Loss Aversion and the Disposition Effect: Why Traders Hold Losers and Sell Winners

There is a pattern that appears in almost every new trader's history: a string of small wins, each closed quickly "to lock it in", followed by one losing trade that is held and held — past the planned stop, past a "it'll come back" — until it wipes out all the small wins and more. It is so common that behavioural economists have names and decades of research for it.

This guide explains loss aversion and the disposition effect, the research that established them, how they show up in day-to-day trading decisions, and — most usefully — the practical techniques that stop them from running your account.

Loss aversion: losses loom larger than gains

In 1979, psychologists Daniel Kahneman and Amos Tversky published Prospect Theory: An Analysis of Decision under Risk in the journal Econometrica. It challenged the standard economic assumption that people weigh outcomes purely by their final wealth. Instead, they found:

• People think in gains and losses relative to a reference point — often what they paid, or where they started — rather than in total wealth.
• Losses hurt more than equivalent gains please. Losing $100 feels worse than winning $100 feels good.
• People are risk-averse for gains and risk-seeking for losses. Offered a sure gain or a gamble, most take the sure gain. Facing a sure loss or a gamble that might avoid it, many take the gamble.

In a 1992 follow-up, Advances in Prospect Theory, Tversky and Kahneman estimated that for typical participants, losses weighed about 2.25 times as much as gains of the same size. Exact values vary between people and situations, but the asymmetry is one of the most consistently replicated findings in behavioural economics. Kahneman was awarded the Nobel Memorial Prize in Economic Sciences in 2002, in part for this work.

Why it matters in trading

Combine the two halves of prospect theory and you get a predictable trading pattern:

• In a winning trade (a gain), people become risk-averse — they want to secure the gain, so they close early.
• In a losing trade (a loss), people become risk-seeking — they would rather gamble on a recovery than accept a certain loss, so they hold, or even add.

The result is small wins and large losses: the exact opposite of the profile that most viable strategies need.

The disposition effect

In 1985, Hersh Shefrin and Meir Statman published The Disposition to Sell Winners Too Early and Ride Losers Too Long in The Journal of Finance, giving the pattern its name: the disposition effect. They linked it to prospect theory, to a reluctance to admit mistakes, and to the regret people feel when they close a position at a loss.

In 1998, Terrance Odean tested it using the trading records of 10,000 accounts at a large US discount broker, in Are Investors Reluctant to Realize Their Losses?. He found that investors showed a strong preference for selling winners rather than losers — and that the behaviour was costly. The winning stocks they sold went on, on average, to outperform the losing stocks they kept, by around three percentage points over the following year. Selling the winners and keeping the losers was, in aggregate, the wrong way round.

Related effects

• The break-even effect. In a 1990 study in Management Science, Richard Thaler and Eric Johnson found that people who have lost money are often attracted to gambles that offer a chance to get back to even. In trading, this is the urge to hold or add to a loser "until it gets back to where I bought it".
• The house-money effect. The same research found that after gains, people may take more risk with "house money" — profits they feel weren't really theirs. In trading, this shows up as increasing size after a winning streak.

How these biases show up in your trades

Look for these patterns in your own history:

If your average loss is larger than your average win while your win rate is high, the disposition effect is the first suspect.

The cost, in numbers

Suppose a trader's plan risks 1R per trade (one unit of risk) and targets 2R. Following the plan with a 40% win rate gives an expectancy of (0.4 × 2) − (0.6 × 1) = +0.2R per trade.

Now suppose loss aversion takes over. The trader closes winners at an average of 0.8R out of fear of giving them back, and lets losers run to an average of 1.5R before giving up. The win rate rises a little — say to 50%, because early exits catch more small wins — but expectancy becomes (0.5 × 0.8) − (0.5 × 1.5) = −0.35R per trade.

Same strategy, same entries. The difference is entirely in the exits, and it turns a positive edge into a reliable loss. Our risk management guide explains expectancy in more detail.

