Behaviour
Your calm self cannot predict your panicking self
Asked in a quiet moment how you would react to a crash, you will answer as the person who is not experiencing one.

The options around predicting your own reactions are set out side by side below, with the conditions that genuinely favour one over the other.
The difference in one place
- People systematically underestimate how differently they will behave under emotional pressure.
- Stated risk tolerance is collected in exactly the wrong emotional state.
- Pre-commitment works because it does not rely on future self-control.
The gap between states
Decisions made calmly and decisions made under stress are made by what is effectively a different decision-maker with different priorities. While calm, it is very difficult to simulate the urgency, fear and time pressure of a falling market. This gap between emotional states has been studied in several contexts and the general direction is consistent: people underestimate how much the state changes them.
It is why sincere intentions to hold through a crash often fail.
Questionnaires ask the wrong person
Risk tolerance is typically assessed on a quiet afternoon with no money moving, which is the least informative possible setting. The answers describe how you feel about an abstraction, not how you behave when a real balance is falling. This is a structural weakness of the instrument rather than a flaw in any particular one.
On an ordinary week, treating the score as a starting point rather than as a measurement is the honest use.
Making it more realistic
Translating percentage falls into your own currency helps, because currency is more concrete than percentage. Recalling how you felt in a previous decline, using notes rather than memory, helps more. For someone with no prior experience of a fall, the honest answer is that their tolerance is unknown.
That argues for starting more cautiously and adjusting upward with evidence.
Pre-commitment beats willpower
Techniques that work are those that do not require the stressed version of you to make good decisions. Automated contributions, written rules, removed apps and scheduled reviews all shift the decision to the calm state. This is the same logic as any commitment device: bind the future self rather than trusting it.
It also means the plan should be as unambiguous as possible, since ambiguity is where the stressed self operates.
Rehearsal has limited value
Imagining a crash is not the same as experiencing one, so mental rehearsal produces less preparation than it feels like. What does help is having written instructions to follow, since following is easier than deciding.
Deciding in advance who you would speak to before acting is another useful pre-commitment. A required conversation adds delay, and delay is what most impulsive decisions cannot survive.
Some of this will suit you and some will not, and that is the point.
After the event
Post-crash, memory rapidly reconstructs how you felt, usually in a more flattering direction. Contemporaneous notes are therefore the only reliable input to future allocation decisions. If experience showed the allocation was too much, adjust once conditions are calm, with advice if the change is material.
The adjustment should reflect what you did, not what you remember feeling.
Side by side
| Consideration | What it means in practice |
|---|---|
| The gap between states | People systematically underestimate how differently they will behave under emotional pressure. |
| Questionnaires ask the wrong person | Stated risk tolerance is collected in exactly the wrong emotional state. |
| Making it more realistic | Pre-commitment works because it does not rely on future self-control. |
The takeaway
Bind the calm version of you to a written instruction. The other version does not take advice.
The version you keep doing is the version that works.
Questions readers ask
How can I test my risk tolerance before a crash?
You cannot fully. Sizing the fall in your own currency, and starting at a level you would clearly tolerate, are the closest available approximations.
I panicked last time. Does that mean I should hold less risk?
It is strong evidence about you and worth acting on once markets are calm. What allocation suits you depends on your wider circumstances and is a question for regulated advice.
Also by Bethan Rees
- What a bond allocation is actually forRisk & Allocation
- Overconfidence is quietest when it is most expensiveBehaviour
- Why the same money gets treated differentlyBehaviour
- Hindsight makes every crash look obviousBehaviour





