Hindsight makes every crash look obvious

What follows is the working version of hindsight bias: the decisions in the order you actually meet them, with the reasoning attached.

Knowing the answer changes the question

Once an outcome is known, the path leading to it appears far more inevitable than it did beforehand. The effect has been demonstrated repeatedly and applies to experts as well as to everyone else.

It works by reorganising memory around the outcome, so the signals that pointed elsewhere fade. This is why every crash has an obvious cause afterwards and none has one at the time.

Everyone remembers having known

People genuinely recall having been more concerned than they were, without any intent to deceive. Commentators who made the correct call are remembered and the far larger number who did not are forgotten. The surviving record therefore suggests a level of foresight that did not exist.

In practice, this creates a false impression that prediction is more feasible than it is.

The cost to your decisions

If past crashes look predictable, the next one seems predictable too, which invites market timing. It also produces harsh self-assessment: regret for not having acted on signals that were not visible at the time. That regret drives over-correction, typically holding excessive caution for years afterwards.

Both effects follow from a memory that has been quietly rewritten.

Explanations are always available

Any market movement can be explained after the fact from the enormous pool of concurrent events. The abundance of plausible explanations should reduce confidence in any single one rather than increase it. A daily commentary that always has a reason is producing narrative rather than analysis.

Noticing that the reason arrives after the price is a useful habit.

Keep a contemporaneous record

Writing down what you expected, when, and how confident you were is the only reliable protection. Reading old entries after an event is a genuinely humbling exercise and more useful than any amount of theory. It also improves future decisions by showing which of your concerns actually mattered.

On an ordinary week, the record needs to be written before the outcome, which is the only difficult part.

Judge decisions by process

A decision made with good reasoning can produce a bad outcome and vice versa, because outcomes contain a great deal of chance. Assessing your process rather than your results is the only way to learn in an environment with noisy feedback. Hindsight actively works against this by making outcomes look like consequences of decisions.

A written record of reasoning is what makes process review possible at all.

It distorts how you judge other people too

The same reorganisation of memory applies when assessing an adviser, a fund manager or a commentator, whose reasoning is judged almost entirely by what happened next. A manager who avoided a fall is credited with foresight even when the same positioning would have looked reckless had the fall not arrived, and one who held through it is judged careless for a decision that was defensible when it was made. This is part of why money moves toward whatever strategy last worked, which is close to the opposite of what the timing evidence supports.

Asking what someone said in advance, and how confident they were, is a far better test than asking what their recent results look like.

The takeaway

Write predictions down before outcomes arrive. Your memory will not preserve them honestly.

The version you keep doing is the version that works.