Happiness, Health, Wisdom and Wealth ^™

It's hard to exaggerate how common and how serious these failures to deal with probabilities correctly are. Here's a real life example. NASA administrators projected that the probability failure of the space shuttle was 1 in 100,000. This estimate was based in large part on heuristics (a fancy word that means what has happened in the past). There had been lots of space flights and no accidents so it must be pretty safe, right? Wrong. When probability experts worked with engineers to develop probabilities based on the known probability of the failure of the hundreds of critical parts that go into a mission; they came up with a probability of 1 in 35. Now, this is not a small difference in numbers. Compare 35 to 100,000; 100,000 is almost 3000 times larger. That meant that the danger in a shuttle mission was 3000 times greater than that predicted by NASA administrators, and they ARE rocket scientists. So far there have been about 130 missions and 2 of those ended in disaster. That's 1 out of 65 that have failed. That compares well with the calculation of 1 in 35. It clearly indicates that 1 in 100,000 was ridiculous.

How can such errors occur. Well, we all make mistakes like this every day. Every time we do something and it works out OK, we begin to think it's 'safe'. The more it safely happens the safer we tend to think it is. That reasoning is fine as far as it goes but there's more to it than that. If something happens safely 10 times that means that the odds are likely less than 1 out of 11 that an occurrence will end badly. The next time it could be a disaster. No matter how many times something has happened safely, the next time it may not. We can't predict probabilities correctly from history unless it has happened at least once. They say nothing is certain but death and taxes. Well, they aren't certain either. There are people who have never paid taxes so there is a probability that a baby born today will never do so. No one has ever lived forever but since about 100 billion people have ever been born; the best we can say are that the odds are greater than about 100 billion to 1 against someone living forever. That's not the same certain.

Let's say you flip a coin 5 times in a row and it comes up heads every time. One of two reactions is likely; you may believe that the coin is rigged and will continue to come up heads or you may believe that the odds against the next flip being heads are huge. The truth is that the chances on the next flip being heads are 50-50. So strong are our biases that many of you reading this are saying, 'you're crazy'. And while that may be true, I'm right about this. The coin can only come up 2 ways and those two events are equally likely. The coin has no memory of what has happened in the past, right? There is only this one flip and there is a 1 out of 2 chance it will be heads. Now, here's the tricky bit. The odds of flipping six heads in a row are 1 in 2x2x2x2x2x2 or 2 to the sixth power or 64. The odds of 6 in a row are 1 in 64 but the odds this flip will be heads is 1 in 2. Seems weird, huh? The reason for the difference is that the first calculations is about a combination of events and the second is about a single event. The probabilities for the combination do not affect the probabilities of the single event. I mean think about it, how could it, it would mean that the behavior of the coin would have to change based on what had gone on before. Do you believe that coins can alter their behavior on purpose? No, of course not.

Now this is not intended to be a course in probability (if you'd like to study probability in a fun informal way try the book Struck by Lightning: The Curious World of Probabilities by Jeffrey S. Rosenthal). This is intended to be a discussion of not how to calculate probabilities but how to think about how they affect our lives. That is to understand that every decision you make is based on your evaluation of the probabilities. If you act in accordance with the probabilities you will lose a lot but you'll win more than you lose. If you make a decision based on a correct evaluation of the probabilities and lose you did not make a bad decision; you made a wise decision that just didn't pan out this time. Make sure you take a minute and think about this. It can go a long way toward eliminating guilt in your life. If you correctly evaluate the probabilities and base you decision on those probabilities you are NEVER wrong.

