Here's a hot investment tip: there's a company out there whose stock goes up when the temperature is between 45 and 50, but goes down when the temperature is below freezing. I don't know what that company is, but I'll bet I can find one. Whatever the trading trick, you can always find a company to fit the new gimmick du jour.

With that in mind, I was very interested in an article published in Investor's Business Daily. This article declared the 200-day moving average rule: when the 200-day moving average starts to trend down, it's time to sell. The case in point, which of course provided a chart to support its argument flawlessly: General Electric. Take a look. The 200-day moving average descends. The stock goes down. Should have sold. Easy.

Good thing they didn't analyze IBM. If investors in IBM had followed this rule, they would have been screwed. A chart of IBM's performance, plotted against its 200-day moving average, tells a different story. The moving average started its decline in early 2000. Had you waited until that decline, you would have been exposed to a precipitous drop in the actual stock price in 1999 from 130 to 100. And if you sold in early 2000, you would have missed a subsequent rise from 110 to 130.

There was another time the 200-day moving average declined a couple of years later. If you waited for this cue, it would have been too late: the stock already tanked from 120 to just below 100.

It turns out that IBM follows another weather rule. If it's sunny or cloudy, choose an index and ignore these other gimmicks.