The mainstream finance industry takes the view that stock markets of advanced countries are efficient (see Efficient Market Hypothesis), and as such it is impossible to consistently outperform the market except through luck. Though holes have been poked in this theory (including the January Effect, the outperformance of low P/E and low P/B stocks, and other anomalies), most in the finance industry are of the opinion that such anomalies are the result of data mining bias, or else disappear as they become known (since many will attempt to profit from them, thereby driving returns to zero).
However, a new field has emerged that helps explain certain stock market phenomena. Researchers of "Behavioral Finance" have suggested the existence of certain deeply ingrained biases in investors that help explain anomalies which have resulted in opportunities for abnormal rates of return (which have gone unexplained by EMH).
For one thing, the findings of Behavioral Finance experts suggest analysts are overconfident in their projections of growth companies, resulting in, for the most part, valuations far and above what they should be for certain high-flying firms. Furthermore, researchers have identified a phenomenon whereby investors ignore bad news, causing them to hold losers for too long (or in some cases acquiring more shares, in an effort to average down costs).
While completely ignoring psychological biases when investing may be impossible, by understanding our inherent biases we put ourselves in a better position to base our decisions on fundamentals rather than emotions. For a more detailed explanation of the various forces and factors comprising Behavioral Finance, here is a link to a useful Wikipedia article.