REPEAT Rudi’s View: The Hindenburg Omen, Scary But Unconfirmed

This story features COMMONWEALTH BANK OF AUSTRALIA. For more info SHARE ANALYSIS: CBA

This story was originally published on Wednesday, 18th August, 2010. It has now been re-published to make it available to non-paying members at FNArena and readers elsewhere.

By Rudi Filapek-Vandyck, Editor FNArena

Ultimately, it is the dream of all market analysts: discovering that one indicator that always applies and never fails. Alas, daily reality is a bit different; a lot different, really. At best certain indicators work most of the times and none of them applies everywhere or regardless what the circumstances.

A good example is my personal market indicator, one I developed throughout years of reading stockbroker research reports and observing market movements. I cannot remember when exactly, but at one point I discovered Australian banks could function as the proverbial canary in the coalmine.

I discovered that when investors would become too exuberant, they would push share prices for the Big Four Banks in Australia past consensus price targets and every time the share market would pull back soon after. My interest was piqued and I researched the matter further.

Turns out every share market correction in Australia over the past ten years has been preceded by one such “banks signal”. I have written extensively and repeatedly about this indicator over the years past. If you happen to be a relatively new reader of my market analyses, and still unaware of my personal indicator, I apologise for not spending more time on the subject this time.

I will do so, I promise, when the time has arrived to again start paying attention to my personal market indicator, but with bank shares (except CommBank ((CBA)) still double digits away from their respective consensus price targets, this indicator is not going to be of much use anytime soon.

My personal indicator had a 100% hit rate until share markets started rallying again in March 2009. It wasn't long before bank share prices were again at their respective price targets and I predicted it was time for a breather. This appeared correct at first, but after the pause share markets, banks included, continued their rally.

My indicator had failed. Or had it?

Turns out I had been too stupid to pay attention to the broader context. What my banks-price-targets indicator requires is that securities analysts are up to date with their estimates and their forecasts, at least more or less. Back in 2009 they'd all been sitting behind their desks, frozen by fear, waiting whether the market would sell down their much beloved stocks even further.

Hardly an environment to be able to rely on their valuations and forecasts, and as it turned out, it wasn't.

That was the only time my personal indicator failed to predict a market correction. In January this year, and again in April, it worked like a well-trained canary should; right on time and before share prices started to drop.

Oh, and did I mention late 2007?

The problem with my indicator is, alas, that share market corrections can also happen when share prices are not overvalued. But under such circumstances my indicator is simply no good.

Similarly, during the Great Bear Market Sell-Offs of 2008 I started writing about the Coppock Indicator. My stories at the time generated a lot of interest, even though the indicator only proved its true value in 2009, signaling it was okay to get back into the share market.

Certainly, to those who paid attention at the time, Coppock's Indicator once again proved its value. However, when some readers enquired about this indicator earlier this year I told them they were looking in the wrong direction. Coppock's Indicator only applies when equities come out of the Bear Market Sell-Off Phase, not when they went back up and pull back by 10% or so.

Rule number one: you have to know when a particular indicator applies.

Enter the Hindenburg Omen. This indicator was originally developed to predict share market disasters. One could say, it functions as the exact opposite purpose of Coppock's. Sure, we all would like to know when the next potential sell-off will occur, but as said in the opening sentences of today's View, there is no such thing as an always working, always accurate market indicator.

Similar to my personal market indicator though, and to the Coppock Indicator, the Hindenburg Omen comes with an admirable track record. It has been said not one single share market sell-off has occurred in the US since 1985 without first having generated a warning signal from the Hindenburg Omen.

That's pretty impressive.

Alas, the track record starting from every warning signal onwards is far less impressive. To go straight to the core: not every warning since 1985 has been followed up with a sell-off.

And so it is that in August 2010, at one of many market inflection points the calendar year has brought upon us, the Hindenburg Omen has issued its first warning signal. Should we get worried?

No, say those market technicians with a bullish market view. The Hindenburg Omen is as yet unconfirmed (true) and its track record is far from perfect.

Yes, say many others, pointing at the reverse track record (no sell-off without a warning since 1985) while pointing out this month's warning comes on top of a plethora of other equally negative market indicators.

Apparently, since Friday, when the Hindenburg Omen issued its warning in the US equities market, financial blogs and chat rooms across the world have been virtually overheating as bears and bulls discuss the merits and flaws of this technical indicator.

Before we look with more detail into what this particular indicator actually measures, let's have some stats to admire:

Note that Friday's Hindenburg Omen warning is as yet an “unconfirmed” signal (will explain a bit further). But were this signal to be “confirmed”, historical data suggest we have a 77% probability of witnessing a 5% correction within the next 41 days.

However, 5% is not necessarily what the bears are looking for. Still assuming we get “confirmation”, the probability of a panic sell off is 41%, and that of a genuine stock market crash is 25%.

Now is probably a good time to point out that we are only a few weeks away from the September-October period which historically tends to generate most share market “accidents”, for whatever reason.

The above statistics, and suggested time lines would, believe it or not, fit in perfectly with another dreadful September-October experience. Which begs the obvious question: if share markets do run into a “correction” over the next 70 days or so, is that than evidence of another October-effect, or is it because of the Hindenburg Omen?

As far as the indicator itself goes, the Hindenburg Omen is, similar to the Coppock Indicator I guess, not as simple and straightforward as my personal market indicator, or as many other indicators such as the “Cross of Death” which equally received a lot of attention these past months.

What the Hindenburg tries to do is measure so-called “distribution” in the share market, when overall confusion leads to a deterioration in market breadth; this then precedes the next move down. Similar to the previous two indicators I mentioned today, the proponents of this technical indicator have a solid conviction in its accuracy.

So how does it work?

The Hindenburg Omen is in essence a combination of various major signals and technical patterns and each of them has to be fulfilled on the same day to trigger a valid signal. (So one can understand why it is not so easy to generate a valid signal at all).

The criteria of the pattern are as follows:

1) Daily number of NYSE new 52-week highs and the daily number of new 52-week lows must both be greater than 2.2 % of total NYSE issues traded that day
2) The smaller of these numbers is greater than or equal to 69
3) The NYSE 10-week moving average is rising
4) The NYSE McClellan Oscillator is negative
5) New 52-week highs are not more than twice the new 52-week lows, which is an absolutely mandatory condition.

Note that even if all of these five criteria have been fulfilled at the close of a given day, we still only have an “unconfirmed” Hindenburg Omen. Which is the situation we're in at right now.

However, if the same thing happens again within 36 days this becomes a “confirmed” Hindenburg Omen. Mid-September is thus the ultimate cut-off deadline.

Those who want to read more, here are two easy links to further information:

At Wikipedia: http://en.wikipedia.org/wiki/Hindenburg_Omen

At Investopedia: http://www.investopedia.com/terms/h/HindenburgOmen.asp

P.S. In case anyone wondered, the last incorrect signal was generated in December 2005.

P.S. II – All paying members at FNArena are being reminded they can set an email alert for my Rudi's View stories. Go to Portfolio and Alerts in the Cockpit and tick the box in front of 'Rudi's View'. You will receive an email alert every time a new Rudi's View story has been published on the website.

Click to view our Glossary of Financial Terms