MOST OF THE TIME SPENT on our investments is dedicated to figuring out what to buy. We pore over research, listen to conference calls and scrutinize a company's annual reports to make a decision whether or not to include XYZ in our portfolio. Some we buy and some we pass up.

And because there's a bid for everything under the sun, one can make a bullish case for almost every security on the board. It's a good critical-thinking exercise: Find a stock you absolutely detest and try and figure out why someone would want to buy it.

Of course, for as many reasons as there are to buy a stock, there are quite a few reasons not to as well. And although no system is perfect, I have a few criteria I specifically try to avoid using when allocating assets. These are some of the main reasons why you shouldn't buy a stock. Many are common traps that hinder traders of every size.

Buying a Stock for the Dividend

Since the tech-led bull market of the 1990s fizzled, there has been a sea-change rush back into dividend-paying securities. Even now, people often feel a sense of safety or cushion when buying a dividend-paying stock. Even if they are wrong on the timing, they think "at least I'll get my 4% dividend."

Not exactly. Buying a stock for a dividend is a common and foolhardy error. To start, dividend policies are never set in stone, but are constantly reviewed by a company's management. Experienced investors know it's quite common for a company to lower, delay or cancel its dividend altogether. This was precisely what happened to many investors buying General Motors early last year for the 8%-plus dividend. In an effort to save money, the company cut its payout by more than 50%. So much for the safe 8%.

Even if the dividend isn't cut, the benefits of an attractive payout are quickly erased if a security drops from your purchase price. That was most certainly the case for Big Pharma stocks like Merck or Pfizer from 2001 through last year. Investors who bought the stocks for the 3% to 4% dividends quickly realized that didn't mean much when shares dropped double digits in just a few months' time.

I look for strong securities — period. If a dividend comes along with the deal, so much the better. But because price risk is a given, and dividend yields are subject to change, it is the price that should most influence your purchase decision.

Buying Low-Priced, "Cheap" Stocks

There is something about a low-priced stock that draws us like moths to a flame. We're almost chemically attracted to the fantasy of buying 10,000 shares of a 50-cent stock in the hope it jumps to 75 cents or 90 cents in just a few hours' time. We feel powerful by the allure of buying a large number of shares and fantasize about the notion of a penny stock becoming an established blue chip.

This misguided fantasy is aided by the sleight of hand effect of stock splits. For example, when you look at a long-term chart of well-known stocks such as Microsoft or Intel, you might notice that, 15 or 20 years ago, most appear to have been trading at or near $1 a share. Take Intel: In the early 1990s, Intel appears to have been trading between $1 and $2 a share. This prompts investors to scour low price stocks, looking to find "the next Intel" that's poised to climb from $1 to $20.

Of course, Intel wasn't trading at $1 back then. It appears that way because the company has split its stock five times over the years.

As we've pointed out before, a stock's price, the numerical cost per share, doesn't mean much of anything at all. Using stock splits (or reverse splits), a company can peg its share price just about anywhere it wants. Look at a long-term price chart of JDS Uniphase, for example, and you'll see the stock appears to have been trading at over $1,000 a share in the early 2000s. That's only because a reverse split in 2006 reduced the number of outstanding shares from 1.7 billion to 221 million. Without the sleight of hand, the stock would be trading significantly lower (but holders would have eight times as many shares).

Low-priced stocks aren't necessarily "cheap" and high-priced stocks aren't "expensive." What should influence your trading decision isn't a stock's numerical price, but its price action.

Buying "Good, Quality Companies"

I have to chuckle when I hear an investment advisor assert that he buys only "good, quality companies" for his clients. A quality stock is one that goes up after I buy it. Strange as it sounds, the financial condition of the company at my time of purchase is largely moot.

Why? Well, play the game long enough and you'll begin to see that, despite the financial media's insistence to the contrary, markets don't reflect the present as much as they predict the future. Before a company's fiscal deterioration ever shows up in its 10Qs, you'll almost always see it reflected in a weakening stock price.

Enron is the classic example. Recall how Enron's stock, which had been touted by almost every major business publication and brokerage house, had already fallen sharply before news of the fraud had been exposed. By the time the details finally came public, the stock had already dropped significantly from its highs. Although there's been no suggestion of financial impropriety, a similar scenario has played out over the past few months in many of the subprime mortgage lenders such as Novastar Financial or Accredited Home Lenders. Sure, they've fallen in recent days as delinquencies have begun to rise, but a longer-term perspective shows that many of these names have been weak for the better part of a year. The "news" was just confirmation of what the market had been saying for months.

A stock is not a company. A stock is a piece of paper to be traded. The sooner you are able to separate the two, the sooner I think you'll integrate the objective discipline that successful trading requires.