Value investors come in all shapes and sizes. Some invest in real estate, others in bonds (generally distressed ones), still others in stocks. Some specialize in certain industries or countries, while others are generalists. But most value investors have two things in common: (1) They seek to buy at a significant discount to intrinsic value (the proverbial dollar bill for 50 cents); and (2) They generally sell once intrinsic value has been reached.

Over the course of my investment career, I've faithfully followed these two tenets, but over time I've tweaked them as I've become more cognizant of what I call "embedded options." By this, I mean something that might cause a stock to appreciate many times over -- in other words, buying a dollar bill for 50 cents, then experiencing the pleasure of selling not at $1, but $2 or more. While not losing money is the single most important factor in long-term investment success, a close second is having a few stocks appreciate many times over.

Rebounding stars
What are the sources of embedded options? What causes stocks to go parabolic? There are many reasons, of course, but the two most common situations I encounter are: (1) A very good business is hit by some sort of disappointment, the stock gets clobbered, but then the business returns to its former economic characteristics; and (2) The company is in a sector with highly irrational investors who are prone to bursts of pessimism (in which no price is too low to sell) and euphoria (in which no price is too high to buy).

Over roughly three years, shares of McDonald's (NYSE:MCD), which I consider one of the world's great businesses, fell from nearly $50 to $12.12 in March of this year. Over approximately the same time period, another fabulous business, Home Depot (NYSE:HD), fell similarly, from $70 to $20.10. From these lows, McDonald's has more than doubled, and Home Depot has nearly done so.

When stocks I own appreciate so quickly, I generally start thinking about selling; the positions have grown larger, while my margin of safety has shrunk. Yet I'm still hanging on to both. Why? Mainly because I never cease to be amazed at how great, well-managed businesses are able to outperform my expectations (which tend to be quite conservative -- one of the keys to value investing). I think McDonald's and Home Depot remain great businesses and that their turnaround efforts will yield rich rewards for at least another year or two. Were these lower-quality businesses, I'd likely have already sold.

Exploiting the herd
I feel a little guilty admitting that I factor the herd into any investment decision, as I fundamentally believe that Mr. Market should be my servant, not my guide. (Warren Buffett described the parable of Mr. Market about halfway through his 1987 annual letter.) But it's been painful to watch ValueClick (NASDAQ:VCLK) soar past $10 (I sold in the $3 range) or E*Trade (NYSE:ET) spike above $11 when I had sold only a few months earlier below $7. I made healthy profits on both stocks, but I left a lot of money on the table as well.

What was my mistake? In part, the businesses performed better than I expected (though I still don't think either is a very good business), so when I thought I was selling at roughly intrinsic value, I was a bit too conservative. But the bigger mistake was failing to realize that these are tech stocks and, therefore, once they returned to favor and their stocks headed higher, all sorts of dimwitted "investors" playing the momentum game would pile in, driving the stocks significantly above intrinsic value. Even without the benefit of hindsight, rather than selling both positions all at once, I should have sold roughly half and then at regular intervals as the stocks rose.

Mediocre businesses in boring industries
It may seem obvious that one should seek to invest in high-quality businesses and, even better, in sectors where investors have been known to pay ridiculously high prices. But in practice, many value investors end up with a portfolio of mediocre businesses in boring industries, typically for two reasons:

First, the stocks of good businesses rarely get cheap (I waited years for Home Depot to get cheap enough to buy). Second, boring industries tend to be easy to understand -- there are few circle of competence issues -- and the future can be predicted with a reasonable degree of accuracy. Both characteristics are lacking in many of the sectors, such as technology, frequented by the most foolish members of the herd.

It's not necessarily bad to own a portfolio of mediocre businesses in boring industries. Investment legends Walter and Edwin Schloss have outperformed the market for decades by owning nothing but ultra-cheap "cigar butts." But ultra-cheap is key. Extensive and painful experience has taught me that if you're going to bottom fish among such situations, you'd better have an extra-large margin of safety because this is the only way you're going to make money.

Consider Universal Stainless & Alloy Products (NASDAQ:USAP), a niche specialty steel manufacturer whose stock I was rash enough to purchase near its peak early last year. I think the company has a sound strategy, management is doing all the right things, and the company is far better positioned than its peers. But as I discussed in a column last year, it's in one of the worst industries imaginable. Management simply has no control over highly cyclical end-user demand, commodity prices, tariffs, imports, and so forth. Thus, even when things go right, Universal Stainless' economic characteristics are decent at best. And investors almost never get excited about steel companies.

So why am I still holding the stock? Because I think that if the economy continues to pick up steam, the company could earn $2-$3 a share. If this happens, I don't need to hope for any investor enthusiasm. Even assuming a low P/E multiple of 10 yields a $20-$30 stock, many times today's price around $8.

Merck and Liberty Media
Two out-of-favor stocks that I believe have both types of embedded options discussed above are Merck (NYSE:MRK) and Liberty Media (NYSE:L). I don't currently own either stock because I'm not yet convinced that the problems ailing these businesses are temporary. However, if you can convince yourself that, say, Merck's legendary R&D department will once again regularly develop blockbuster drugs, then this stock will be a huge winner. The underlying economic characteristics are wonderful and investors often get very excited about pharma stocks.

When deciding whether to buy a stock or when (and how much) to sell, it pays to think carefully about embedded options.

Whitney Tilson is a longtime guest columnist for The Motley Fool. He owned shares of McDonald's, Home Depot, and USAP at press time, though positions may change at any time. Under no circumstances does this information represent a recommendation to buy, sell, or hold any security. Mr. Tilson appreciates your feedback at The Motley Fool is investors writing for investors.