How many stocks should you own?

(MONEY Magazine) -- With the stock market as bouncy as a beat-up couch, you may be thinking it's time to focus on a small number of stocks that you know really well. What better way to keep returns up and risk down?

Conventional wisdom and new academic research certainly seem to suggest that this is the way to go. Many financial planners and brokers will tell you that a portfolio of as few as 12 stocks (and up to 30) will sufficiently diversify your holdings.

Jason Zweig is the author of the new book 'Your Money and Your Brain.'

More from Money Magazine

Correcting your credit report gets easier Cut your cooling costs by 30% or more After Obamacare's launch, a look ahead

Best Places to Live Current Issue Subscribe to Money

And three recent studies have found that individuals who own fewer stocks do better than those who own many.

However, as is often the case with conventional wisdom (and academic research), there's a lot more to the story. Fact is, if you build your portfolio entirely on the principle of "less is more," you're a lot more likely to end up with less than more. Here's why and what to do instead.

The idea that 12 to 30 stocks are all you need dates back at least to the 1960s, when academics, including Burton Malkiel, author of the classic "A Random Walk Down Wall Street," concluded that that's what it took to eliminate most of the risk from a portfolio. (They usually defined "risk" as the chance of suffering big swings away from the average market return.)

But back in the days of hula hoops and transistor radios, and before computer-generated trading became common, stocks didn't bounce around the way they do today. In 2001, Malkiel found that it took 50 stocks to get the risk reduction that 20 used to provide. Others estimate that true diversification requires hundreds of stocks.

Just recently, however, researchers studying the performance of individual investors have discovered something that, at first glance, seems electrifying: The more concentrated a portfolio is, the higher the returns. One study found that investors whose portfolios were dominated by one or two stocks outperformed the most diversified stock owners by 0.8 to 4.8 percentage points annually on average. That's a huge gap. And roughly 8 percent of the top performers had portfolios concentrated in a single stock.

So the heck with diversification, right?

Well, not exactly. First, the least-diversified investors frequently lagged the market; they just lagged by less than investors who held more stocks.

Second, because stock returns are so uneven, the "average" undiversified investor doesn't really exist. At any given point, there are something like 10,000 stocks in the United States. Most of them are mediocre, but a handful are what Bill Bernstein of Efficient Frontier Advisers dubs "superstocks," capable of delivering gargantuan returns for years. Think Microsoft (Charts, Fortune 500) in the '90s, when it returned 9,000 percent. (More often superstocks are lesser-known companies.)

Across a large group of people whose portfolios are mostly in one or two stocks, the lucky few with superstock portfolios will make the group's average return look great, even if the vast majority of individual members have lousy or middling results.

On the other hand, investors who spread their bets across dozens of stocks have only a slightly better chance at catching a superstock. And if they do land one, it won't have nearly as much impact on their portfolios, or on the group's average return.

So the real story is that you need a lot of stocks to be adequately diversified, and you need a concentrated portfolio to give yourself a shot at striking it rich. An unsolvable catch-22? Hardly. In fact, you can have it both ways by employing a straightforward, two-part strategy.

First, direct 90 percent of your U.S. equity allocation into a total stock market index fund that automatically gives you a stake in thousands of companies. That guarantees you a piece of every superstock that already exists or might emerge later - and, more important, it means you'll be adequately diversified and your investing costs will be at rock bottom.

Then pursue your search for the next Microsoft or Google (Charts, Fortune 500) by researching the daylights out of a very small number of companies and putting the remaining 10 percent of your portfolio into your one-to-three best ideas. This way you'll let yourself have a little fun. You will also minimize your risk and maximize your hope.