Why Traders Hold Losing Positions: What the Evidence Actually Says

By Sofia Harchich | Trading Psychologist & Behavioural Finance Writer | thewealthmirror.com

Why traders hold losing positions has a standard answer, and the standard answer is doing less explanatory work than it appears to.

Loss aversion, the argument runs: losses hurt about twice as much as equivalent gains feel good, so the trader avoids realising the loss and holds. It is cited constantly, it is genuinely well-evidenced, and it accounts for perhaps half of what actually happens at a desk.

Why traders hold losing positions has a second half, and that half is what this article is about, because the missing half is where the intervention lives. A trader who believes the whole explanation is loss aversion will try to override a feeling. A trader who understands the rest will change something structural, and only one of those two approaches survives a bad Tuesday.

Finding one: the asymmetry is real, and it is not about money

What the evidence supports: strongly.

Prospect theory established that people evaluate outcomes as gains and losses relative to a reference point rather than in terms of final wealth, and that the loss side of the value function is steeper. This is one of the more robust findings in behavioural economics and it replicates across contexts that have nothing to do with markets.

Two details get dropped in the trading version, and both matter.

The first is the reference point. The asymmetry is measured relative to a reference, and the reference is chosen, not given. A position is a loss relative to entry price. It is something else entirely relative to the day’s starting balance, or to the position’s planned invalidation level. Traders who set the reference at entry price have constructed the steepest possible version of their own value function, and most platforms encourage exactly that by displaying unrealised profit and loss per position by default.

The second is that the asymmetry describes a feeling about an outcome, not a decision rule. It explains reluctance. It does not explain why the trader who is reluctant at 10:20 is still holding at 15:40.

Finding two: realisation, not the loss, is what gets avoided

What the evidence supports: strongly, and it is the most useful finding here.

The disposition effect — the documented tendency to sell winners early and hold losers — is not really about the money either. The account is already down. Closing the position changes nothing about the balance.

What closing changes is the status of the loss. Open, it remains provisional; something might still happen. Closed, it becomes a fact with a number attached, entered into a record.

This reframes the whole problem. The behaviour is avoidance of resolution rather than of loss, and that is a considerably more tractable target, because resolution is a mechanical event that can be pre-scheduled while a feeling cannot.

The position stays open because closing it makes the loss true, and a provisional loss can still be argued with.

Finding three: the pain is anticipated, and anticipation is beatable

What the evidence supports: moderately, with a caveat trading content ignores.

The neural findings usually cited here are real but routinely over-read. What can be said with reasonable confidence is that anticipated loss engages processing associated with aversive states, and that the anticipation is often more intense than the event.

The caveat is that this varies enormously between people and within the same person across states. Sleep, prior outcomes that day, and baseline arousal all move it. Anyone presenting a specific brain region as the reason you held a position is compressing a messy literature into a slogan.

What survives the caveat is practically useful: the moment of maximum aversion is before the close, not after. Traders consistently report that closing was easier than expecting to close. That asymmetry is why a pre-committed mechanical exit works — it does not require the trader to be brave at the worst moment, only to have been honest at a better one.

Finding four: the account is standing in for something else

What the evidence supports: weakly as formal research, strongly as observation — and stated as such.

Here the honest position is that the literature thins out and the pattern in behavioural records is nonetheless consistent.

Holding times on losing positions vary within the same trader in ways that price action does not explain. The same person will cut a loser in nine minutes on one instrument and hold a structurally identical loser for two sessions on another. What differs is rarely the setup. What differs is what the instrument carries — gold behaves differently in records than an index does, and it does so because scarcity, permanence and prestige are attached to it in a way they are not attached to a basket of equities.

This is the layer where the loss stops being financial. A position held past every rational exit is usually protecting a claim about judgement rather than capital, and self-serving bias shows how readily the mind protects that claim: wins get attributed to skill, losses to circumstance. A closed loser makes that attribution harder to maintain.

Taken together, the findings place the hold inside the instability loop. Leaving the position open is the behaviour that stops the anticipated pain of realising the loss — for now — and every hold that recovers teaches the loop to repeat it.

Why traders hold losing positions: what the four findings change

Read together, they point somewhere specific, and it is not cut your losses faster.

