Decision theory serves as a platform for helping businesses make wise decisions.
How would you like to be charged with making decisions about this economic crisis or the influenza pandemic? We all know that decisions surrounding such extremely serious and impactful issues can have catastrophic results if not made based on sufficient knowledge. Millions of people rely on such powerful decision-makers to be fully informed and make wise decisions.
Fortunately, most people are charged with making more everyday type decisions. Even so, business leaders and managers make important decisions daily, and obviously, they want to make wise decisions. In my first HRTools.com Insight titled, “Two Toolbox Essentials: Decision Theory and Knowledge Management,” I introduced the topic of decision theory. What is this concept and how does it apply to making decisions?
In order to make wise decisions, business decision-makers must be willing to take time to apply what they know. To help explain, and if you were participating in my class, here is what you might hear.
- First of all, we have what I call the “decision space,” and I’ll share an example. Most people are familiar with the bell curve (if not, see this information and diagram at Wikipedia.com).
In this example, I am working in the field of finance, and I have 1,000 stock issues. I track their returns. I find that roughly 650 to 700 of these stock issues fall within a standard deviation above or below the main return for those stocks.
This is a probabilistic estimate. As we observe things in nature, we know they’re distributed in that fashion, or they’ll tend to fall that way.
So this example represents one way of looking at a decision. For instance, then, if I have a return that’s very, very far outside the bell curve—an outlier, as we call it—then what happened?
- A second example relates to competitive modeling or competitive analysis. If you saw the movie, “A Beautiful Mind,” you saw the classic two-player game with a matrix. The character, John Nash, is a mathematical genius who comes up with the idea that if you have two players, they will tend to gravitate to a situation where one tries to minimize the other’s maximum gain. They then reach what is called a “steady state.” This game analogy represents competitive modeling/analysis.
- A third example is what we call a “collision matrix,” which is really a triangle with Player A, Player B and Player C and their respective competitors. Eventually two of those players will team up. They may create, for instance, a strategic alliance.
They might merge, or one may become part of the other; and they will take these actions in order to deal with the third party. Whoever has the most to trade will usually be the one who will come out; who will be the winner. So it’s another way to visualize using what you know to make a decision.
Probability Vs Likelihood in Decision Theory
Next, it may be helpful to touch on the differences between the terms “probability” and “likelihood.”
Back in the 1960s, some people tracked discovery wells out in the ocean. They might go out into the middle of the North Atlantic or the Gulf of Mexico to drill an oil well. Back then, you might hit oil about one-eighth of the time. So this would mean a probability exists that you would hit oil one out of eight times, or five out of 40, or two out of 16, and so on.
Now, if I hit a well in a certain area, and then I returned to that same area, there would be about a 75 percent chance that I would hit a second, third or fourth well, etc.
So if I found a well in a certain area, the basis for this reoccurring is because there is a likelihood that more wells would be found in that same area.
Bell Curve + Expected Value = Decision Tree
In closing, let’s return to the bell curve discussion—that bell curve is a continuum of a whole set of data. Decision theorists understand that when you are looking at a range of potential actions—some of those can be really good, some could be really bad and some can be what you kind of expected to get.
That said, now the key part of this discussion turns to “expected value.” For instance, if I raise the price of my product, what will happen to its value? What will I get? Will I kill off the chance for some sales? Will I make more profits?
And there’s another concept: the expected value of perfect information, whereby I take a model and call it a “decision tree,” which is a diagram used to illustrate possible outcomes. Here, I work through the possible outcomes of a situation. I might project something such as, “OK, is it worth it for me to spend $500,000 to do a study, if there is a possibility that I may lose $50 million?”
It doesn’t take a genius to say, “I better not spend the $500,000.”
Again, decision theory is used to help business decision-makers apply their knowledge to make wise decisions. Wiser decisions usually translate to higher performance levels and results.