How Do I Stop Revenge Trading?

What does sizing up after a loss actually do?

It removes units of risk from the account, and units are the only thing standing between you and zero. Risk of ruin for a fixed-fraction bettor is approximately ((1 − A) / (1 + A))U, where A is your edge and U is how many units of risk the account holds.

Hold the edge at 10% and vary only the unit count:

Units of risk	Risk of ruin
20 units (normal size)	1.8%
10 units (double size)	13.4%
5 units (quadruple size)	36.7%

Doubling size makes ruin 7.4 times more likely. Quadrupling it — two revenge trades deep, which is the normal depth — makes it 20.3 times more likely. Your edge did not change. Your forecasting did not get worse. The account simply stopped being able to survive being wrong.

The cost of the extra activity is measurable. In the largest study of individual brokerage accounts, the households that traded most earned 11.4% a year against a 17.9% market^[1] — and revenge trading is that behaviour concentrated into the hours when your judgement is worst.

Why does the loss trigger it?

Because a realised loss is treated as a debt rather than a cost. The account is down $400 and some part of the mind starts working on recovering exactly $400, which quietly changes the objective from "make good decisions" to "get back to a specific number by a specific time". Every trade that follows is sized and timed against that deadline instead of against the setup.

Two details make it worse than it sounds:

• It is faster than deliberation. The next trade is usually taken within minutes, which is not enough time for anything reflective to intervene.
• It often works. Recovering the loss on the next trade is the outcome that trains the habit, and it happens often enough to be memorable. The failures are rarer and much larger, which is exactly the shape of a distribution people misjudge.

What actually stops it?

Deciding size before the session, in units of risk rather than dollars, and making the number something you would have to visibly overrule rather than quietly adjust.

A daily loss limit works for the same reason: it is a decision made when nothing is at stake, binding a decision made when everything feels like it is. Two units down, the day ends. Not because two units is a magic number, but because the rule has to fire before your judgement is the thing being consulted.

Is trading after a loss always revenge trading?

No, and treating it that way produces its own problem — a trader who cannot take the next valid setup because the last one lost is running a different superstition. The test is mechanical rather than emotional: is this trade the size it would have been if the last one had won, and would it have been taken at this time regardless? If both answers are yes, take it. If either is no, the previous trade is still making decisions.

The behaviour has been isolated in a laboratory, and the paper is almost named after it. Thaler and Johnson showed that a prior loss changes subsequent risk-taking: people accept gambles they would otherwise refuse when the gamble offers a chance to break even.^[2] Prospect theory gives the shape — risk-seeking below the reference point^[3] — and the reference point, after a loss, is the balance you had this morning.

kappi is a trade recorder: you commit a trade before the fact, it is sealed on your device for a time-capsuled delay you choose, then kappi publishes it on a Merkle-anchored log. The seal is the whole mechanism — once a stop, a target and a thesis are sealed, revising them is no longer a private edit. $15/month, no free tier.

Sealing the size alongside the entry matters more here than anywhere else in this cluster, because size is the field revenge trading actually touches. A stop that moves is one trade damaged; size that doubles is the account's survivability halved.