Posted by on Jun 28, 2014 in Articles & Advice, Blog, Columns, Posts |

By Jason Zweig | Jun 27, 2014 7:04 pm ET
Image Credit: Christophe Vorlet

Let me guess. You are sick and tired of earning nothing on your bonds and cash, and the stock market still scares you.

I have the answers to your investing prayers: simple strategies that have delivered gigantic returns, at moderate risk, for decades. You can invest this way in the comfort of your own home and dump your overpriced, underperforming financial adviser.

First is my Graham and Buffett Portfolio. This strategy, named after two of the best investors of the past century—investment analyst Benjamin Graham, who died in 1976, and his greatest student, Berkshire Hathaway Chairman Warren Buffett—has returned an average of 12.5% annually for the past 20 years. That dwarfs the average annual return of 9.2% of the S&P 500.

If you had invested $10,000 in this portfolio two decades ago, you would have $105,971 now. Meanwhile, the losers who stuck with the stodgy S&P 500 ended up with only $58,347.

Still not sure you’re ready to reach for investing greatness with me? My Presidential Political Party Portfolio makes Mr. Buffett look like a birdbrain. It has crushed the S&P 500 over the past one year, five years, 10 years, 15 years and 20 years. If you had invested $10,000 in the PPPP in January 1994, you would have $127,220 now. That is more than double the results of the stock market.

How did I do it? I designed these portfolios with the help of Chris Covington and his colleagues at AJO, an institutional money manager based in Philadelphia that, I hasten to add, doesn’t invest its $24.4 billion in assets this way.

Here’s why:

The holdings in the Graham and Buffett Portfolio consist entirely of stocks whose ticker symbols contain only letters that are included in the names Benjamin Graham and Warren Buffett. Some examples: Badger Meter (BMI), Ubiquiti Networks (UBNT) and Weatherford International (WFT).

The companies in the PPPP have ticker symbols derived exclusively from letters in either the word “Republican” (Rubicon Technology, or RBCN) or “Democrat” (Caterpillar, or CAT). The portfolio holds only Republican-tickered stocks in years when the GOP holds the presidency. When a Democrat is in the White House, the PPPP holds only stocks with “Democrat” tickers.

These portfolios aren’t real, of course. But the results are. Had you invested an equal amount in each stock in these portfolios 20 years ago, you would be far richer today.

But it never would have occurred to you—or anybody—to buy them, because Mr. Covington and his colleagues went “data-mining” to find something, anything, that would produce groups of stocks with high returns between 1994 and the end of 2013.

Fortunately, Mr. Covington’s research team already knew exactly what every stock returned over those years, so they just kept digging, with 20/20 hindsight, until they found random variables they could use to form portfolios that would yield the highest performance.

Given long enough time and deep enough data, you can find a seemingly impressive linkage between almost any two factors. But correlation isn’t causation. Stocks with these letters in their trading symbols did beat the market—but their outperformance wasn’t caused by their tickers. You also could have beaten the market over the past two decades by buying companies based in ZIP Codes whose digits add up to 21. Invest like this, and you are kidding yourself.

The Graham and Buffett and PPPP “strategies” came out of the process known in the investment industry as “backtesting.” It is often legitimate—seeing how a strategy did in the past can be informative if there is reason to believe it will persist—but backtesting also can be hard to evaluate.

An ad that appeared earlier this month in The Wall Street Journal and elsewhere declared in bold type that “100% is the historical probability of outperformance” for a new stock index.

That is because this index beat a rival benchmark over every one of the three-year periods measured weekly over the past 10 years, says Noël Amenc, a finance professor at EDHEC Business School in Nice, France, and chief executive of ERI Scientific Beta, the Singapore-based firm that ran the ad.

“Popes have been trying to achieve infallibility for 2000 years, and these people have finally done it,” says David Bailey, a retired mathematician and computer scientist at Lawrence Berkeley National Laboratory in Berkeley, Calif., who has published several articles that criticize financial backtesting. If many trials are done on the same data set, he says, something will inevitably be found to outperform—even if it is a “false positive” caused by luck alone.

Prof. Amenc stands by his claims. He told me by email that Scientific Beta takes several “precautions against overfitting”—or capturing variables that won’t predict future results. The firm also publishes performance and risk data on its website for all its indexes. That ability to analyze returns, he told me, enhances “the informational efficiency of the market.”

Be that as it may, investors should form the habit of asking every financial adviser or money manager: Are these results backtested? How many other strategies did you test before you settled on this one? How far back did your historical data go? When did you begin running this strategy live, with real money? Have you made changes to the strategy since, and if so why?

If you don’t ask questions about performance that was plucked out of the past, you are likely to end up disappointed about the returns you get in the future.

Source: The Wall Street Journal

See also: False Profits