Thursday, August 26, 2010

WSJ Disses the Hindenburg

WSJ this morning has a piece pooh, poohing the Hindenburg Omen.

Although the piece contains a nice profile of the creator, Jim Miekka, the positives to the article end there:
Mr. Miekka's foray into stocks began after he was injured while conducting experiments at a Massachusetts mine, where he was trying to find a better way to extract minerals from rock. There was an explosion from chemicals he was working with, and was blinded by complications during an ensuing eye operation. "The last thing I saw was the eye chart going into surgery," he said.

As he recuperated, Mr. Miekka said he began listening to television shows that focused on investing, and began actively putting his money into the market.

He came up with his first trading "system" in 1989...

Market indicators aren't his only inventions. He also says he created artificial-vision technology that uses sounds to help him identify targets better than m any sighted shooters. He can hit a National Rifle Association target at 100 yards, and this week he hit a bowling pin at 200 yards.
WSJ then  blasts his indicator:
"People are grasping at straws and always looking for someone who might have all the answers," says Jeremy Siegel, finance professor at the University of Pennsylvania's Wharton School of business
Siegel, just by coincidence, according to his web site, seems to 'have all the answers." He tells us that he is the "Wizard of Wharton". Jim Cramer thinks he is God:
Jeremy Siegel is one of the great ones. [His article at the market top was] one of the most stark and prescient calls I have ever seen
What market top is not exactly clear, that insightful fact is missing from the Cramer quote.
I highly suspect it wasn't the recent top.

You see Siegel had this to say about the economy just 4 months ago:
Yes, the recession has definitely ended. What they were more uncertain about was what month it ended, but that is quibbling about history. My feeling is that it was July or August of last year. But there is no question that we are out of it. In my opinion, we are not going to have a double dip. It is just a question of whether we are going to come out of this with moderate growth or surprisingly rapid growth.
I wonder if he thinks the same now, given housing, unemployment and production are tanking? Oh well, you can't get them all.

Moving on, the other expert WSJ turns to is Barry Ritholz. Ritholz also does a heavy "analytical" attack on Miekka's indicator. He tells WSJ that it is "recession porn". Thinking that Ritholz has a desire to get in the paper and so threw out a sound bite for WSJ, and has his real analysis at his website, I turned to his web site to see the deep analytical work behind his classifying the Hindenburg Omen as "recession porn".

His complete analysis of the Hindenburg Omen is a chart that shows the term trending up in Google trends. I am not making this up. Yup, because the economy is clearly in the tank and people are trying to learn about the "Hindenburg Omen", Ritholz disses the indicator because, get this, 280 people searched for the term. You are more likely to run into someone suffering from kuru disease than you are someone who has searched for the term "Hindenburg Omen".

Ritholz is trying to indicate that because a term has become "popular" (Apparently he considers 280 searches popular) that the rule of  thumb that being on the opposite side of crowds should be kicking in. Of course, any trader who has traded more than a week knows that crowds can grow very large and that they can take a very long time to reverse. Even given Ritholz's absurd implication that 280 searches in the sea of investors is a crowd, the crowd indicator as far as timing is concerned is a terrible indicator. Cash flow into the market, new highs and lows and current momentum (all part of the Hindenburg Omen) tell you a lot more. Thus, WSJ is really quoting Ritholz using a very weak indicator (the weakness amplified by the  absurd tiny size of the "crowd") to diss the Hindenburg Omen which is rich in valuable data. Amazing.

Who knows if WSJ reporters have an agenda or are just clueless, from experience I know that most reporters love sound bites over analysis. There is nothing necessarily wrong with a sound bite, as long as there is analysis behind the sound bite. Ritholz has none and Siegel can't even get the direction of the economy right.

I have long been a preacher of being very careful of empirical indicators. You need to know what is behind the indicators (which I did for the Hindenburg Omen here), but you also need to know what the analytical work (or lack of it) is behind the "experts" WSJ quotes. There's zero analytical work by the "experts" in this WSJ article.


  1. I dissed the Hindenburg omen due to its awful track record -- its worse than a coin flip (25% accuracy).

    When something with that poor a record gets all this press, its value is of a sentiment / contrary indicator.

