Evidence Trading
The psychology of trading · Chapter 3 of 9

Tilt: how it starts and what it looks like

Foundation3 min read · part of The psychology of trading

Tilt is the state where you are still trading but no longer trading your method. The word comes from poker, and the useful part of the idea is that it is a state, not a mood — you can be calm and on tilt, and you can be annoyed and perfectly disciplined.

It rarely starts with a big loss. More often it starts with something that feels like unfairness: a trade that was right and lost anyway, a stop hit by a wick before the move you predicted, a missed setup that ran without you. What follows is a subtle shift in what you are trying to do. Instead of executing a method with an expectancy, you begin trying to get something back or prove something.

The observable signs are consistent, which is what makes them useful. Trade frequency rises. Position size drifts up. Time between trades falls. Setups get looser — you take the one that is "close enough". Analysis gets shorter. And crucially, you can usually justify every single one of these individually.

That last point is why self-diagnosis in the moment does not work. The state that impairs your judgement is the same state you would need good judgement to detect. This is not a failure of honesty; it is structural.

So the countermeasure has to be external and mechanical: a limit that acts without your agreement. A trade count cap, a daily loss limit, a cooldown timer after a loss. Evidence can flag the pattern from your own data, but the flag only helps if the constraint was already set — a warning you can dismiss is a warning you will dismiss.

What to take away

  • Tilt is a state, not a mood: still trading, no longer trading your method.
  • It usually starts with perceived unfairness, not with a big loss.
  • Signs: more trades, bigger size, shorter gaps, looser setups, less analysis.
  • You cannot reliably detect it from inside it — the constraint must be external and set in advance.

This chapter, measured against your own trades

In the app the same chapter ends in your figures rather than an example: how often you did the thing it describes, over your last ninety days. You pick one change to make, and Evidence checks afterwards whether it actually changed — from your journal, arithmetic, no opinion involved. Questions you get wrong come back a week later and again a month after that.

Open the free plan →

Check that it stuck

Answers shown — in the app these are asked before you see them, and the ones you get wrong come back after a week.

Which sequence is the most reliable early sign of tilt?
Trades getting more frequent while setups get looser — A single loss or a losing day is normal variance. The combination of rising frequency and falling standards is the behavioural signature, and it is measurable from your own journal.
Why is "I will notice when I am on tilt and stop" unreliable?
The state that impairs judgement is the same one you would need good judgement to detect — It is a structural problem, not a character one. That is exactly why the countermeasure has to be a limit that acts without needing your agreement in the moment.
A trader takes a setup that is "close enough" after two losses. What is the most accurate description?
A standards drop typical of tilt, individually justifiable — The defining feature of tilt behaviour is that each individual step can be justified. That is what makes it hard to catch without a pre-set rule.