If you haven't heard of the Hindenburg Omen, here's a quick description from Wikipedia...
"The Hindenburg Omen is a combination of technical factors that attempt to measure the health of the NYSE, and by extension, the stock market as a whole. The goal of the indicator is to signal increased probability of a stock market crash.
The Omen is largely 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.
The rationale is that under 'normal conditions' a substantial number of stocks may set either new annual highs or new annual lows, but not both at the same time. As a healthy market possesses a degree of uniformity, whether up or down, the simultaneous presence of many new highs and lows may signal trouble."
In a Wall Street Journal story on June 11, Steven Russolillo interviewed Jason Goepfert, founder of Sundial Capital Research and author of the SentimenTrader Daily Report. Goepfert "said there have been a cluster of five Hindenburg Omens over the past few weeks. As a student of investor psychology and sentiment trends, he's worried about what potentially lies ahead," according to the article.
"That's a heavy concentration that we haven't seen too many other times over a span of nearly 50 years," Goepfert wrote in his Monday June 10 report. "And when we have, it hasn't been good."
In a study by Goepfert going back to 1965, which found 11 instances of Omen "clusters," the S&P 500 has risen only twice in the following 3 months with the average decline over 2%. The biggest drops were 10.5% in February 1980 and 9.8% in November 2007. "That's not a pretty short-to-intermediate-term picture," said Goepfert.
There's a Patriot Missile Aimed at Your Blimp
So far in the second half we are enjoying a mild continuation of the bounce off of S&P 1560 from last week. The first week of July has historically been bullish so this shouldn't surprise. Call it a patriot rally, if you will. It's the Q2 earnings fireworks that get things going ugly after Bastille Day.
But over the weekend, I found a stock market study that could take the Hindenburg down from its ominous perch without so much as poof in your portfolio.
A stock market historian by the name of Wayne Whaley looked back at 7-month rallies, followed by an 8th month in the red (what we just had in June) and he found a very powerful, and bullish, pattern.
"Since 1930, there have been 16 previous occasions of seven or more consecutive positive S&P months. The internal market strength necessitated to post a march of seven consecutive monthly wins merits respect, as all 16 previous cases resulted in higher S&P prices one quarter after the down month which concluded the streak."
Intuitively this makes sense to me. When institutions are in demand mode for equities and strength begets strength, it's hard to stop that train.
Now, I'm not so naive about statistics as to think they are always predictive, especially in markets where "wild randomness" is always afoot.
Strong Starts Lead to Strong Finishes
But I look at market behavior from multiple angles including institutional reactions to fundamental, technical, and sentiment data. And the bottom line here is that when the bulls want to run, you better not be in their way.
In his four-page report, Whaley cites a second study about powerful beginnings...
"A second angle in play in 2013 is the tendency for the S&P to finish a year in similar fashion from hence it began. There have been 23 previous post-1930 years in which December, January and February were all positive. The S&P was positive in the second half of 19 of those 23 years, gaining on average 7.72%, which is more than the average return (7.11%) for an entire year over the test period. There were 12 second-half-of-the-year double digit gains after this setup with no double digit losses."
Mary Poppins to the Rescue?
What does Whaley do with all this number-crunching? He pretty much thinks we can fly.
"Of the 16 previous cases of seven consecutive positive months outlined on page one and the 23 cases of positive Dec/Jan/Febs noted on page two, 11 managed to show up in both studies, thus satisfying my long standing requirements for a Mary Poppins signal."
Formally stated, a Mary Poppins Signal, as defined by me, occurs when:
1. The S&P experiences a negative month and,
2. There were at least seven consecutive preceding positive months,
3. Of which December, January and February participated.
Whaley's article has plenty of tables displaying his evidence, and he muses over his work this way...
"When Mary Poppins blows into town, the S&P was 'practically perfect in every way' measurable over the following six months."
Two questions to get us started:
1) How much weight do you give these bullish historical patterns vs. the Hindenburg Omen?
2) What are the chances in your mind that we are headed toward (a) a market correction of greater than 10% (S&P 1500ish) and (b) new highs in the next six months? Give your probability for each please.
I'll tell you what I think after lunch.