TL;DR
Revenge trading is a specific sequence, not a mood. A loss threatens how you see yourself as a trader, and that threat — not the market setup — drives a fast, oversized re-entry meant to undo the loss rather than execute a plan. It feels like conviction from the inside. In your trade log it looks like a size spike, a compressed exit-to-entry gap, and an entry that skips your normal criteria. Because the whole sequence runs in seconds, you almost never catch it while it's happening. You catch it afterward, by reviewing for these three tells.
Most traders think they'd recognize revenge trading if they were doing it. They wouldn't — not in the moment. That's not a character flaw, it's how the underlying trigger works. Understanding the mechanism, and knowing exactly what it leaves behind in your trade data, is the only reliable way to catch it.
It's a threat response, not an emotion
The common description — "trading out of anger" or "letting emotions take over" — is close enough to be useless. It tells you the feeling without explaining why it produces a specific, repeatable action.
Here's the actual sequence. A loss lands. For a trader who ties competence to outcomes — which is most traders, whether they'd admit it or not — a loss isn't just a number on the P&L. It's evidence against a belief: I'm good at this. That's a threat to self-image, and self-image threats produce urgency. The brain doesn't file the loss as "data point, update the model." It files it as "unresolved," and unresolved things demand closure. The fastest available form of closure is a new trade that proves the loss was a fluke.
That's the whole mechanism: loss → threat to self-image → need to undo it now → a trade selected for its speed of relief, not its edge. Anger, if it shows up, is a symptom of the threat response, not the cause of the bad trade. This is also why revenge trading concentrates on the same instrument you just lost on, in the same direction you were originally leaning — undoing this specific loss is the point, not finding the best opportunity on the board.
Once that need to undo takes over, the decision-making system that normally screens trades — checklist, setup criteria, risk sizing — gets bypassed. Not overridden by a conscious choice to break the rules. Bypassed, because the process running the show is no longer "does this meet my setup," it's "will this make the bad feeling go away."
Three checkable tells in your own trade history
Because you can't feel yourself doing this reliably, you have to look for it after the fact, in data. Three markers, in order of how reliable they are:
Position size spike right after a loss. Compare the size of a trade to your own trailing average, not to some abstract idea of "big." A trader who normally risks a consistent amount per trade and then, immediately following a loss, puts on 1.5x or 2x that size on the next one — with no corresponding change in signal quality — is sizing to the emotional need for recovery, not to the setup. Size should track conviction in a setup. When it tracks the size of the previous loss instead, that's the tell.
A compressed gap between exit and re-entry. Your normal process has a rhythm — time to scan, time to confirm a setup, maybe time to check other markets. Revenge trades collapse that rhythm. A trade placed two or three minutes after closing a loser, on the same instrument, is a trade that skipped the part of your process that usually takes longer than two or three minutes. The gap itself is diagnostic, independent of whether the trade wins or loses.
Entry without your normal setup criteria present. This is the one traders lie to themselves about most, because it's easy to retroactively justify almost any entry. The honest check: if you write down your setup criteria in advance (specific price action, specific confirmation, specific risk-reward), did the trade in question actually meet all of them, or did it meet "the market is roughly near where I was trading before, and I want back in"? A revenge trade often has a story attached after the fact — "it was retesting support" — that wouldn't have qualified as a reason to enter if the prior trade had been a winner.
Check your own re-entry pattern
Auraxon flags trades placed within minutes of a loss automatically, alongside the position-size and setup data, so you're checking a pattern instead of relying on memory.
What separates this from a legitimate re-entry
Not every quick re-entry into a losing instrument is revenge trading, and conflating the two makes traders second-guess entries that were actually fine. The distinction isn't speed or size alone — it's what's driving the decision.
