Psst...want to make money—guaranteed!—in the stock market? Well, without fundamental reasoning to back it up, even the hardiest statistical trends will make liars of us all eventually, writes John Heinzl, reporter and columnist for The Globe and Mail.

I've discovered a system that's so amazing and so foolproof that even your dog could get rich using it. What's my secret?

Invest only in years ending with the number 5.

That's right. Be prepared to be blown away by what I'm about to show you.

Starting in 1935, the S&P 500 has posted a positive return in each and every year that ended with the number 5. The gains have been nothing short of spectacular: the average advance during those eight years was a staggering 25.3%—more than double the stock market's long-term average return of about 10%.

You're probably asking: How do I get in on this can't-miss opportunity?

Well, all you have to do is plow all of your money into the stock market on January 1, 2015, then sell everything on December 31, 2015. When 2025 rolls around, repeat the process. And so on, until you and your dog are relaxing on your own private yacht.

In case you haven't guessed, I'm being facetious. Although the outsized returns cited above are real—and, indeed, pretty amazing—clearly there is nothing special about the number 5 that would account for such outperformance, and no guarantee the trend will continue.

It's a fluke, an accident, and nothing more. Nobody of sound mind would build an investing strategy around a pattern that is so clearly a product of chance. Yet people take seriously other trends and patterns that are equally suspect.

Every four years, for example, the press is filled with stories about the US presidential election cycle, which claims that the first two years of a president's term are generally bad for stocks, while the third and fourth years are typically good.

The theory is that, early in their terms, presidents enact tough legislation, while in the third and fourth years—as the next election looms—they focus on feel-good measures that lift stock prices.

The election cycle theory has a plausible veneer, but I'm not buying it. Why not? Because the pattern, like years ending in 5, could just as well be a product of chance. If you slice and dice stock market data enough ways, patterns are bound to emerge that look compelling, but don't signify anything.

If you then try to make money based on the belief that those patterns will repeat, the only sure outcome is that you'll drive up your trading costs and possibly your taxes, too. Worse, the theory could stop working one day.

Anyone who followed the election cycle mantra in recent years, for example, would have gotten slaughtered. In 2008, George W. Bush's last full year in office, the S&P 500 plunged 38.5%. Oops.

In 2009, Barack Obama's first year in office, the S&P 500 soared 23.45%. It rose again, by 12.8%, in 2010, Obama's second year. All of these returns are the opposite of what the theory would have predicted.

You'd have done far better by simply staying invested, instead of trying to outsmart the market based on some half-baked theory. There are myriad other examples, ranging from the seasonal "Sell in May and Go Away" strategy to the patently ridiculous "Super Bowl Indicator."

These theories are fine as a source of entertainment, but as investment indicators, I put them on a par with fortune tellers and the daily horoscope. Avoid them if you know what's good for you.

Read more from The Globe and Mail here...