Revenge Trading: How to Catch Tilt in Your Own Trade History

Revenge trading has a reputation as a discipline problem, something you fix with willpower and a sticky note on the monitor. The sticky note fails for a specific reason: tilt does not feel like tilt while it is happening. It feels like conviction. The trade you take four minutes after a stop-out feels urgent and well-reasoned, and only looks unhinged the next morning, if you look at all.

So this post is not about feeling your way to discipline. It is about the marks revenge trading leaves in your trade history, which are measurable, and how to find them in your own data.

What it is, mechanically

Revenge trading is re-entering the market to recover a loss rather than to take a setup. The trigger is the loss itself. The tell is that the trade would not have been taken if the previous trade had won.

That definition matters because it makes the behavior detectable. A trade motivated by a setup is independent of what happened an hour ago. A trade motivated by a loss is correlated with it, and correlation shows up in data even when it hides from memory.

The signatures in your data

Pull your trade history, sort it by time, and look for these five patterns. Each one is checkable with timestamps, sizes, and PnL, no self-honesty required.

Clustering after losses. Measure the time gap between each trade and the one before it, then split the gaps by whether the previous trade won or lost. Tilted traders re-enter faster after losses, often dramatically faster. If your median gap after a win is forty minutes and after a loss it is six, the six is the loss placing the next order.

Size spikes after losses. Compare your position size on trades that follow a red trade against your overall average. The revenge pattern is a size increase, sometimes a doubling, in the trade immediately after a stop-out. At the screen it feels like conviction. In a table sorted by time it is a size spike four minutes after a red trade.

Performance decay after losses. Compute the expectancy of trades taken within an hour of a losing trade, and compare it with everything else. For most traders with a tilt problem this single split is the most expensive line in their history. It is also the most useful, because it converts a vague character flaw into a dollar figure.

The late-night entries. Slice your results by hour. Crypto trades 24/7 and your discipline does not. If your losses concentrate in a specific window (very often the late-night hours after a red day), your history is telling you when you should not be at the screen.

Setups that do not exist. If you tag your trades with the setup that justified them, revenge trades are the ones with no tag, or a tag you had to invent at entry. An untagged cluster right after a loss is about as clear as the evidence gets.

Running the audit

You can do this in an afternoon with exports and a spreadsheet, and if you keep a manual journal that is exactly how to do it: add a column for time-since-previous-trade and a flag for previous-trade-was-a-loss, then pivot expectancy on the flag.

The catch is that the audit is only as good as the record. Memory will not reconstruct last month's timestamps, and screenshots do not carry position sizes. If your history is scattered across exchanges or has gaps where the bad weeks were, the analysis quietly excludes the evidence it most needs. The traders who most need this audit tend to have the least complete records, and that is not a coincidence; not wanting to look and not logging are the same behavior.

This is the strongest practical argument for a journal that fills itself from your exchange history: the record exists whether or not you were in the mood to keep it, including the weeks you were not.

What traders do with the answer

We are careful here, because no journal fixes tilt, and anyone who promises otherwise is selling something. What the data changes is the feedback loop. Some patterns that show up among traders who run this audit:

• A cooldown rule with a number on it. No entries for 45 minutes after a stop-out, where 45 came from their own gap analysis rather than from a resolution to be more patient.
• A size cap after consecutive losses. If the data shows size spikes after two reds, the rule targets exactly that state.
• A trading window. If the 2am trades have negative expectancy across six months of history, stopping at midnight stops being a discipline question and becomes arithmetic.
• Writing the why at entry. One sentence, at entry time, about what the setup is. Its absence later is the tag that catches the trades that had no why.

None of these require becoming a different person. They require noticing the pattern, and the pattern lives in the record, not in your self-image.

The next red trade

The uncomfortable version of this post is one sentence long: you already know whether you revenge trade, and you have been choosing not to check. The comfortable version is also true: checking is a one-time afternoon of work, the patterns are usually obvious once plotted, and traders who put a number on the habit tend to find it much easier to break than traders who fight it as a character flaw.

Viktury computes the raw material for this audit automatically: complete timestamped history across your exchanges, position sizes, per-hour breakdowns, and a mistake tag whose cost it totals for you. Start your free trial, sync your history, and look at what the hour after your losses has been costing.