"One of the advantages of a fellow like Buffett is that he automatically thinks in terms of decision trees." -- Charlie Munger

Decision trees, decision trees, decision trees. I believe one of the biggest errors (myself included) that investors make is attributing results to decisions, rather than to decision trees.

Let me give you an example. Suppose someone offered you a bet: If you win, you get a billion dollars. If you lose, you have to do the Macarena in a crowded mall. The odds of winning are 75%. Would you take this? Heck, yes, you would! The downside is minimal, and the upside is astronomical.

Now let's tweak that a little bit. Instead of doing the Macarena in a crowded mall, you have to run blindfolded through freeway traffic. Now the decision is not so clear, because the decision tree (if I win, I'm a billionaire; if I lose, I have a high probability of dying or suffering debilitating injury) has a grim downside.

Heyyyy, Macarena!
As investors, we spend too much time focusing on only the most likely outcome, and too little time on the different scenarios that might occur and their corresponding probabilities. In the two scenarios above, the most likely outcome is overwhelmingly positive -- you win a billion dollars. However, the decision trees are drastically different, and thinking through the probabilities of the second scenario (with a 25% chance of possibly dying) leads you to make the correct choice to reject the bet.

In investing, the future for even the simplest of companies has countless possibilities and probabilities. As Buffett says, never lose money, so let's focus only on downside risk -- in a decision tree, this would be the arrow that goes down. Assessing our downside risks involves asking three simple questions.

1. What is the worst-case scenario that could develop?
A while back, some very smart people like Marty Whitman and Eddie Lampert figured out that Kmart's investment decision tree was heavily tilted in their favor. For example, the company was already in bankruptcy, but they realized they could still recover billions of dollars worth of real estate, so they would make money even in the worst-case scenario. However, not only did they sell a lot of real estate, but Lampert was also able to turn the business around, and the company, now Sears Holding (NASDAQ:SHLD), has gone up about 1,300% in the past four years.

However, not all worst-case scenarios are quite so rosy. Industry juggernaut USG (NYSE:USG) hit a perfect storm and went bankrupt in 2001, and Time Warner's (NYSE:TWX) AOL, once an Internet titan, now offers its services for free. Worst-case scenarios can and do unfold, and investors need to strongly consider the consequences if they do. (And although this is obviously said with perfect hindsight, perhaps this would have helped Time Warner sidestep the AOL merger.)

2. What is the probability of the worst-case scenario?
For Coca-Cola (NYSE:KO), I'd imagine the worst-case scenario is that people stop drinking Coca-Cola products. If this happened, the business would be worth its liquidation value. However, the probability of that happening is so low that we can pretty much ignore it. Coke has been the dominant carbonated drink for ages.

It can be really hard to pin down an answer on this question. For instance, what's the probability that ethanol will be a viable alternative energy source in five years? What's the probability that the price of a barrel of oil declines to \$40? What's the probability that Google will still have the dominant search engine in five years? Unlike dice, where we know the exact probabilities, businesses have too many factors. Often, the best we can do is guess. The trick is to only make guesses where you have enough knowledge to get somewhere in the ballpark.

3. What's the company worth in the worst-case scenario?
Sometimes, as with Kmart in bankruptcy, a company has substantial value even in the worst-case scenario. Other times, as with a one-drug biotech company, the company is virtually worthless if the drug doesn't work. A quick and dirty way to value a company in its worst-case scenario is the liquidation value of its assets, or the lowest price one could reasonably expect a competitor to pay to buy the company.

For companies with nearly impenetrable moats like Coca-Cola or American Express (NYSE:AXP), the worst-case scenario might be flat growth -- so my worst-case valuation would be ascribing a "normal" multiple to a high-quality company.

It's darkest before dawn
If you invest only in companies where you're paying less than or equal to the company's worth in the worst-case scenario, then it's pretty hard to lose money. You can either do this by paying less than liquidation value for a deeply distressed company, or by paying a mediocre multiple for a great company. This is a methodology that takes time to understand and apply correctly -- as well as an enormous amount of discipline and patience -- but I believe it's the best way to invest your hard-earned money.

Related Foolishness:

Coca-Cola and USG are Motley Fool Inside Value recommendations. Time Warner is a Stock Advisor recommendation. Try any one of our investing services free for 30 days.

Fool contributor Emil Lee is an analyst and a disciple of value investing. He doesn't own shares in any of the companies mentioned above. Emil appreciates your comments, concerns, and complaints. The Motley Fool has a disclosure policy.