Wouldn’t it be great to have a money machine? Just type some commands and it would spew out neat stacks of hundred-dollar bills while you got in a leisurely round of golf or had your nails done.

If the idea sounds preposterous, idiotic or, at the very least, illegal talk to some investors. Scratch the surface and you’ll find that many, perhaps most, have “systems” just as you’d find with roulette players.

And now, Charles Schwab, the big discount brokerage, has introduced its own system using a stock-picking computer instead of the traditional Wall Street analyst. Human analysts have lost favor amid accusations that some tout bad stocks to please corporate banking clients. Schwab says its system is an alternative to the conflict-ridden advice offered by rival firms.

There have been many automated, or virtually automated, stock-picking systems in the past. At the extreme, you find chartists, who believe money can be made by studying the up and down patterns seen when a stock’s price changes are traced on a sheet of graph paper. If a stock hits a low, recovers a bit, then hits another low and starts back up, that W shape on the chart is a signal to buy, they figure.

Not all systems are bad, of course. After all, everyone needs a way to sift the tens of thousands of stocks, bonds and mutual funds in the marketplace. It might be perfectly rational, for example, to say: “Well, I’m not going to buy stocks in companies that have never shown a profit.”

Schwab system flawed

Schwab has unveiled its own system. Well, I shouldn’t say unveiled, since the details are, in fact, secret. Schwab says its computer crunches 24 types of data “correlated with stock performance” in four categories “fundamentals, valuation measures, momentum and risk.” Among the factors are standards such as price-to-earnings ratios, earnings-estimate revisions by Wall Street analysts, stock-price volatility and free cash flow at the companies studied.

The result is an A-to-F ranking for each of the 3,000 stocks covered, based on prospects for the next 12 months. The best 10 percent get an A rating, the next 20 percent are Bs, the middle 40 percent are Cs. After that, the nearly failing 20 percent are Ds. The worst 10 percent get Fs.

Because Schwab won’t describe the exact criteria, it’s impossible to know whether the system makes sense. The company has “back tested” looked at how it would have performed had it been used in the past. But industry regulations bar the release of those data, according to Greg Forsythe, a Schwab vice president and director of equity research.

He also said he was barred from disclosing test results showing the statistical probability that the A stocks would beat the market average. But he noted that other studies have shown that even the best traders people like Warren Buffett have only a slightly better than even chance of beating the market.

Presumably, Schwab is in that ballpark: Buy all 300 A stocks and you’d have, perhaps, a 55 percent chance of beating the market assuming the factors that determine winners and losers are the same in the future as in the past.

Forsythe said Schwab will release performance data on its system in the future. But it will take some time for the data to be useful, since tracking only began May 1.

For the time being, the skimpy facts Schwab did release are cause for concern. For one thing, the 12-month performance horizon is pretty short. It really suits only active traders rather than long-term investors.

Also, the system automatically gives a top rating to the 10 percent of stocks with the highest scores. If you took a universe composed of nothing but lousy stocks, 10 percent of them would get top ratings nonetheless. There are times when stocks are not a good investment, period when the entire universe is, in fact, lousy.

The biggest problem with any automated system is that it flies in the face of what life teaches us about excellence. Great decision-making cannot be reduced to a formula, else we’d all be great generals, winning NFL coaches or billionaires.

Even if Schwab’s top-rated stocks do, on average, beat the market, you can bet that many individual A stocks will do worse than the market. To have a good chance of matching the A-list performance, you’d have to own about 40 of its stocks, Forsythe said. That’s an awful lot for a small investor.

You’d also have to be ready to jump in and out of them at a moment’s notice. That’s because, if the A list really is an effective predictor, investors will jump on stocks as they are added to the list. That demand will drive the prices up until those stocks aren’t bargains anymore. You’d have to beat the rush.

Schwab acknowledges the ratings should mainly serve as an initial screen to be followed by an evaluation of “more qualitative perspectives and company news.”

But that reads like a line the lawyers made them write, just in case a big A-list investor loses money and sues. Legal caveats aside, Schwab is pushing its ratings system as a great place to find winners.

A game of luck

History has shown that effective stock picking, year in and year out, is an extremely rare talent. In fact, it may not exist at all. With thousands of experts picking stocks all the time, some are bound to stumble on successive winners through bum luck, just as a roomful of coin flippers will have a few who flip all heads.

For most small investors, picking individual stocks is a losing game. Even if you can pick market-beating stocks, you’re likely to lose your extra gains to trading commissions and taxes.

If you hit upon a really big winner, it won’t be from a system because the really big winners are the ones that defy the precedents on which systems rely.