'Past performance is no guarantee of future results.'

This phrase, ubiquitous in the small print of financial products, often falls on deaf ears, according to Adam Reed, a finance professor at the University of North Carolina at Chapel Hill. 'Investors have a tendency to rush into those funds that are at the top of the previous year's performance rankings,' he says, citing numerous studies.

They shouldn't. The evidence is that last year's top performers will lag behind the market in 2014--if not lose a lot of money.

You should instead be focusing on strategies with superior track records over far longer than just the last 12 months. More like 15 years, in fact.

Consider a hypothetical portfolio that, each Jan. 1, began following the strategy espoused by the investment newsletter that--among the more than 500 tracked by the Hulbert Financial Digest--had the best record over the previous year. Over the past two decades, this portfolio would have been a disaster, losing an average 17% a year.

Such a 'follow the winners' strategy didn't trail the S&P 500 every year. But in the majority of years it did, and in a handful of cases it suffered huge losses. That is because one-year performance rankings will almost always be dominated by high-risk strategies that hit it big. When they don't pan out, those same strategies also can lose big.

Take the portfolio with the best 2013 return, according to the Hulbert Financial Digest: the 'Traders Portfolio' of the Medical Technology Stock Letter, edited by John McCamant. It gained 216% in 2013 by investing in emerging biotechnology companies. The S&P 500, by contrast, returned 32%, including dividends.

When judged by the volatility of its returns, a common measure of risk, the portfolio has been more than three times riskier than the market as a whole. Its worst single calendar-year performance, in 2008, was a loss of 68%, versus a loss of 37% for the S&P 500 index, assuming dividends were reinvested.

Mr. McCamant, in a telephone interview, acknowledged the volatility inherent in the biotech sector, though he added that, in the future, he will be 'doing his best to manage that volatility.' He says that 'biotech stocks are one of the premier growth sectors going forward,' and that even though another 216% gain in 2014 is unrealistic to expect, 'a triple-digit gain is still achievable.'

So does the 'first shall be last' pattern mean you should instead follow the lead of the worst performer of the previous year? Absolutely not. A portfolio that did that over the past 20 years would have lost even more than the follow-the-winners strategy.

That shouldn't be particularly surprising, since the bottom of the one-year performance rankings also will be dominated by risky strategies whose bets didn't work out.