Logical Thinking · Lesson 4 of 4

Enough probability to stop being fooled

A good decision and a good outcome are different things. Judging only outcomes teaches the wrong lesson.

6 min read · Free — no account needed to read

You do not need mathematics to reason better about chance. You need three habits. First, ask how common the thing is to begin with — a test that is 99% accurate for a condition one person in ten thousand has will still be wrong most times it says yes.

Second, be suspicious of confident stories built on few cases. Small numbers produce extreme results easily, which is why the best and worst performers are so often the smallest groups.

Third, separate the quality of a decision from the quality of its outcome. A sound decision can end badly and a reckless one can end well. If you judge only by outcomes, you will learn the wrong lesson roughly as often as the right one.

See it happen

You skipped insurance on a trip to save 60. Nothing went wrong.

The read that arrives on its own

Good call — that was 60 saved, and nothing happened. You skip it next time too.

The same situation, with this idea

The outcome was good. Was the reasoning? What was I exposed to, how likely was it, and what would it have cost me if it had happened? If the answer is "more than I could absorb", the decision was poor and the outcome was kind.

What changed

Good decisions and good outcomes are different things under uncertainty. Grading yourself on outcomes teaches you a rule that will eventually be expensive.

The takeaway

A good decision and a good outcome are different things. Judging only outcomes teaches the wrong lesson.

Easy to take the wrong way

Using this in reverse, to excuse any decision that went wrong. The test is whether the reasoning was available at the time, not whether you can construct it afterwards.

Try this

Recall a decision that ended badly. Ask honestly whether it was a bad decision or bad luck.