For most people, mutual funds and exchange-traded funds are probably the best options for IRAs. However, I must admit it is a bit discouraging to embrace a strategy that simply performs in line with the overall market.

Sure, a stock portfolio invested in an index fund will participate fully in the growth of the economy over time. But it seems positively un-American to accept mediocrity in investment results. Isn't there a way to obtain better-than-average results given the amount of risk taken? And can't those good results be achieved without spending an inordinate amount of time working on investments?

I personally think it is possible by doing three things:

• Commit to being an unemotional and independent investor.
• Concentrate your search for investment ideas to a universe of smaller, less widely followed companies.
• Be a mechanical investor.

The ability to trust in yourself and follow your set of investment principles no matter what is happening in the news is critical. The problems with the Iraq war, the Japanese carry trade or subprime mortgages should not even be on your radar screen when it comes to your investments. In addition, the future as seen in the crystal ball of your favorite analyst should be given very little weight in your investment decisions.

If you do go this route, I believe your best opportunities will be found in the smaller, less widely followed companies. This seems like pure common sense to me, since enormous companies have huge followings that generate even more efficient pricing than is normally present in the market. The best hope to find an attractively valued company is to look where most others are not looking.

Notice that I am not advocating a relaxation of quality standards when evaluating an investment idea. There are plenty of smaller, less-well-known companies that have solid financial structures and good market positions in their industries.

Being a mechanical investor means that you establish criteria that are never violated. These are quantitative requirements that define your universe of potential investments. For example, you will buy only those stocks that sell at a price/earnings ratio below 20, have a dividend yield greater than 1.0% and have positive free cash flow.

While it's easy to be seduced into buying great "stories," having a set of quantitative requirements that make sense based on investment history is the most sensible way to screen your investment universe.

Also, when you purchase a stock, I believe that a set of sell criteria automatically should be put in place. For example, you will sell when the P/E ratio increases to a level higher than the industry median, when quarterly earnings are reported down from last year or when you have held the stock for 12 months.

How can you arrive at a set of quantitative requirements that actually work and are not dependent on someone else's vision of the future? I'm going to recommend a book called What Works On Wall Street by James O'Shaughnessy. When I read this back in the late '90s, I felt like I finally understood where I should be looking to obtain superior investment performance. This is a very statistically oriented book that goes back about 40 years to see how specific investment metrics can be used to achieve superior results.

For example, buying and holding for one year, the top individual strategies over the entire period from 1955 to 1996 were:

• Buying stocks with the lowest price/sales ratio
• Buying stocks with market cap between $25 million and $100 million
• Buying stocks with the lowest price/book value ratio
• Buying stocks with the lowest price/cash flow ratio
• Buying stocks with the highest price change over the previous year

Conversely, the worst strategies were:

• Buying stocks with the highest price/earnings ratio
• Buying stocks with the highest price/book ratio
• Buying stocks with the highest price/cash flow ratio
• Buying stocks with the highest price/sales ratio

And the very worst strategy was buying those stocks that had declined in price the most over the prior 12 months.

O'Shaughnessy combines some of these metrics to arrive at systems that have performed very well over time. In fact, he started a couple of mutual funds that use his approach. They have subsequently been sold to another manager and are now part of Hennessy Funds. I have a small investment in the Cornerstone Growth that has definitely outperformed the market over the last several years.

Another similar approach to the market was outlined by Joel Greenblatt in his book, The Little Book That Beats The Market. He uses another valuation metric, enterprise value/EBIT, along with a return measure to rank stocks, and posts the results on his Web site. He claims a back-tested 35% annual return for his system, but I have not followed it over the last year or so, and I don't know what recent real-life results have been.

The point is that these quantitative approaches use what has worked in the past and will help you take the emotional response out of the investment equation. When someone asks if you own Google (GOOG) , you will smile, knowing that a portfolio full of stocks with valuation characteristics similar to Google is almost certain to disappoint over time. It sure has a great story though.

In the next and final installment, I will get more specific about what investment approach I actually use in my own personal IRA.