Classical economic theory rests on a foundation of highly questionable assumptions. The most pernicious of them is that humans behave rationally.

Thanks to the emergence of a relatively new field of study known as behavior economics, however, many of these assumptions have been turned on their head. Dan Ariely's book Predictably Irrational presents a case in point, arguing that humans are not only irrational but, as the title suggests, predictably so.

The significance of this for investors couldn't be more dramatic. It changes not only how we perceive the actions of others, but also how we understand our own. In this article, the second in a series, I discuss how to avoid one predictably irrational mistake that I suspect every investor has made and many continue to.

The irrationality of anchoring
Anchoring describes our tendency to rely on one trait or piece of information when making decisions. Two of its earliest theorizers were Amos Tversky and Daniel Kahneman, the latter of whom received the 2002 Nobel Prize in Economics for his work in prospect theory, which describes decisions between alternatives that involve risk.

In an early study on the effect, Tversky and Kahneman asked two groups of people to guess the percentage of African countries that were members of the United Nations. When asked if it was more or less than 10%, the first group guessed 25% on average. When asked if it was more or less than 65%, however, the second group guessed 45% on average. And where did the original 10% and 25% anchors come from? The researchers spun something akin to a wheel of fortune in each participant's presence, thus demonstrating that anchors are not only powerful, but potentially random as well.

The randomness of anchoring, which came to be known as arbitrary coherence, was studied further by two researchers at the Massachusetts Institute of Technology. They began the study by asking a group of 55 MBA students at MIT's prestigious Sloan School of Management to write down the last two digits of their Social Security number. The students were then asked whether they would hypothetically pay more or less than that amount for a number of products, including an ostensibly valuable bottle of wine. Following this, the students were asked to actually bid on the items in an auction.

The question the researchers sought to answer was: Did the digits from the Social Security numbers serve as anchors? And remarkably they did, for the students with the highest-ending Social Security numbers (from 80 to 99) bid highest, while those with the lowest-ending numbers (01 to 20) bid lowest. Indeed, when all was said and done, the students with Social Security numbers ending in the upper 20% placed bids that were 216% to 346% higher than those of the students with Social Security numbers ending in the lowest 20%.

The relevance to investing
While the significance of anchoring to investing may not be obvious immediately, you can be sure its insidious presence lurks beneath many a poor investment decision. How many times have you heard somebody say that a particular stock is cheap because it's trading below its 52-week high? Or that it's expensive because it's above its 52-week low? Or how about when a person chooses to invest in one stock rather than another based on their respective share prices alone -- stocks like Berkshire Hathaway (NYSE: BRK-A) (NYSE: BRK-B), whose A-shares trade for over $100,000 a share, aside. Indeed, in each of these cases, investors are not only anchoring, but are under the influence of arbitrary coherence.

One of the greatest stories about the cost of anchoring involves what Warren Buffett claims to have been his biggest investment mistake ever at Berkshire. At some point in the early 1980s, Buffett wanted to increase his position in Wal-Mart (NYSE: WMT), which at the time was still a fledging retailer compared to what it is today. Seeing the price of the company rise (purportedly by an eighth of a point), however, he decided to wait until it had fallen again before buying the shares. Needless to say, it never regained his target price and led Buffett to claim that his "thumb-sucking" cost Berkshire Hathaway upward of $8 billion in lost profit.

Speaking from my own experience, I'd go so far as to say that any stock chart you look at is a figurative graveyard of anchoring errors. Take Netflix (Nasdaq: NFLX) as an example. It was on top of the world six months ago. At the beginning of June, it was trading for over $270 a share, almost tripling in one year. Later that month, it fell to below $250 -- an apparent bargain for people who were following the stock. Yet, the stock now trades for a measly $71.

The same thing can be said about SodaStream International (Nasdaq: SODA), the maker of the homemade soda machines you see displayed at virtually every retail store. Indeed, had you purchased it based on relative dips in the stock's movements at any point in July and held it until today, you'd be out over half your investment, as the stock went from nearly $80 at the beginning of August to only $31 today. And that's despite the fact that it's been a hot-selling gift this holiday season.

So how do you avoid this?
The key to avoiding anchoring's sinister effects is to rededicate yourself to the fundamentals. Look at value, not price. Ask yourself whether you like the management and believe that the company, not necessarily the stock, is going in the right direction. Pick up Benjamin Graham's The Intelligent Investor or Peter Lynch's One Up On Wall Street and learn that the trick to great investing is thinking like a business owner and not a speculator -- that's what Vegas is for. And finally, look for advice from the best in the business. One place to start is a free report our top analysts drafted about specific stocks the smartest investors are buying. It explains what sector Buffett and others are interested in, and tells why the Oracle of Omaha would be interested in Bank of Hawaii (NYSE: BOH), among others, if he were a small investor. To access this free report while it's still available, click here now.