Nine cognitive distortions quietly killing your trading account

Here's the thing about cognitive distortions: they're not personality flaws. They're not signs that you're bad at trading. They're standard equipment in every human nervous system, useful in normal life, catastrophically expensive when you're holding leveraged positions.

The single most useful thing you can do as a trader is learn to name them. Once you can identify a pattern as “loss aversion” — out loud, in the moment, while it's happening — you've put a wedge between you and the impulse. That's the gap where the better decision lives.

What follows is the nine that show up most reliably in trading data. None of them are exotic. All of them have killed accounts you've heard of.

1. Loss aversion

The pain of losing $100 hits roughly twice as hard as the pleasure of making $100. Kahneman called this the most well-replicated finding in behavioural economics. In trading, it shows up as: holding losers too long because closing the position means making the loss real. The position is already a loss the second the price ticked against your stop — the trade is just renegotiating with you about whether you have to look at it.

The fix isn't willpower. It's pre-decided exits and a system that respects them.

2. Revenge trading

You took a loss. The next entry is the same setup, double size, no plan. Your dataset will tell you, if you're honest with it, that the trade taken inside three minutes of a stop-out is almost always one of your worst-performing entries. The market doesn't owe you anything. You can't “make it back” on the same instrument that just took it.

Concrete fix: a consecutive-loss-lock rule. Two losses in a row triggers a soft lock. Three in a row triggers a session-ending hard lock. The threshold isn't arbitrary — it's the inflection point where the trader stops being the trader and starts being a different, worse one.

“The market doesn't owe you anything. You can't make it back on the instrument that just took it.”

3. FOMO

Fear of missing out. The setup ran without you. Now everything looks like the next setup, and you're ready to take a worse trade because you're still mentally trying to catch the one you missed. The honest read: the trade you missed wasn't actually in your playbook — you're post-rationalising a decision you would have made for the wrong reasons.

FOMO is hardest to catch in real time. The most reliable signal that you're in it: your reasoning is built around what just happened, not what's happening now. If you find yourself saying “I should have got in earlier,” you're probably about to take a FOMO trade.

4. Sunk cost

“I'm already in this far.” You're a leg into a vertical spread that's gone the wrong way and now you're thinking about rolling it. You held a futures contract through the news event because closing it would lock in the loss you took mentally an hour ago. You doubled the size on a losing position because at the new average you'll “just need a small move to get back to even.”

Money already lost is gone. Every decision from this point should be made on the situation in front of you, ignoring the entry price entirely. The “average down” trade is the most reliably wrong move in retail trading, and it's sunk cost wearing a different jacket.

5. Anchoring

The first price you saw becomes the price the rest of the trade is measured against. You shorted at 21,850. The market is now 21,930. You're mentally looking for it to come back to 21,850 because that's where you got in — instead of asking what the actual current setup is. The entry price isn't a magnet. The market doesn't care where you got in.

This is why disciplined traders journal entries against the setup, not the price. The setup is the reason the trade exists; if the setup is invalidated, the price is irrelevant.

6. Recency bias

The last three trades dominate your read of your own ability. Three winners in a row and you're sizing up. Three losers in a row and you're paralysed. Both reactions are wrong because three trades is statistically nothing. Your edge plays out across hundreds.

The honest correction is to look at distribution-level metrics — R-multiple distribution over 100+ trades, win rate by setup over 50+ trades, hold time vs R over the full dataset. Three trades is noise. Acting on it as if it's signal is a recurring source of risk drift.

7. Confirmation bias

You're long. The chart pulls back. You're looking for reasons it's a healthy pullback, not reasons it's the start of a reversal. The signals that contradict your position get downweighted; the ones that support it get upweighted. Same chart, different read, depending on which side you're on.

The discipline here is asking, before every continuation decision: if I weren't already in this trade, would I take it? If the answer is no, the position should be closed, not held. The fact that you're in it is not a reason to stay in it.

8. Premature exit

The opposite failure mode: you cut a winner too early because you're scared of giving back the gain. The trade was working, the setup was valid, the target was 2R, and you closed at 0.4R because the position wobbled and you wanted the win locked in.

Premature exit is loss aversion playing forward — pre-emptively turning paper gains into a smaller real gain because the prospect of the gain disappearing feels worse than the smaller gain feels good. It's the same mechanism, in the opposite direction. And over a hundred trades, it's the difference between a profitable system and a flat one.

9. Overconfidence

A winning streak feels like a permanent change in ability. It isn't. The streak is one of two things: variance working in your favour, or genuine signal. Without enough trades to tell the difference, you should default to assuming variance, because the cost of being wrong about variance (you size up, get blown out) is much higher than the cost of being wrong about genuine signal (you're slightly under-allocated for a few extra weeks).

The most expensive trades in any account's history almost always cluster on the days right after the most profitable trades. The trader is the same. The setup is the same. The size isn't.

What naming gets you

A surprising amount of the value of behavioural-trading work comes from the simple act of identifying a pattern by name in real time. Not because the name fixes anything — because the name forces the meta-cognitive layer on. You stop being the impulse. You become the trader watching the impulse, with about three seconds to decide whether to act on it.

Three seconds is enough. Three seconds is the entire window that separates the trade you take and the trade you almost took.

Most journals catch these patterns three days later, three weeks later, three months later. By then the dataset is thicker but the trade is gone, the lesson is abstract, and tomorrow at 10:32 ET the same pattern will fire again because nothing in your system intervened the moment it counted.

Kaxse's AI coach is built around exactly this — naming the distortion in your live session chat the moment the pattern shows up, so you get the three seconds. The taxonomy is documented at /features/ai-coach. The full philosophy at /risk-management.