Good Decisions vs. Luck

Today I’d like to discuss something profoundly important regarding decision making.

Ninety-nine percent of people do not grasp these concepts, as they are subtle and nuanced.

My question is this: What makes a decision a good one?

It’s a simple question.

Most people say a good decision is one where there is a good outcome.

When good information exists about a situation, that’s often true. 

The problem is that in life (and in business), many decisions are made with imperfect information, where many important and relevant things are unknown. 

This has been the traditional Achilles’ heel and weak spot of the MBB consulting approach. Analytically-derived decision making (which is what MBB is known for) is premised on there being data to analyze. When data isn’t available or doesn’t exist, the value of the analytical approach drops considerably.

For example, urban legend has it that all the top consulting firms concluded to electronics manufacturers that the mobile phone market would never exist… they supposedly made this conclusion in the 1970s.

This is because in 1970, no concrete data existed to prove that consumers would value and ultimately demand mobile phones. And data did exist to strongly suggest there would be very low demand for such devices (which incidentally turned out to be true for the subsequent 20 years).

So this brings me back to my original question.

Under high informational uncertainty, what makes a decision a good one? Is it a good outcome? Or something else?

If you’re dying of some disease with a 90% chance of dying within 12 months, but a surgery could save your life, would you take the surgery? What if the surgery had a 50% chance of killing you? What if the surgery was only possible if you did it in the next 30 days?

The only thing more difficult than making a decision under uncertainty is making such a decision when stakes are extremely high.

Let’s say you weigh the pros and cons of the life and death situation I described above and decide to undergo the surgery… and then you die. Is that a bad decision or merely a good decision with a bad outcome?

THAT question is a profoundly important one to grasp.

You won’t have many such decisions to make in your life. But of the decisions you make that have a major impact on your life, many will tend to be ones like these — highly uncertain, with high stakes.

To tackle decisions like these, it’s useful to borrow a concept from the field of statistics — the concept of expected value.

To illustrate this profoundly useful concept in a simple way, let me share a simple example.

Let’s say I flip a coin.

Fifty percent of the time it lands with side A facing up, and the other fifty percent of the time it lands with side B up.

Statistically speaking, there is equal probability of getting a side A outcome versus a side B outcome.

Now let’s say you play a game with a friend.

Every time the coin is flipped and side A is face up, you earn $2. However, every time side B is face up, you lose $1.

Should you play this game?

• $2 gain if you win.
• $1 loss if you lose.

• 50% chance of winning
• 50% change of losing

Intuitively, most people (including most kids) recognize playing the game as a “good decision.”

The “expected value” of this decision is as follows:

Expected value of outcome = (50% chance of winning x $2 benefit for winning) – (50% chance of losing x $1 cost for losing) 

Expected value of outcome = $1 – $0.50 = $0.50 

In other words, each time you play you would expect to earn $0.50. If you play two times, you’d expect to make $0.50 x 2 = $1.

One time you win, and you make $2. Next time you lose, and you pay $1. Your “profit” after two plays is $1.00 (or $0.50 for each attempt)

Is this a good decision?

What happens if you play this game, and you lose 10 times in a row?

Did you make a good decision or a bad decision?

Would you play an 11th time? Would THAT be a good decision?

Does your answer change if you play the game 100 times and lose 100 times in a row?

Was THAT a good decision?

Would you play the game a 101st time? Would THAT be a good decision?

Statistically speaking, IF you can afford to play the game many times and sustain “unlucky” losses, it WAS a good decision to play the game 100 times and is STILL a good decision to keep playing the game.

With uncertainty, the quality of a decision is not based on outcome, but on weighing the expected value of your outcomes.

If you drive your car blindfolded down the freeway at 100 miles per hour (160 km per hour) for 20 minutes and you did not die, was that a good decision or merely a lucky one?

If you eat healthy, exercise regularly and get good sleep but get a heart attack anyway, was that a bad decision or a good decision with a bad outcome?

If you prepare for case interviews thoroughly but don’t get a job offer, was that a bad decision or a good decision with a bad outcome?

If you do NOT prepare for case interviews, but DO get a job offer, was it a good decision to not prepare, or a bad decision that happened to have a good outcome?
It’s worth thinking about.

Questions like these cross multiple fields, including statistics, decision theory, and life/business planning. It’s something I have a personal interest in, for a few simple and practical reasons. Looking back on my own life, much of where I am today (the good, the bad and the ugly) seems disproportionally determined by a few pivotal decisions.

I call this “making the difficult decisions.”

It is PROFOUNDLY important to do this well.

I’ve made some great “difficult decisions,” and some colossally stupid ones — and paid dearly for them. I’ve learned a few things along the way — one of which I just shared with you.

Sometimes good decisions have bad outcomes, and sometimes bad decision have good outcomes. Do NOT get the two confused.

In short, be smart enough to realize when you just got “lucky.”

I tend to write and teach about the topics that I find interesting that others do too. I have many esoteric interests, and “how to make difficult decisions” is one of them.

If you share this interest and would like to learn more, vote with your preference by filling out the form below.

In the meanwhile, did you make a good decision today, or did you happen to get lucky/unlucky? Give it some thought.