I was in Orlando this weekend at the American Association of Individual Investors annual conference. Investor James O'Shaughnessy, who I greatly admire, gave a wonderful talk called "What works on Wall Street." It got me thinking about something important.

It's actually one of the things that drives me crazy about finance.

O'Shaughnessy began noting that in Benjamin Graham's day, the first half of the 20th century, good financial data was extremely hard to come by, and good historical data going back more than a few years was nonexistent. This made studying finance close to impossible. Graham (Warren Buffett's early mentor) once said that until more data was available, investors couldn't dare call their trade a profession. It was an art, but not a profession.

Faster-forward almost a century, and now we're drowning in financial data. A child in the middle of Africa with a $100 cell phone and internet access can access historical market data going back decades for free on Yahoo! Finance. Ben Graham couldn't dream up such an incredible resource if he tried.

When computers took off in the mid-1990s, O'Shaughnessy gathered reams of historical market data to figure out what kind of investing techniques have actually outperformed over the long run. He spent years crunching the data and came to some smart conclusions, which I'll oversimplify as:

  • Buy cheap stocks, ranked on various metrics like price to earnings and price to sales.
  • Make sure they're financially sound stocks. Avoid highly levered companies.
  • Buy a basket of them.
  • Rebalance your portfolio once a year, selling what's expensive and buying what's cheap.

Doing this over time works. You can beat the market. The data on it is persuasive.

So, why doesn't everyone do it?

According to O'Shaughnessy, it's partly because few have the discipline to follow a formulaic approach. Investors like to tell stories about their stocks. They invest in what's flashy, or companies they know and love. And those companies, O'Shaughnessy says, tend to be expensive. Twitter. Facebook. Tesla. People are crazy about these companies, and their shares are pricey. So as a group, they tend to underperform over time.

The cheap companies – whose stocks do well over time – tend to be companies you're embarrassed to hold, or haven't heard of. O'Shaughnessy mentions DirecTV, which currently comes up on his list of cheap stocks. "People hate the company," he said. "Every time they send someone out to my home, they look like ex-cons. My wife doesn't want them in the house. They show up and they look like they just got out of Folsom prison." It's not a sexy company. But it's a cheap stock. And cheapness is what could make it do well over time.

The crux of investors' problem, O'Shaughnessy says, is that most people think of a stock's name, not its underlying characteristics. "A great company is not always a great stock, and a great stock is not always a great company."

I tend to agree with this.

But it also kind of drives me nuts.

Why?

Because other equally smart, and equally successful, investors could give the exact opposite presentation and be just as convincing.

When Warren Buffett once talked about why he bought Coca-Cola stock, he said, "Forget about share of market, I'm talking about share of mind." People love Coke. It's the company's most valuable asset. And that intangible, warm-fuzzy moat explains why Coke stock has done so well over time.

The central thesis of Buffett's investing approach over the last 30 or 40 years is buying good businesses. Ideally, you want to buy them at a bargain price, but Buffett clearly states that quality can trump value. He wrote to shareholders in 1989 (emphasis mine):

Shortly after purchasing Berkshire, I acquired a Baltimore department store, Hochschild Kohn, buying through a company called Diversified Retailing that later merged with Berkshire. I bought at a substantial discount from book value ... How could I miss? So-o-o-three years later I was lucky to sell the business for about what I had paid ...

I could give you other personal examples of "bargain-purchase" folly but I'm sure you get the picture: It's far better to buy a wonderful company at a fair price than a fair company at a wonderful price. Charlie understood this early; I was a slow learner. But now, when buying companies or common stocks, we look for first-class businesses accompanied by first-class managements.

Philip Fisher, another of Buffett's early idols, emphasized the "scuttlebutt" approach to investing, studying a company's intangibles that don't show up on financial statements -- things like management integrity, openness to change, supplier relations, and customer awareness. He, too, wanted good businesses as opposed to just cheap businesses. Peter Lynch followed a similar approach when jumping on the beds at LaQuinta Inn and drinking the coffee at Dunkin' Donuts before investing. There's even long-term data showing the most admired companies outperform the market over the long run.

If any one of these investors gave a presentation about how warm-fuzzy business factors can trump numeric value, I guarantee they'd be just as persuasive as O'Shaughnessy.

Even though we have more financial data than ever, investing is just as much an art as it was when Ben Graham criticized it almost a century ago. Investing is not a hard science like chemistry or physics. There are no laws or unbreakable rules. Equally smart people can have opposite views and, oddly, be equally successful. Keep this in mind when determining what works. 

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