However, there is another aspect of probabilities we need to discuss and that's the cost-benefit ratio. Most of the decisions we face in life involve an unequal balance of cost and benefit. In our umbrella example, the cost (getting wet) and the benefit (not having to schlep your umbrella) may vary from person to person and situation to situation. If you hate getting wet 9 times more than carrying your umbrella all day then if you carry your umbrella on all days where the chance of rain is greater than 1/10 (because 10 is 9 + 1) your satisfaction will 'break even'. Huh? Let's look at a really high number of days, say 1000. Out of those thousand days that had a 10% chance of rain it would have rained 100 of them (10% of 1000 is 100). On those days you got wet and were 9 times more unhappy than if you had carried you umbrella. So your dissatisfaction rate is 9 times 100 or 900. On the other 900 days it rained and your satisfaction rate is 1 times 900 or 900. So your total satisfaction is equal. If you don't like getting wet why not carry an umbrella every day? Well let's see how that works out. You carry and umbrella everyone of the 1000 days. So your cost is the amount you hate carrying an umbrella (1)times the number of times you carried it (1000) or 1000 dissatisfaction points. The benefit is that on the 100 days it rained you didn't get wet so on each of those days you get 9 satisfaction points for a total of 900. So you end up with 1000 dissatisfaction points and 900 satisfaction points. The amount of dissatisfaction 'beats' the amount of satisfaction so in the long run you lose. Now, I'm not suggesting that you should or even could make these kinds of calculations for every decision. But it is critical to understand the concept because most people are off by HUGE amounts; amounts so large that an understanding of the principle would change the decision without having to do a calculation. For example, what if there's a 10% chance of rain and you've got an important job interview. Many people would say, 'well, there's only a 10% chance of rain so I don't need an umbrella'. Wrong! If there were only a 1% chance of rain you should still carry an umbrella because benefit of showing up to your job interview not looking like a drowned rat is FAR higher than the cost of carrying the umbrella. We don't need to know the exact numbers. We don't need to do a precise calculation we only need to understand the principle that a small cost (carrying an umbrella) we can reap a large benefit (showing up looking great).

How does all of this affect our lives? In two seriously important ways. Learning to make better decisions and not feeling guilty when it doesn't work out. As you begin to use this probability principle your decision making will improve. As your decision making improves your life will improve. If you want to improve your life you MUST improve your decision making process. As you gain a greater understanding of this principle you will find it easier and easier to forgive yourself for making the right choices. Let's do a quick exercise. Write down the ten things you most feel guilty about in your life. Now, one at a time think back to how you made that decision. Evaluate your decision not based on the outcome but on the process. Did you follow the course of action that the probabilities suggested was wisest? If you did you have nothing to feel guilty about. Sort the list into two categories, one in which you made the decision the probabilities favored and one in which you didn't. Use only the information you had at the time you made the decision. The list of those that the probabilities favored you should be proud of; you made the correct decision. In fact, you should be proud of any decision that is made wisely, regardless of the outcome. Now, let's look at the other category. These were mistakes. But again, ignore the outcome. Focus on where the decision making process went astray and see if you can devise a plan to prevent it happening again. That's how we learn from our mistakes.

You're probably reading this on a computer, most likely your computer, when did you last do a back up? Let's do some estimates. We aren't going to try and get the numbers exact. The average life expectancy of a hard drive is about five years. That's almost 2000 days. That means there's about a 1 in 2000 chance your hard drive will fail today. If the data on your computer since you last backed it up is more than 2000 times more valuable to you than the cost of a backup; you should do a backup. This brings up questions like, 'Can I reduce the cost of a backup by getting better software or hardware?' and 'Should all files get equal backup protection?'. I can't answer for anyone else but some files I backup daily, some files I have five backup copies of in different places, some files I don't backup at all. That didn't happen randomly; it's the result of a decision making process.

When discussing the value of fitness people occasionally ask me about Jim Fixx. Fixx was a fitness advocate and marathon runner who died at 52 of a heart attack. They use this as evidence that exercise doesn't extend your life. The fallacy in their 'evidence' is that Fixx's fitness increased the probability of a longer life; it didn't guarantee it. There are no guarantees in life, all there are is probabilities. Fixx made a wise decision. He 'played' the probabilities. It just didn't turn out that time. That is the nature of life. All you can do is take your best shot, that is, play the probabilities. It will turn out how it turns out but you will have done the best anyone can do no matter how it turns out.