August 20, 2010

"Hindenburg Omen" is Nonsense

You may have seen some news lately about a "Hindenburg Omen" which is a complicated pattern of stock signals that experts say signals a market crash.    By "complicated pattern of stock signals" I really mean a bunch of random coincidences that are basically meaningless when put together and by "experts" I mean scare mongering idiots who somehow got people to listen to them.  

Think I'm being too harsh?  Just read this bit about the so called omen from a WSJ article on the topic:
"The Omen was behind every market crash since 1987, but also has occurred many other times without an ensuing significant downturn. Market analysts said only about 25% of Omen appearances have led to stock-market declines that can be considered crashes."

OK lets parse that.  The Omen has happened "every market crash since 1987".   But also happened "many other times" without a crash.   Ok so right there its obviously not very dependable.   Then analysts say it precedes a crash "only about 25%" of the time.  So the Omen happens before "every" market crash except the "many" times that it doesn't and it is wrong 75% of the time.  

Wow.  Thats about the dumbest "omen" I've heard of.   If 75% of the time the market DOESN'T crash then isn't this Omen a more accurate prediction that there WON'T be a market crash???    75% > 25% last I checked.    I sure hope 25% correct is not taken by anyone as a 'good' rate.    25% correct gets you an F in middle school and it should get you an F in stock market analyst circles too.

Here's an example.   This article The Past Performance of the Hindenburg Omen Stock Market Crash Signals 1985 - 2005  was written on Oct. 2, 2005.  He says: "During the past two weeks, we have had six Hindenburg Omen signals".    Lets look at the stock market in the period from Sept 2005 to Oct 2006:


Source: Yahoo Finance


S&P 500 is in blue, Dow Jones in Red and NASDAQ in green.    Do you look at that chart and declare to yourself that it would have been a good idea to sell everything in October 2005?   I don't.   You can see that from September to October of 2005 the market did drop a few % points.   But by December it had recovered that loss and was up a few points.    Then 12 months after the Hindenburg Omen events occurred the market was up 5-10% points.     

That same article also says: "Our research noted that plunges can occur as soon as the next day, or as far into the future as four months."   I don't really find it too useful for someone to say that the stock market might go down sometime in the next 4 months.  In fact I think its a fairly good bet that will happen.   The stock market is down 5% in about 1/8 months.    Over the past 80 years if you took any random month and said that in the next four months the market would drop 5% then you'd be right about 33% of the time just randomly.

You can call a string of coincidences an Omen but it doesn't mean anything.  It isn't necessarily anything more than a string of coincidences stringing out into the past.   If I slap my knee and then flip a coin and it comes up heads that doesn't mean anything.  If I do it 3 times in a row it still doesn't mean anything.  I could call it the "knee slap omen" if I want but its still meaningless coincidence.  If I slap my knee and then flip a coin a 4th time then the chances it will come up heads are 50/50.


 Bottom Line:  The Hindenburg Omen has preceded a crash only 25% of the time.   Seriously folks this so called Hindenburg Omen is total nonsense.    

Blog Widget by LinkWithin