1. Set the reference point away from entry price. Track the position against its invalidation level, and the day against its starting balance. This is a display setting, and it changes the steepness of the value function you are operating on
2. Pre-schedule the resolution. The exit condition is an observable event, written before entry, with no escape clause — because what is being avoided is resolution, and resolution can be automated in a way that courage cannot
3. Decide at the moment of lowest aversion. Before entry, not during. The pre-trade minute is the only point in the trade when your judgement is operating at full capacity
4. Log the instrument on every position. If holding times cluster by symbol rather than by setup, the fourth finding is running and the fix is not a tighter stop

Finding five: what the traders who exit on time do differently

Supported: as observation across behavioural records, and stated as such.

There is a contrast group worth examining, and it is more informative than any amount of study of the problem itself.

Why traders hold losing positions is best answered by looking at the people who do not. Some traders close losing positions at invalidation, consistently, without visible struggle. The obvious explanation — that they feel the loss less — appears to be wrong. Asked directly, they describe the same reluctance, the same pull to wait one more candle, the same dislike of being wrong. Whatever separates them is not affective.

Three structural differences show up instead.

The exit was defined as an event, not a judgement. They wrote closes below this level on this timeframe before entry, so the live decision is a matter of noticing rather than of deciding. Why traders hold losing positions is, on this reading, largely a question of what was left undefined before the position opened.

The reference point was never entry price. They track distance to invalidation rather than unrealised profit and loss, which means the position is not continuously reporting how much has been lost relative to the steepest possible baseline.

The exit is scored as compliance, not as failure. Closing at invalidation increments a number they care about. There is a scoreboard on which cutting the position is a win, and it is visible.

None of the three requires feeling differently, and that is the whole point. Each one moves a decision out of the moment when the compound of reluctance and self-justification is strongest, and into a moment when nothing is at stake.

There is a fourth difference, harder to pin down and more tentative than the first three: describing the position in the third person. The trade was invalidated rather than I was wrong. Whether the language causes the behaviour or simply reports it is unclear, but it costs nothing to adopt, and it is worth checking how your own journal describes the positions you held too long.

Worth saying plainly, because the comparison can land badly: these traders are not more disciplined than you in some general sense. They have front-loaded a set of decisions you are currently making live, under load, with a degraded working memory and a reference point that platform defaults chose for you. That is a difference in preparation rather than in character, and preparation is available.

The reading that is not in the data

One thing worth separating out, because it is what usually turns this from a technical problem into a heavy one.

Holding a losing position past its invalidation is not evidence of weak character, and traders who conclude that it is tend to add rules rather than change structure — which reliably produces a trader with stricter rules and the same holding times. What the four findings describe is a system doing exactly what it evolved to do: protecting a reference point, deferring resolution, and defending a claim about competence. All three are adaptive nearly everywhere except in a market.

That is also why the interventions above are structural rather than motivational. The behaviour is rational with respect to something other than the account, and the fix is to make the account’s requirements explicit before the other thing gets a vote.

Where this connects

The same architecture appears with the polarity reversed in why traders exit trades too early, and it compounds with identity in loss aversion, ego and the trading account.

That is the practical answer to why traders hold losing positions. For the loop this sits inside, start with the cluster hub, how to stop revenge trading. The pattern quiz identifies which emotional pattern your behaviour under pressure fits, which determines which of the four protocol changes above will do the most for you.

The question that replaces “why am I like this?”

Why traders hold losing positions is a better question than the one most traders are actually asking at ten in the evening, which is usually some version of what is wrong with me.

The five findings answer the first and dissolve the second. A reference point set by a platform default. A resolution being deferred rather than a loss being avoided. An anticipation that peaks before the act. An instrument carrying weight that a spreadsheet cannot see. A contrast group who prepared rather than endured.

None of that is a character description, and all of it is adjustable in an afternoon.

Understanding why traders hold losing positions changes what you go and do next: a display setting, a written invalidation, a second scoreboard, and an instrument column. Four structural changes against a problem that has been treated as a moral one for as long as anyone has been writing about markets.

---

About the Author

Sofia Harchich is a Trading Psychologist and Behavioural Finance Writer with a Master’s in Psychology. She works at the intersection of Jungian shadow work, neuroscience, and market behaviour — helping traders understand the psychology driving their decisions, not just the strategy.

Read more at thewealthmirror.com/about