Standard & Poor's 500-stock index plummeted 8.6 percent in January, leading many prognosticators to conclude that stocks will likely continue to drop the remainder of the year. Don't count on it. After falling in January, stocks historically have tended to rise in the subsequent 11 months.

Like every myth, the so-called January barometer contains a grain of truth -- but only a grain. The stock market's performance last year provided ammunition for this barometer's fans: The S&P 500 plunged 6 percent in January, and the index lost 38 percent for the entire year.

The January barometer started as a belief that if stocks rise in January, they will rise during the subsequent 11 months of the year. At first glance, this indicator seems to have some validity. When the market rises in January, two-thirds of the time it goes up the rest of the year.

But Mark Hulbert, editor of the Hulbert Financial Digest, put the kibosh on this nonsense. The fact is that the stock market goes up about two-thirds of the time -- regardless of what happens in January. So the barometer is practically worthless.

Next, devotees turned the barometer on its head: If the market falls in January, it tends to fall for the rest of the year, the theory's advocates argue.

Once again, reality says otherwise. Hulbert, whose publication tracks the performance of investing newsletters, looked at returns of the Dow Jones industrial average from 1897 through 2008. He found that when the Dow fell in January, its average monthly return for the subsequent 11 months was 0.25 percent.

Still, a negative January does contain a smidgen of predictive value. In an average month, the stock market rises 0.57 percent -- more than twice as much as it does over the 11 months following a losing January.

Indeed, Hulbert found that when the market rises in January, on average it gains 0.69 percent per month from February through December. That's a soupcon more than the 0.57 percent the market returns in an average month. But here's the catch: The return for most months exhibits a slight tendency to predict which way the market will head next.

It turns out that market performance in December has been slightly more accurate at predicting the subsequent 11 months than its performance in January (the S&P 500 rose about 1 percent last December). November is the third most accurate month (the S&P plunged 7 percent in November).

Why does one month's return predict what will happen over the ensuing 11 months? Momentum works, to some degree, in the market. When the market is falling, odds are it will continue falling. When it's rising, odds are it will continue rising. Continued...