# Revenge trading: what to check before your next order

> Revenge trading is entering a trade to recover a loss. Use a written pause-and-review checklist to decide whether the next order is a fresh, planned trade.

**Short answer:** Revenge trading is placing an order to recover a loss rather than because it meets rules you set before that loss. The test is not the size of the trade and not whether it makes money; it is whether you can describe the order without referring to the previous outcome. If you cannot, write the review down before placing anything: what is still open, what the loss actually cost, which limits you agreed in advance, and how the proposed order compares with them. Stopping is one of the two valid answers.

## What counts as revenge trading?

A trade taken to get back what the last one cost. The label describes where the decision came from, not what happened afterwards.

Three distinctions are worth keeping separate.

**Motive, not outcome.** A loss-chasing order can make money and still be revenge trading. A planned order can lose and still have been a planned order. The result does not reclassify the decision, which is why “it worked” is not evidence that the process was sound.

**Size is a symptom, not the definition.** Doubling up after a loss is the recognizable version. An order at your normal size, in a market you do not usually trade, taken because you want the balance back where it was, is the same behaviour wearing a calmer outfit.

**The trigger is usually a realized loss.** Sometimes it is a position closed too early, or a gain given back. In each case the next order is aimed at an earlier event rather than at current conditions.

Two terms used below. A **setup** is the set of conditions you decided in advance would justify an entry. A **discretionary order** is one you decide to place yourself in the moment, as opposed to one an existing rule or an automated system places.

## Why does a loss make the next order feel urgent?

FINRA’s investor guidance on volatile markets makes a general point that applies here: “the steeper the rise or the fall, the more tempting it can be to derail a long-term strategy with a snap decision”, and impulsive reactions are especially likely when markets fall sharply. Its advice is to focus first on your goals and your investment timeframe before acting ([FINRA, Market Volatility: Check Your Emotions at the Door, published 13 December 2018](https://syndication.finra.org/content/market-volatility-check-your-emotions-door-0)).

That source is general guidance written for long-term investors in stock markets. It is not a study of crypto trading behaviour, and it is not evidence that the checklist below improves anyone’s results. It supports one narrow proposition: a steep rise or fall makes a snap decision more tempting, and impulsive reactions are especially likely when markets fall sharply. The recommendation that follows from it — write the review down before placing anything, so there is a reference point to return to — is ours, not FINRA’s. Nothing cited here shows that writing it down works.

The other half of its framing does not carry over unchanged. “Keep your long-term view” assumes you are able to keep holding. In a leveraged crypto position — one where the exposure you hold is larger than the margin you posted, so price movement in both directions is amplified against that margin — the exchange can close the position if the collateral supporting it falls below its required level, and that closure happens at the market’s price, not yours. Taking the long view is not a reason to remove a protective order or to leave a position unmanaged while you wait.

## Four warning signs that the next order is about the last one

These are editorial warning signs: patterns worth noticing in your own behaviour. They are not diagnostic criteria, and they do not predict results.

**1. The size was derived from the loss.** If the quantity came from asking what would make the number go away, rather than from the trade’s own stop distance and your risk budget, the size is a statement about the past.

**2. The entry is outside your setup.** A market you do not normally trade, a timeframe you do not normally use, or conditions you would have skipped an hour ago. The setup did not change; your willingness to ignore it did.

**3. The rules keep changing after each loss.** A stop moved further away, a loss limit raised, the definition of a confirmed entry loosened. Each edit can be argued for individually. Together they mean there is no rule left to compare the next order against.

**4. The previous account peak reads as a debt.** A peak is a historical high point in your balance. The market does not owe a return to it. Drawdown — a fall from a previous peak — describes what already happened; it is not an amount due on a schedule. The recovery arithmetic is covered in [drawdown in trading](/blog/drawdown-in-trading): a deeper fall needs a disproportionately larger gain to undo it. That is why getting back to even feels urgent, and why the feeling is a poor input to position size.

## The post-loss checklist: pause, review, then decide

This checklist is an editorial risk-control aid written for this article. It is not a clinically tested intervention, it does not diagnose anything, and it does not promise better results. Its purpose is narrower: to produce a written record you can hold next to rules you set before the loss, at the moment when comparing the two is least appealing.

