4 / 6 · 9 min
Tilt: why winning it back is worse than simply stopping
After a loss people go to recover the money in the same coin — and we measured that the market at that moment swings six times wider than usual. Plus the arithmetic that makes the hole deepen faster than it fills.
Tilt is a word from poker, and it is more precise than any translation. It is not anger or upset; it is a state in which the GOAL IS QUIETLY SWAPPED. A second ago the goal was to trade well over a distance. Now the goal is to get back these particular dollars, preferably in the same coin where they were lost. The person keeps performing sensible-looking actions while solving a different problem.
Three signs by which tilt is recognised from outside
- Position size grew without any change in the calculation: «I am sure about this one, I will take more».
- The trade is opened in the same coin where the loss happened, and the justification appeared after the decision.
- The gap between trades shrank: what used to take an hour of thought now takes a minute.
The measurement: why recovering there is the worst choice
We took the same 60 liquid coins and 83 hours of minute candles and compared two quantities: the ordinary hourly range of price, and the range in the hour that follows immediately after a drop of 3% or more within five minutes. That is, right after the very move that usually makes a person want their money back.
Hourly price range (median)
| When | Range | Observations |
|---|---|---|
| An ordinary hour | 1.61% | 2460 |
| The hour right after a 3%+ drop | 9.88% | 427 |
Six times wider. This means your usual stop and usual size represent an entirely different risk at that moment: a stop that yesterday caught a meaningful move today sits inside noise. Revenge trading is dangerous not only because the decision is emotional — it also lands in the very hour when the market is objectively least predictable.
The arithmetic of the hole
The second reason is pure mathematics and does not depend on mood. Loss and recovery are asymmetric: to return to the previous account you must earn a larger percentage than you lost.
How much you must earn to get back
| Lost | Must earn |
|---|---|
| 10% | +11.1% |
| 20% | +25.0% |
| 30% | +42.9% |
| 50% | +100% |
| 70% | +233% |
Read this not as a scare story but as a map of where the boundary runs. Up to 10–20% a drawdown is still curable by ordinary work. Past 50% you must double the account — that is, do something you had not done in all the time before the drawdown. This is exactly why stopping must happen early rather than «when it gets really bad»: early is cheap, late is nearly impossible.
Doubling size after a loss: where the account ends
The commonest form of revenge trading is to increase size so one trade brings back what was lost. Suppose risk starts at 1% of the account and doubles after every loss. The cumulative risk of a streak is then simple: 1, 3, 7, 15, 31, 63, 127 percent of the account on the first, second and so on consecutive loss. On the seventh loss there is no account left. Now let us see how close that is: at a 50% win rate, a streak of seven consecutive losses occurs within a hundred trades with probability 31.8% — roughly every third hundred. This scheme is not «risky», it is designed to end in ruin; the only question is when.
Where n is the length of the losing streak. The formula explains why doubling looks convincing and works for a while: six losses in a row cost 63% and leave hope, while the seventh costs 127% and leaves nothing. The scheme delivers many small wins and one loss that closes the account entirely.
«Today I win it back, tomorrow I return to the rules»
The phrase feels like a temporary exception, but it inverts the whole construction: the rules apply while everything is fine and are suspended exactly when they are needed. One question tests you: had you seen this same trade on a clean account, with no morning loss behind it, would you have taken it? If not, the trade is being chosen by your drawdown, not by the market.
What actually works — a stop set in advance
Inside tilt the decisions are made by a person with impaired judgement, and asking him to «weigh things up» is useless. Only a rule works: one written in a calm state and requiring no judgement at the moment it fires — a plain number at which the trading day is over. Note the difference from the risk course: there a drawdown limit protects capital, here it interrupts a state. That is why the daily limit sits noticeably earlier than the monthly one.
Stopping rules that require no thinking
- A daily limit in money or in R: once reached, the terminal closes until tomorrow, no discussion.
- No more than N trades a day: the number is chosen in advance, usually half of what feels necessary.
- Three losses in a row — pause until the next day, even if the money limit was not reached.
- Once the limit has fired, changing the limit itself on the same day is forbidden.
- Return to normal size only after a review has been written, not after the first profitable trade.
The easiest place to test stopping is where a mistake costs nothing: in paper trading, set yourself a daily limit and see whether you actually stop at it.
Open paper tradingSet your two numbers today
Write two quantities on paper: how much money you are willing to lose in one day, and how many trades you are willing to take. The numbers should be slightly uncomfortable — if they feel convenient, they will not work. Put the paper where it is visible from your desk. A month later count how many times you broke them: that is the measured size of your problem, rather than a feeling about it.
Why is trying to win it back in the same coin right after a loss especially expensive?
Because two things combine. The decision is made in a state where the goal has been swapped for «get these dollars back», and it lands in an hour when the price range is six times the usual (9.88% against 1.61%). A habitual stop sits inside noise in such an hour, and a habitual size represents a completely different risk.
Doubling size after a loss produces many small wins. Where is the catch?
In the cumulative risk of a streak growing as 2ⁿ − 1: the seventh consecutive loss costs 127% of the account. At a 50% win rate the chance of meeting such a streak within a hundred trades is 31.8% — roughly every third hundred. The scheme is not risky but designed for ruin: it merely postpones it and makes it feel sudden.