Does emotion affect trading returns? Here's how to find out on your own record

Last reviewed 17 September 2026.

You have sold in a panic on a red morning and watched the position recover by lunch. You have chased a stock three days into a run because everyone on your feed was up and you were not, and you have held a loser far too long out of a stubborn refusal to be wrong about it. The question is not really whether emotion moves your hand — you know it does. The question is how much it costs, and whether it costs enough to be worth doing something about.

We are not going to hand you a number and ask you to trust it. A number like that, unless it comes from your own trades, tells you almost nothing about your own behaviour — averages from other people's decisions wash out exactly the pattern you are trying to find in yours. So instead, here is the method in full: how to tag your decisions honestly, how to compare the groups once you have enough of them, and how many trades "enough" actually means.

The method: tag at the decision, not after

The rule to hold yourself to is simple: the tag has to be attached at the moment of the decision, not reconstructed afterwards. This is the whole method, and it is also the part almost everyone skips, because it is inconvenient in exactly the moment you are least inclined to be careful. Hindsight rewrites how we remember feeling — a trade that worked out gets remembered as "confident," a trade that didn't gets remembered as "I knew it was risky." Neither memory is reliable.

Step 1 — Pick a fixed list of emotions. Something like calm, confident, anxious, fearful, FOMO, revenge, bored, hopeful. A fixed list stops you from inventing a new, more flattering word each time.

Step 2 — Tag at entry and at exit, in the moment. When you open a position, write down the emotion and a conviction score from 1 to 5. When you close it, do the same for the sell decision specifically — a calm entry can still end in a fearful exit.

Step 3 — Record the realised return when the position closes. The maths is just:

Return % = ((Sale proceeds − Purchase cost − Fees) / Purchase cost) × 100

So a position bought for £2,000, sold for £2,240 with £8 of combined fees returns ((2240 − 2000 − 8) / 2000) × 100 = +11.6%.

Step 4 — Group into two buckets and compare. A "steady" bucket (calm, confident, hopeful) and a "reactive" bucket (fearful, FOMO, revenge, anxious) is enough resolution to start. Take the simple average return within each bucket. The gap between them — steady average minus reactive average — is the whole study, done on your own book.

Step 5 — Don't trust the gap until you have enough trades. A handful of trades in a bucket can swing wildly on one lucky or unlucky outcome. As a rough floor, treat anything under about twenty tagged trades per bucket as too thin to draw a conclusion from — interesting, not yet reliable. The more trades you accumulate, the more the noise averages out and the more the gap (or lack of one) starts to mean something.

What the research says, and what it doesn't

You don't have to take this on faith that it's worth checking — the underlying patterns are among the most-studied in behavioural finance, going back decades. Two are worth knowing before you tag your first trade, because they tell you roughly what to expect to find:

The disposition effect. Investors tend to sell winning positions earlier than optimal and hold losing positions longer than optimal — realising gains too fast and deferring the discomfort of admitting a loss. It is one of the most replicated findings in the field, first documented rigorously in large brokerage datasets in the 1990s and reproduced many times since. If your own "reactive" bucket skews toward early, panicked exits on positions that later recovered, this is very likely what you are seeing.

Overconfidence after a string of wins. A run of good decisions tends to inflate confidence faster than it inflates skill, and the trades placed at peak confidence are, on average, not the best ones — they are simply the ones taken with the least hedging and the largest size, so when they go wrong they go wrong loudly. If you track a conviction score alongside emotion, watch for this specifically at the very top of your conviction scale, not the middle.

What the research does not give you is a number you can apply to your own trading. Population-level effect sizes vary hugely by the market, period and investor population studied, and none of them describe you specifically. The value of the method above is that it replaces "research says investors do this" with "here is whether I do this, and by how much" — which is the only version of the question that can actually change what you do next.

One pattern worth watching for once you have enough tagged trades: many people find their reactive-bucket share of trades falls the longer they keep tagging — not because the market got calmer, but because naming the feeling before you act creates a half-second of friction that a surprising number of bad decisions don't survive. You cannot un-feel an emotion, but writing it down before you commit money is often enough to catch yourself.

What this can't tell you

Your tags are self-reported, which means they capture what you believed you felt, not a clinical measurement — and even with an at-the-moment design, people sometimes rationalise in real time. "Confident" and "FOMO" can be the same feeling wearing different clothes. Correlation is not causation: a calm tag does not make a trade return more — calm may simply be what a well-researched, appropriately-sized position feels like, and a reactive tag might correlate with smaller or later entries for reasons that have nothing to do with the feeling itself. This method also will not tell you whether any specific holding was a good idea; it tells you about the reliability of a pattern in your own decision-making, which is a different and, for most people, more useful question.

None of this requires you to become a monk about your investing. The practical takeaway is narrow and doable: build a two-second pause into reactive decisions, name the feeling before you place the order, and review the tag when the trade closes. The value is not in the label — it is in the friction the label creates.

Frequently asked

Does emotion actually affect trading returns?

Behavioural finance research says yes — the disposition effect (selling winners too early, holding losers too long) and overconfidence after a run of wins are among the most replicated findings in the field. Whether it costs you specifically, and how much, is only answerable by looking at your own trade history: tag how you felt at the moment of each decision, then compare realised returns across the tags once you have enough closed trades to average out the noise.

How do I measure this on my own trades?

Tag each trade with an emotion (for example calm, confident, anxious, fearful, FOMO or revenge) at the moment you make the decision, not afterwards — hindsight rewrites how you remember feeling. When a position closes, record the realised return. Once you have a few dozen tagged trades, group them into a "steady" bucket (calm, confident) and a "reactive" bucket (fearful, FOMO, revenge) and compare the average return in each. The gap, if there is one, is specific to you.

This article is for information only. It is not financial, investment or tax advice, and nothing here is a recommendation to buy or sell any security. StackViewer analyses decisions you have already made; it does not predict returns. Past performance does not indicate future results.