The study of probability began in France in the seventeenth century. The mathematician Blaise Pascal was friends with two gamblers who were obsessed with a dice game that was the forerunner to craps. One of them asked Pascal how payment should be distributed if a dice game were interrupted before its completion. Pascal started a
correspondence on the subject with his friend Pierre de Fermat, a judge who studied mathematics as a hobby. This formed the basis for the mathematical study of probability. After a religious experience, Pascal stopped working in mathematics and turned his attention to writing about spiritual matters. When he passed away at age 39, Pascal left behind some notes he had begun to organize to publish as a book. It was eventually published as Pensees (“Thoughts”) which has continuously been in print since 1670.
In his book, Pascal makes a mathematical argument for belief in God. During the Enlightenment, many educated people frowned upon religion. However, Pascal had noticed that these same people would would their money and even their social standing on the rolls of dice, so they should also be willing to wager their souls on the possibility of eternal life. Pascal used the logic of mathematics and an analogy to the games of chance to argue that a rational person should believe in God.
Pascal’s Wager considers the risks involved in belief and unbelief. Everyone makes this choice as they actively believe in God or decline active belief. The declination can either be a belief that there is no God (atheism) or an unwillingness to decide (agnosticism). Belief and unbelief can be viewed as two sides of a wager. On one side, suppose someone does not believe in God. The potential long-term consequence if God does exist is eternal damnation. If there is no God, nothing is lost. On the other side, if someone chooses to believe in God, the potential long-term loss if there is no God is nothing. The potential gain in belief in God, if God does exist, however, is eternal life. Pascal cast this as a comparison of expected values. However, the real difference in the outcomes is not the likelihood but the consequences. Even if someone believes that the probability of the existence of a supreme being
is low, say 1%, the real risks are the potential long-term consequences.
Even if the probability of existence is less than 0.0001%, the consequences far outweigh any consideration of probability. The purpose of this example is to stress that in assessing risk we need to let go of our intrinsic need to be correct and consider consequences. Risk has two dimensions – the likelihood of occurrence of a bad event, and the consequence. There is an innate tendency to focus on the likelihood of occurrence. However, as this example shows, consequence should be considered as well. Project risk management also suffers from this defect. The focus on S-curves and confidence levels leads to excluding the consideration of extreme consequences, which, while unlikely, should be incorporated into a realistic risk analysis.
I go into more detail on this topic in my forthcoming book Solving for Project Risk Management: Understanding the Critical Role of Uncertainty in Project Management. It will be published on November 3rd (election day in the U.S.), but you can pre-order now from Amazon or Barnes and Noble. You can also read Chapter 1 of the book for free here, and read my technical conference papers on risk here.