Do Markets Follow a Script?

The S&P 500 dropped sharply from January 23 to February 4, only to mount an equally sharp rally that erased the entire 6% loss.

Investors are now trying to figure out the implications of that V-shaped recovery. At the simplest level, the question is: “Are stocks going higher or lower from here?”

As always, history serves as a good guide.

Erik Swarts, from, is one analyst I follow regularly. His analyses of current market trends and price patternrns are always set against the backdrop of similar looking trends and patternrns from the past. He then uses those historical price patternrns, called analogs, to forecast the likely direction of future prices.

You can think of this concept as the current market “following a script” that past markets wrote.

Here’s one example, comparing the patternrn traced out by the S&P 500 in 2004 (in black) and recent S&P 500 prices (in blue):

See larger image

I made one annotation to Mr. Swarts’ chart, highlighting the V-formation patternrn that we just experienced.

Interestingly, that same patternrn occurred during March 2004. Back then, the S&P 500 traded lower after the formation, printing lower lows and lower highs into September 2004.

So can we expect the S&P 500 to follow the same path this year?

Before I answer that question, let me make one thing crystal clear: I do not think that market analogs are predictive. They only provide a road map of what prices might do.

There are plenty of dissimilarities between today’s market and 2004.

For one, the V-formation in 2004 was preceded by a 12-month gain of 44%, while the current S&P 500 patternrn follows a gain of only 25%.

More importantly, 2003 was a recovery year, as stocks rallied strongly out of the late 2002 bottom. As such, it’s difficult to compare that market to the current market, which is now in year six of a bull run.

Finally, if you look at the dates on the chart above, you’ll see the months don’t match up. The 2004 V-formation began on March 5, while the current V-formation began in mid-January. So, seasonal differences will affect this comparison to some degree.

The limitations of market analogs are not meant as criticism of Mr. Swarts’ methodology. They only serve as a reminder that, while historical patternrns can be very useful, they’re not a fool-proof way of predicting future prices.

The same limitations apply to statistical analysis of historical prices — a methodology on which I rely heavily.

I recently shared some statistical analysis with Cycle 9 Alert subscribers that showed V-formations have historically spurred bullish moves in stocks.

To quantify the formation, I established two criteria:

  1. A one-week gain of 2% or more in the S&P 500, which was preceded by…
  2. A high-to-low loss of 5% or more in the S&P 500 over the preceding three weeks.

These criteria were met on February 14, when the S&P 500 rallied up to 1,835. They were also met 37 times in the last 14 years… and 70% of the time, the S&P 500 was trading higher two to three months later, gaining an average of 5.5%.

The chart above cuts off all data after September 2004, but if it were extended beyond, you’d see the S&P 500 was higher by the end of the year. In fact, the market gained 5.5% and 7.2% over the two and three months following the 2004 V-formation.

We’ll see what 2014 holds. But with history as our guide, it makes sense to maintain a cautiously bullish bias while being highly tuned to any signs of an impending crash.

P.S. Cycle 9 Alert subscribers received this research a week ago. Click here to learnrn how you can receive my timeliest research and recommendations.