1. **Pause new discretionary entries.** The pause covers orders you would place yourself to open new exposure. It does not cover managing risk that already exists — those two pull in opposite directions, and step 3 is why.
2. **Review what is open before changing anything.** Current positions and their notional (the total value the position is exposed to, which is not the same as the collateral you posted). Pending orders still waiting to fill. Any collateral or margin alert from the exchange. And the protective orders attached to each position — the stop or take-profit orders intended to close it at a defined level.
3. **Do not cancel protective orders as part of the pause.** Pausing new entries while leaving existing positions unmanaged is the opposite of the point. A protective stop stays unless you have a reason to change it that you can write down, and even then it is replaced rather than simply removed. A stop is not a guaranteed maximum loss: in a fast move it can fill worse than its trigger price, and a limit-based stop can trigger and remain unfilled.
4. **Record the actual loss.** The realized loss on the closed position, plus fees on entry and exit, plus funding — payments you made or received for holding the position, netted into one signed figure — plus any other charge or credit attached to it. Then write down your remaining exposure, and the original rationale — the reason you entered, as you wrote it before the trade, not as you remember it now.
5. **Compare the proposed order with your pre-agreed limits.** Three comparisons: does it match a setup you defined in advance; is the size within your per-trade limit; and is it within your session-loss limit, meaning the total loss after which you stop trading for that session.
6. **If you have no pre-agreed limits, you cannot invent them now.** A limit written after the loss and sized so that this trade fits inside it is not a limit. Record it as absent and treat the absence as the finding. Setting limits is separate work, done when nothing is at stake; [crypto position size](/blog/crypto-position-size) covers how a loss budget and a stop distance turn into a quantity.
7. **Do not increase size solely to recover the loss.** A larger position because the setup justifies it and your limits allow it is a decision. A larger position because the previous one lost is not.
8. **Apply the explanation test.** Write out the proposed order: the market, the setup, the size and why that size, the level at which the idea is wrong, and what you will do when it is wrong. Then remove every reference to the previous trade. If the description no longer stands up, the order is about the loss.
9. **Decide, and record the decision.** Two valid outcomes: restart under the original rules, or stop. Restarting is not required, and stopping is not a failed review. There is no universal cooldown period that fits every trader, market or account, so the test is not that a particular amount of time has gone by. Restarting is available only when the explanation test passes *and* every pre-agreed limit still allows the order. If a pre-agreed rule has already ended the session — the session-loss limit is reached, or a rule you wrote beforehand stops trading for the day — the session stays closed. The review records that; it does not reopen it.

## A filled-in example: what the record looks like

The example below is constructed for this article. “T” is a label, not a person, not a Cointrapper participant and not a customer case study. The numbers are invented to show the form being filled in. It deliberately shows no profit outcome, because the artifact records a decision, not a result. The numbers in the left column are record rows, not the checklist step numbers above: some steps are written across more than one row.

| Field | Filled in |
| ----- | --------- |
| When | Second session of the day. The date and the times of the pause and of the review belong in the record; the values here are constructed. |
| 1. Paused | No new discretionary entries from the moment the position was closed. |
| 2. What is open | One long position, notional 800 USDT, protective stop in place. One pending limit buy, cancelled during this review: it was sized for the position just closed, not for anything T would open now. No margin alert; collateral above the required level. |
| 3. Protective orders | Unchanged. The stop on the open position stays where it was set before the loss. |
| 4. Actual loss | 120 USDT price loss + 4 USDT fees on entry and exit = 124 USDT. |
| 5. Remaining exposure | 800 USDT notional, unchanged by this review. |
| 6. Original rationale | Written before entry: a break and retest of the range high, invalid below the range midpoint. The position was closed at the invalidation level, as planned. |
| 7. Pre-agreed limits | 150 USDT session loss, 800 USDT maximum notional per trade, entries only in the two markets on the list. All three were written before the session. |
| 8. Proposed order | 2,400 USDT notional in a market that is not on the list, entered because it “has to move back”. |
| 9. Comparison | Size is three times the per-trade limit. The market is outside the setup list. 124 of the 150 USDT session-loss limit is already used. |
| 10. Explanation test | Fails. With no reference to the 124 USDT, there is no stated reason to open this order. |
| 11. Decision | Stop. No new entries this session. The open position keeps its stop. The session is closed by this decision, so nothing reopens it during the session. In a later session, a new entry still requires the explanation test to pass and every pre-agreed limit to allow it. |

The same form can end in “restart”. That version has to show a proposed order matching a setup on the list, a size inside the per-trade limit, a session-loss limit not yet reached, no other pre-agreed rule that has already ended the session, and an explanation-test description that still makes sense with the previous trade deleted from it. Neither version says anything about whether the next order makes money.

## What the checklist cannot do

- **It does not make losses smaller.** It can keep an unplanned order out of the account. It cannot turn an approach that loses into one that does not.
- **It is not a diagnosis and not medical or psychological advice.** Nothing here identifies a condition, and no warning sign above is a symptom of one.
- **It does not guarantee that a protective order executes.** A stop can fill worse than its trigger in a fast move, and a limit-based stop can trigger without filling. The exchange’s current documentation for its own order types is the authority on that, not this article.
- **It offers no safe loss cap and no standard cooldown.** Any number you use is a rule you chose for your own account. There is no duration or threshold that applies to everyone.
- **It does not validate your limits.** If the pre-agreed size was already too large for the account, following it faithfully still loses money. The checklist compares behaviour with rules; it does not judge whether the rules were sensible.

If loss-chasing has become frequent, hard to stop, or is affecting your sleep, your mood or money you need for other things, stepping away from trading and talking to a qualified professional is a reasonable step. This article does not assess that and is not a diagnosis or a treatment recommendation.

## What this means for your capital

A realized loss is information about a position that is already closed. The next order is a separate decision with its own setup, its own size and its own level of invalidation, and it is entitled to be judged on those alone.

The checklist exists to make that separation visible in writing, at the moment when it is hardest to feel. Fill it in and the answer is sometimes “no new entries”. That is a complete outcome, not an incomplete one.

We run Cointrapper, a private, invite-only algorithmic trading pool on Binance. Participants invest USDT into the pool’s shared accounts and hold units in the pool; they do not hold individual futures positions of their own, they do not place these discretionary orders themselves, and the pool is not a bot running on a participant’s exchange account. That changes who makes the entry decision. It does not make the trading risk-free, and automation is not protection from loss.

Futures trading and pooled investing are high risk. Your entire investment can be lost, and capital is not insured. Nothing here is individual investment advice, and exchange rules, margin requirements and order behaviour change — check the current documentation for the exchange and contract you use.