Andy Kilpatrick is a Warren Buffett maniac. He wrote a book -- no, the book -- on Buffett that's over 2,000 pages long. (War and Peace, by the way, is only about 1,400 pages.)

He's also made a fortune by putting his money where his mouth is. As Berkshire Hathaway's (NYSE: BRK-A) (NYSE: BRK-B) other famous face, Charlie Munger, once put it: "Andy was decisive. He bought all the Berkshire stock he could on margin. It went up, and he bought more. It went down and he sold just enough. In due course, he became quite rich."

Such stories of successfully buying stocks on margin -- using borrowed money to invest -- illustrate the findings of two Yale professors. Their study, titled "Life-Cycle Investing and Leverage: Buying Stocks on Margin Can Reduce Retirement Risk," has a simple recommendation: That young workers should supercharge their retirement accounts using 2-to-1 margin (two dollars of debt for every dollar of their own money), goosing returns in their early years, and hence reducing the need for risky stock exposure as they approach retirement age.

Oh no you didn't
Calling this idea controversial is being polite. Leverage is a blast on the way up, but can ruin lives on the way down -- as many homeowners can now attest. The arithmetic is simple. If you're leveraged 2-to-1 and your investment loses 50% of its value, you're toast. Game over. Insert more quarters. You've lost everything. And at various lesser stages of losses, brokers ask you to put up more money as collateral, called margin calls.

More important, we know big drops happen more frequently than we'd like. Both the S&P 500 and Dow Jones fell over 50% from Oct. 2007 to March 2009. From 2000 to 2002, the Nasdaq lost over 76%. Two of the most popular stocks in the world, General Electric (NYSE: GE) and Citigroup (NYSE: C), lost over 83% and 98%, respectively, from their 2007 highs to 2009 lows. And guess which company was one of the most widely held retirement account stocks in 2008? AIG (NYSE: AIG). You get the idea.

But what's amazing is that this study, which is the subject of a new book, recreated just about every possible combination of scenarios going back to 1871. Even factoring in downturns where leveraged investors would be decimated, the long-term returns of this strategy still beat non-leveraged returns -- provided investors dust themselves off and keep investing.

Pretty math, fuzzy assumptions
That's where I became a skeptic. See, I don't doubt this study's empirical findings. I'm sure the math works out nicely. I just question how people will apply the findings. What works on the chalkboard and what works in reality are often entirely different species.

In this case, the implied assumption that investors can suffer large losses and not be completely embittered with the market isn't realistic. Most of us aren't Andy Kilpatrick. The gut-wrenching losses of leverage would make our heads spin, our palms sweat, and, here's the kicker, our propensity to keep investing take a nose-dive.

Most people don't invest in steady amounts. They get excited and plow lots of money in near market tops, and throw in the towel after big losses. Confidence usually isn't recouped until big gains have returned -- usually near the next market top. That accentuates the risk and depresses the benefits of investing with leverage. And while I don't have the computing horsepower or freshman-intern workforce to run the numbers, I'm willing to bet that a more realistic scenario of how and when investors invest (and when they don't) would throw this study off track.

Consider this: From 1980 to 2008, the S&P 500 gained an average of 12.1% a year, assuming all dividends are reinvested. Yet if you missed just the top five best days, average returns fall to 10.9%. Miss the 10 best days, and you get 9.9%. The top 30 days, and you're down to 6.9%. Miss the top 50, and you're down to 4.3% -- worse than bonds.

And those disenchanted investors waiting out the storm probably will miss the best days. Of the 20 biggest one-day percentage jumps in the Dow, 19 occurred either during the Great Depression or in late 2008 and early 2009. The masses, it's safe to say, missed these gains entirely.

Our natural tendency to try and spot patterns and time the market is largely to blame for this. Humans have the unfortunate inclination to think big loses will be followed by even bigger losses, and big gains will be followed by everlasting bliss. As Jason Zweig writes in the book Your Money And Your Brain:

For decades, psychologists have demonstrated that if rats or pigeons knew what a stock market is, they might be better investors than most humans are. That's because rodents and birds seem to stick within the limits of their abilities to identify patterns, giving them what amounts to a kind of natural humility in the face of random events. People, however, are a different story.

Leave it to the pros (and idiots)
Now, the recommendations of this study might work for those with iron guts and unflappable focus. Indeed, the world's best investors aren't just financial geniuses, they're psychological Jedis. And what makes below-average investors fail isn't always that they're financially inept; it's that they're human beings that can be psychological basket cases.

If you're convinced that you're a member of the first group, have at it. But if you're humble enough to admit that you're part of the other 99% of us, steer clear of leverage.

Check back every Tuesday and Friday for Morgan Housel's columns on finance and economics.