A legitimate re-entry happens because the setup is still valid, or a new one has formed, independent of what the last trade did. If you took a long on a breakout that failed, and price pulls back to a clean level with fresh confirmation, entering again is following your process — the fact that you just lost on the same name is coincidental, not causal. The size should be normal or smaller, because the failed breakout is itself new information that arguably warrants more caution, not less. And you should be able to state the setup for the second trade without referencing the first trade's outcome at all.
A revenge trade fails that last test specifically. Ask yourself: "would I be taking this trade, at this size, right now, if my last trade had been a scratch instead of a loss?" If the honest answer is no — if the trade only makes sense as a response to the loss — it's a revenge trade regardless of whether the setup happens to look plausible on a chart. The market doesn't know or care what your last trade did. If your entry logic implicitly does, that's the signal.
Why willpower doesn't work here, and review does
The trigger mechanism above runs faster than conscious deliberation. By the time you're aware you're "thinking about" re-entering, the threat response has usually already selected the trade — you're rationalizing a decision that's substantially made, not weighing one that's still open. Telling yourself "I won't revenge trade" before a session is a bit like telling yourself you won't flinch before someone claps near your face. The intention is real. It doesn't reach the part of the system that actually fires.
This is why the fix that actually works isn't in-the-moment discipline — it's structural, decided in advance, and checked afterward rather than felt in real time. Two things do the job: a mechanical rule that doesn't require a decision under pressure (a fixed cooling-off period after any loss past a defined size, enforced by literally stepping away, not by willpower), and a regular review of your own trade log against the three tells above, so the pattern gets surfaced with numbers attached instead of staying a vague suspicion. Revenge trading is nearly invisible from inside a single session and fairly obvious once size, timing, and setup are laid out side by side across a week of trades — which is a data problem, not a discipline problem.
Walking through it
A trader is short a Nifty put, sized normally, on a setup that had worked three of the last four times. It stops out. Ninety seconds later, they're in again — same strike, roughly double the size, no new confirmation, working from a version of the original thesis that hasn't actually changed except that it's now personal. That trade also loses, and thirty seconds after that there's a third entry, still bigger, in a slightly different strike "to reduce correlation," entered with less certainty about the actual reasoning than either of the first two.
None of these three decisions felt like recklessness while they were happening. Each one felt like a reasonable adjustment. Laid out afterward — three trades in under three minutes, size climbing each time, setup criteria present on the first and absent on the second and third — the pattern is unambiguous. That gap, between how it feels at the time and how it reads afterward, is exactly why the trade log is the tool, not the memory of the session.
It's also worth being clear about what this piece isn't covering: how the resulting losses actually compound trade over trade — that's a distinct question with its own math, covered separately. What matters here is catching the sequence early enough that it doesn't get the chance to compound at all.
What is revenge trading, exactly?
A specific behavioral sequence: a loss threatens your sense of competence as a trader, and that threat drives a fast re-entry — usually larger, usually on the same instrument — meant to undo the loss rather than follow your normal process.
Is revenge trading the same as trading emotionally?
It's a subset. Emotional trading is a broad category; revenge trading has a specific trigger (a preceding loss), a specific target (undoing that loss), and a specific, checkable signature in trade data — size, timing, and missing setup criteria.
How do I tell a revenge trade apart from a legitimate re-entry?
Ask whether you'd take the same trade, at the same size, if your last trade had been a scratch instead of a loss. If the trade only makes sense as a response to the prior loss, it's revenge trading, even if the setup looks plausible on a chart.
Why can't I just catch myself doing it in the moment?
The trigger mechanism runs faster than conscious deliberation. By the time you're aware you're considering the trade, the decision is usually already substantially made. That's why retrospective review of your trade log works better than in-the-moment willpower.
What should I actually look for in my trade history?
Three things: position size that spikes right after a loss relative to your own average, a short gap between closing a loser and opening the next trade on the same instrument, and entries that don't meet your stated setup criteria.
Does revenge trading always make things worse?
It removes the process that normally screens trades for quality, so the trades it produces are, on average, worse than the ones your normal process would have taken — independent of whether any individual one happens to win.