Newsletter Sign-up

Do you believe in luck?

I am guessing that most of us have done at least one, if not all, of these things as points in our lives. If you don't believe in luck yet, then consider this: In order for you to be sitting wherever you are reading this column, you have had to survive (insert your age) years and any of the dangers you have encountered in your life ... preceded by all of the dangers that your parents, their parents and the rest of your ancestors encountered back to the beginning of time, including world wars, plagues and diseases and primitive medical techniques.

In short, we are all lucky to be here ... or are we?

When you think about it, it is pretty amazing how ingrained the concept of luck is in our lives, yet there is not a lot of clarity on what luck actually is and how to recognize it when you see it. So what is luck, and what does it have to do with strategy and risk in business?

Merriam-Webster Dictionary defines "luck" as "a force that brings good fortune or adversity." Does this mysterious force actually exist in business, or it is just a perception? If it does (or doesn't), how should we account for it as we set strategy and manage risk?

When asked if the force of luck exists in business, the analytical side of my brain says, "Absolutely not." Let's take a simple, hypothetical example where there are two nearly identical companies in a market, each with completely substitutable and identically priced products and 50 percent market share. Everyone agrees that there is a 50/50 probability than a customer will purchase Company Alpha's product over Company Beta's product, and vice versa.

On March 17, Paddy Murphy, the new head of sales for Company Alpha, is on his way to his first meeting with a customer when he trips and stumbles, nearly dropping the golden horseshoe in his pocket and almost scratching his sacred Notre Dame class ring. But he comes across a four-leaf clover growing in between a crack in the sidewalk What luck! Paddy is on top of the world and could not be happier. Low and behold, the client decides to purchase from Company Alpha, instead of Company Beta. Paddy could not be happier at his good luck and decides to make a point of tripping on his way into every customer meeting in the hopes that he will come across such luck again.

As much as I would like to keep from bursting Paddy's bubble, we can use a simple decision tree to show that, despite the positive outcome above with Paddy's customer, luck did not change the odds of the customer buying from Alpha or Beta.

According to Wikipedia, there are approximately 10,000 three-leaf clovers for every four-leaf clover. So in the top tree here, let's set the odds of Paddy finding a four-leaf clover at 1 in 10,000 (or 0.01 percent). Additionally, the odds of the customer buying from Alpha or Beta is 50/50 in both the case when luck exists (e.g., Paddy finds a four-leaf clover) and when luck does not exist (that is of course, unless you believe that when Paddy tripped and found the four-leaf clover, something like the infamous "butterfly effect" kicked in and the universe changed, altering the odds of the customer buying from Alpha in Paddy's favor).

The bottom tree here shows the results when we "flip" the tree using a Bayesian revision. Again, there is a 50/50 probability of the customer buying from Alpha or Beta. Additionally, given that the customer buys from Alpha, the probability that Paddy found a four-leaf clover is still 0.01 percent.

This is a long way of saying there is no influence of luck on the customer's decision. Again, if you do believe in luck, then you would have to believe that it alters the odds of the customer buying decision when it exists versus when it does not.

Some may ask, "Wasn't the fact that Paddy found the four-leaf clover and the customer chose Alpha an indication of luck, because the odds of this happening are so small (0.005 percent or 1 in 20,000)?"

On the surface, it appears that they may have a point. However, all it really means is that Paddy happened to find a very rare mutation of a regular three-leaf clover (1 in 10,000 chance) and the customer flipped a coin and purchased from Alpha (1 in 2 chance)--two otherwise unrelated events resulting in 1 in 20,000 when multiplied together.

Yet if luck does not actually exist in business, what's the harm in people still believing in the existence of luck? After all, everyone is entitled to their own beliefs, right?

Unfortunately, there is potential harm. Anytime there is a gap between the actual probabilities of an event and someone's perceptions of the probabilities of an event, there is a chance that the person will make a decision based on their perception of the probabilities. For example, Paddy's decision to intentionally trip before every customer meeting is based on his perception that, one, tripping increases his chances of finding a four-leaf clover and, two, finding a four-leaf clover increases his chances of bagging a sale.

Sure, he may be correct that tripping increases his chances of finding a four-leaf clover; however, even if he found one every time he tripped, the customer would still only purchase from Alpha 50 percent of the time.

Unfortunately, by intentionally tripping and trying to kiss the sidewalk before every meeting, Paddy might actually do something that causes harm to himself. Therefore, we can sum up Paddy's decision to trip as having no upside (neutral at best) and a fair amount of downside.

Paddy Murphy's trip is a highly stylized illustration, but gaps between actual and perceived probabilities are very real and unavoidable in both business and our lives. As we help make strategy and risk management decisions in our own organizations and lives, it is essential to understand:

-- if there could be a gap between the actual and perceived probabilities of a key event. If so, time invested in better understanding or further studying these gaps will often increase the quality of our decisions.

-- if we are potentially ascribing a causal relationship between factors that may not be related at all (or vice versa). If so, time invested in better understanding or further studying the nature and degree of these relationships will also increase the quality of our decisions.

-- if simple decision trees and influence diagrams are a couple of techniques that one can apply to any decision, no matter how basic or complex, to improve one's understanding of a situation and to ultimately improve their ability to make high-quality decisions.

Every March, my wife and I make a concerted effort to each have at least one green beer at a local pub and a tasty Shamrock Shake from McDonald's ... for good luck, of course. So far, it seems to have been a pretty successful formula for us, so it is hard for me not to recommend it to others. Therefore, please grab a green beer or Shamrock Shake and join me in thanking each of our parents and many generations of ancestors for surviving and enduring countless past dangers, making it possible for us to be sitting here now. Plus, we are going to need all the luck we can get to successfully navigate our own dangers to make our future generations possible.

DAVID M. WONG is director of enterprise risk management at CME Group, the world's largest and most diverse derivatives exchange.

March 16, 2011

Copyright 2011© LRP Publications