day was added to the Nasdaq making a total now of a cluster of 5. I exited at a clustering of 4 distribution days even though the moving averages and % increases & drops have still held.
Fridays action of a reversal off new highs is a very negative sign. All the major averages closed near the bottom of their trading ranges and volume was much higher across the board. Overall the market is looking weak. Few stocks are breaking out and producing meaningful gains and the major averages seem to be struggling. We are also seeing the return of the Hindenburg Omen signal. We had one yesterday and just missed one earlier this week. This indicator occurs near the end of cyclical bull markets. We had a bunch of them in 2007 near the end of the last cyclical bull and several more occurred earlier this year. We are now seeing them again. The Hindenburg Omen is not perfect as nothing is. The Hindenburg Omen is based on based on Norman G. Fosback’s High Low Logic Index (HLLI).The value of the HLLI is the lesser of the NYSE new highs or new lows divided by the number of NYSE issues traded, smoothed by an appropriate exponential moving average.
From Wikipedia’s page on the Hindenburg Omen…
These criteria are calculated daily using Wall Street Journal figures from the New York Stock Exchange for consistency. (Other news sources and exchanges may be used as well.) Some have been recalibrated by Jim Miekka to reduce statistical noise and make the indicator a more reliable predictor of a future decline.
The daily number of NYSE new 52 week highs and the daily number of new 52 week lows are both greater than or equal to 2.8 percent (this is typically about 84 stocks) of the sum of NYSE issues that advance or decline that day (typically, around 3000). An older version of the indicator used a threshold of 2.5 percent of total issues traded (approximately 80 of 3200 in today’s market).
The NYSE index is greater in value than it was 50 trading days ago. Originally, this was expressed as a rising 10 week moving average, but the new rule is more relevant to the daily data used to look at new highs and lows.
The McClellan Oscillator is negative on the same day.
The number of New 52 week highs cannot be more than twice the number of new 52 week lows (though new 52 week lows may be more than double new highs).
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, or one signal does not mean very much, but more than one is a confirmed signal, or five or six are even more important. 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”, arguing that the global finance market works on a weekday (Monday to Friday) schedule—leaving about 100 hours where only limited sharemarket trading takes place.
As a rule, the shorter the time-frame in which the conditions listed above occur, and the greater the number of conditions observed in that time frame, the stronger the Hindenburg Omen. If several—but not all—of the conditions are repeatedly observed within a few weeks, that is a stronger indicator than all of the conditions observed just once during a 30-day period.
Structural: New highs and lows are being affected by exchange-traded funds (ETFs). The last two times Hindenburg triggered was due to Bond ETFs making new highs or lows. If ETFs were removed, Hindenburg would not have triggered. When the Omen was originally designed there were no ETFs, so triggering behavior in the 2010s is not the same as in the 1990s to mid-2000s.
Theoretical: It is theoretically possible for those with unlimited financial resources and minimally regulated automated trading systems to keep the omen from triggering. This has been postulated by the creator of the “Vergulde Draeck” Omen.
Triggering: To eliminate false positives some technical analysts have imposed the condition that the Hindenburg Omen
must be triggered three times in a row within a month from the first triggering event for said initial trigger signal to be considered to be valid (i.e. requires double confirmation)
is only valid when “all tightly coupled triggerings are within a fortnight”
will indicate a possible future downturn or correction, depending on the magnitude of any “one off” triggering
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.
In 2013, a Hindenburg Omen signal was initially observed on April 5. Ten days later on April 15th, a second Omen nearly materialized, but narrowly failed to do so as the NYSE 52-week lows on that day were 2.749%, just shy of the necessary 2.8% or greater. As such, the April 5th Omen failed to be confirmed by a repeat Omen within a period of 30 days, which is generally considered a requirement for validation. On May 31, the Hindenburg Omen again appeared; this time with proper validation in the form of three subsequent Omens occurring on June 4, June 10 and June 19. The Omen re-emerged within the first two weeks of August in a series of six occurrences in three back-to-back sets of Omens. The first group of Omens occurred for two consecutive days on August 5 and 6, the second pair on August 8 and 9 and the most recent on August 13 and 14.
Sep 19, 2014: Weekly 10SMA rising, 128 New Highs, 102 New Lows, McClellan -89.797
Dec 11, 2013 and again on Dec 16, 2013 where the OSC stayed negative in between those dates and is still negative as of Dec 17, 2013 (the 50 day prior higher high seems to be valid but can someone confirm these two occurrences.)
July 23, 2012. The Omen was triggered in 2012 on July 23 and was immediately confirmed by a subsequent Omen the next day on July 24 and by a third on July 25th forming a consecutive three-day cluster.
August 12, 2010: The Omen’s creator, Jim Miekka, considered the Omen officially triggered on this date with 92 and 81 new 52-week highs and lows, respectively. The McClellan Oscillator was a negative -120.03 and the 10-week NYSE moving average was rising; the market closed above its open of 50 days prior (May 27). In the ensuing week, the Omen narrowly missed confirmation twice (August 13 and 19).
August 20, 2010: According to the Wall Street Journal, the omen was confirmed on Friday, with 83 new 52-week highs and 95 new 52-week lows on the NYSE. The McClellan Oscillator was a negative -106.46 and the 10-week NYSE moving average was rising; the market closed above its open of 50 days prior (June 11).
August 24, 2010: 166 New Lows, 87 new Highs, McClellan Oscillator was negative, but the 10 week average began to fall. (Non-Confirmation.) (Although the 12 week average is still positive.)
August 25, 2010: 150 New Lows, 90 new Highs, McClellan Oscillator was negative, but again the 10 week average was falling (Non-Confirmation.) (Although the 12 week average is still positive.)
August 31, 2010: 86 New Lows, 164 new Highs, McClellan Oscillator was negative, and the 10 week moving average was up slightly 8.86 (0.13%) but falling (non-confirmation)
December 14, 2010: 113 New Lows, 179 New Highs, 3063 Advancers+Decliners, McClellan Oscillator was negative (-5.36), NYSE Composite Index closed at 7855.22 vs 7272.53 50 trading days prior (October 4, 2010), and the 10 week moving average was rising.
December 15, 2010: 89 New Lows, 156 New Highs, 3044 Advancers+Decliners, McClellan Oscillator was negative (-22.59), NYSE Composite Index closed at 7798.78 vs 7434.18 50 trading days prior (October 5, 2010), and the 10 week moving average was rising. This represents a single confirmation.
We are in a 5 year old bull market with optimism flowing. All too many forget how bad 2008 was. I am not a bull nor a bear. I am a realist and try to read the market as best as I can. The Fed’s QE forever has supported the market for a long time and it is likely the bull would have ended earlier without it. I don’t know when it will end, but when it does it is likely to be ugly.