    People are looking backwards, and sentiment is lagging. The time to worry about a collapse is BEFORE a 55% drop -- not afterwards. For crying out loud, even Tony Robbins is giving dire economic warnings. (

  2. Barry,

    I'm missing your point here. Are you suggesting that all we have to do is the opposite of what Tony Robbins is saying?

    Can you please direct me to the data on the awful track record of the Hindenburg Omen?

  3. Wenzel,

    Great job on Siegel. I don't get this guy and why his opinion is so sought after. Does he successfully manage billions of OPM? Does he unsuccessfully manage billions of OPM (seems to be just as worthy of "expertise" in the media these days)? Or is he just another loud-mouthed college professor, a man of great gentlemanly eminence and vast scholarship, no doubt, but a man who has neither the balls nor the resources to practice what he preaches?

    I would LOVE to see a study of how "Professor" Siegel went from a nobody to the media's go-to guy on all matters concerning the financial markets.

  4. I don't have track record statistics, but other arguments against the relevance of the latest HO cluster are generally based on (a) the changing composition of the NYSE, and (b) the fact that the HO's are coming after an already substantial decline. As to (a), inverse ETFs and more interest rate sensitive issues, such as closed end bond funds, are claimed to distort NYSE breadth statistics. As to (b), Ian Woodward has a good discussion here:

    Also, data in Google Trends graphs are expressed as index values and are relative.

  5. The Google Trends number is not actually the absolute number of searches performed, it is some relative index to the total number of searches which Google does not completely disclose.

    "Google Trends analyzes a portion of Google web searches to compute how many searches have been done for the terms you enter, relative to the total number of searches done on Google over time. We then show you a graph with the results – our Search Volume Index graph. "

    Regardless, pointing to public interest in something certainly can't work to classify anything as right or wrong.

  6. @BobEnglish

    A. If you look at the stocks in the new lows list they are not much related to changes in NYSE. The new highs are all pretty much interest rate related stocks. I would not consider that a comforting indicator for teh bulls.

    B. What can I say? Trends can continue for a long time, that the stock market has been down for some time doesn't mean it can't go lower, even a lot lower.

    Woodward's analysis looks to me like technical mumbo jumbo, discussing extreme fine points of empirical data that can not be explained in terms of human action.

    I am not looking at the HO for anything more than an indicator of broad based action. This analysis along the lines of it needs to cross a value at X and it missed by 0.000001 does not interest me.

  7. Ritholz is deceptively claiming the "horrible 25% accuracy because the HO signals a full blown stock market crash only 24% of the time. The fact that another 65% of the time there is a significant stock market decline is conveniently (dishonestly) ignored by him.

    Nevertheless, that statistical correlation is radically important to investors. That leaves the main question as to why Ritholz would deceptively cherry pick his info. Is he on the take from the FED, or just trained by those who are? Even Milton Friedman warned that the FED corrupts the Econ profession by subsidizing economists.

  8. Regarding my brief summary of anti-HO arguments above, if composition of new NYSE highs and lows are indeed in actual stocks, I would agree that knocks out the first argument.

    Pushing aside the techno-talk of the second argument, schizophrenic market behavior--which is in part what the HO is attempting to measure--after a long rally should be a bigger warning than after an ongoing decline.

    Investor euphoria of being near a high encourages complacency, is the time of the highest concentration of unhedged directional bets (up), and will exacerbate selling once sentiment reverses. In contrast, at present, institutions that weren't wiped out by the flash crash became extremely cautious and very well hedged, which accounts for the mostly sideways, range-bound action over the summer.

    This is not to say the market won't crash, but there are much better and more precise indicators based on hedging activity and liqudity.

  9. Barry Ritholtz is a smart guy.

    He's smart enough not to put any specific stock picks or market calls on his blog - now that he has got some assets to skim, er manage!

    It was a year ago now that he got absolutely destroyed defending *subprime lending*'s role in the housing/mortgage/banking crash:

    Actually, Barry's simply not that bright. Get this, on his Amazon wish list, he has listed among an astounding 849 items(!) the most puerile *climate change* propaganda books. You know....because there's not enough of the bunk out there in the public dialogue for him to wrap his mind around!

    In the end, there's nothing special about him. He's hardly the first clown out there to mistake making a little money for *intelligence*...

  10. however great it is, it's still just technical analysis.

    miekka ran from stocks, fine. (he should have in 2007 or 2008, but didn't. oh well.) but he didn't then give his money to charity, right?

    so did he buy gold?