How to counter loss aversion

You cannot switch off a bias that is built into how people evaluate outcomes. You can design your trading so that it has fewer opportunities to act.

1. Decide exits before entry

Write the stop-loss and target before you enter, and place them with the order. A decision made before the trade is made in calm, gain-and-loss-neutral conditions. A decision made during the trade is made under the full force of the bias.

2. Make stops one-directional

Adopt a rule that stops may only move in the direction of the trade — to reduce risk or lock in profit — never further away. Write it in your trading plan.

3. Size positions so losses are bearable

The larger a potential loss feels, the harder it is to accept. Risking a small fixed fraction of your account per trade makes each stop-out tolerable, which makes following the plan realistic.

4. Think in R and in samples

Express every result as a multiple of the initial risk (R) and judge performance over dozens of trades, not one. A −1R loss is the planned cost of a trade that didn't work; it is not a verdict on you. This reframing directly weakens reference-point thinking.

5. Use mechanical exits for winners

If you tend to close winners early, use a pre-set target, a trailing stop or a rule such as "close half at 2R and trail the rest". Then stop watching the trade tick by tick. Price alerts let you step away without losing track.

6. Separate the decision from the outcome

Review each trade by asking "did I follow the plan?" rather than "did I make money?". A loss taken at the planned stop is a good decision. A loss held past the stop that happened to recover is a bad decision with a lucky outcome — and it trains the habit that will eventually be expensive.

7. Track the bias directly

In your trading journal, record for every trade:

• Planned exit versus actual exit.
• Whether the stop was moved, and which way.
• Your emotional state when you exited.

After a few weeks, compare your average win and average loss with the plan. The gap is the measurable cost of the bias.

Loss aversion beyond the single trade

The same bias appears at larger scales, too:

• Sticking with a failing strategy. After months of effort, abandoning a strategy can feel like realising a loss on the time invested. Traders keep trading approaches that their own data shows are not working, because stopping would make the loss "real" — a version of the sunk-cost trap.
• Refusing to pause after a drawdown. Stepping back after a large drawdown feels like locking in the loss, so traders keep going, often with larger size, to win it back.
• Avoiding the journal. Reviewing losing trades in detail is uncomfortable, so some traders stop recording them. That removes the very information needed to improve.

The fix is the same at every scale: decide in advance what evidence will make you change course — a number of trades, a drawdown level — and follow that rule when the time comes.

A quick self-test

Answer honestly, using your last twenty trades:

• Is your average losing trade larger than your planned 1R loss?
• Is your average winning trade smaller than your planned target?
• Have you moved a stop-loss further away at least once?
• Have you added to a position that was losing?
• Do you feel relief, rather than neutrality, when you close a trade at break-even after it was losing?

Each "yes" suggests loss aversion is influencing your exits. That is normal — it affects almost everyone — and each one points to a specific rule you can add.

A worked example

A trader buys gold at a support zone with a stop 15 points below and a target 30 points above (a 1:2 plan). Price rises 10 points, dips 5, rises again. Anxious about losing the gain, the trader closes at +10 — a third of the plan.

The next trade goes against them. As price nears the stop, the trader thinks "it's just a wick" and moves the stop 15 points lower. Price keeps falling and they finally close at −30 — twice the planned loss.

Two trades, both "reasonable" in the moment. Net: −20 points, where following the plan might have produced +15 (if the first trade reached its target and the second was stopped out). Neither trade's analysis was wrong; both exits were driven by loss aversion.

Key takeaways

• Prospect theory shows that people judge outcomes as gains and losses and feel losses more strongly — roughly twice as strongly in Tversky and Kahneman's estimates.
• The disposition effect — selling winners too early and riding losers too long — is well documented in real investor data.
• The bias produces small wins and large losses, which can turn a positive strategy negative.
• Counter it by design: set exits before entry, only move stops in your favour, size positions so losses are tolerable and think in R over many trades.
• Track planned versus actual exits in your journal to see the bias's real cost.

Continue with FOMO and revenge trading, or practise rule-based exits on a free RumiEX demo account.