Bill Miller, manager of Legg Mason Value Trust (LMVTX), has done something that no other fund manager can claim. Since taking over sole management of the fund, he has beaten the market each and every year, for 15 straight years. Is he just lucky like some efficient market supporters would claim? Or does he know something that others don't?

The answer is neither. Markets are not completely efficient. All they do is incorporate the consensus view of investors and use that to arrive at a prevailing market price. The conventional wisdom is collected in an efficient manner, but such wisdom is wrong more than it is right. Every quarter when public companies report their earnings, about 70 percent will either miss or exceed the consensus estimate.

Miller's most recent letter to shareholders outlines his strategy. His ideas will seem reasonable, logical, perhaps even obvious to many. However, when we look across Wall Street we see very few who put them into practice. Which begs the question, why?

People ask me all the time what my philosophy is, my approach to investing. These are the concepts investors must grasp to be good at what they do, before you even begin to look at an industry or a specific company's stock. Many individual investors choose to buy what they know, what they like. Many financial advisers at your big name retail brokerages recommend them as well. That can end very badly, for many of the reasons expressed below. These are excerpts from the letter.

"Unfortunately, when we purchase companies we believe are mispriced, it is often difficult to determine when the market will agree with us and close the discount to intrinsic value. Our goal is to construct portfolios that have the potential to outperform the market over an investment time horizon of 3-5 years without assuming undue risk. If we achieve that goal, we believe we will be doing our job, whether we beat the market each and every year or not."

"The most common error in investing is confusing business fundamentals with investment merit. A company that is doing terrifically well, that has great management and returns on capital, and great products and prospects, may be a terrible investment if the expecations embedded in the current valaution are in excess of those fundamentals. A company with poor business fundamentals, a mediocore management, and indifferent prospects may be a great investment if the market is even more pessimistic about the business than is warranted. The most important question in investing is what is discounted, or put slightly differently, what are the expectations embedded in the valuation?"

"Systematic outperformance requires variant perception: one must believe something different from what the market believes, and one must be right. This usually involves weighting publically available information differently from the market, either as to its magnitude or its duration. More simply, the market is either wrong about how important something is, or wrong about when that something occurs, or both."