Traders cut winners early and let losers run because closing a position is treated as a verdict rather than a decision. Banking a gain registers as proof you were right; closing a loss registers as admitting you were wrong. So the profitable trade gets closed to lock in the feeling, and the losing one gets held to postpone it. The market is not consulted either way.
This has a name — the disposition effect — and it is among the most consistently replicated findings in behavioural finance. It is also the single behaviour most capable of turning a genuinely profitable strategy into a losing account, because it damages the part of the equation that does not survive damage: the size of your average win relative to your average loss.
What the brokerage data actually shows
Terrance Odean examined the trading records of 10,000 accounts at a large discount brokerage and measured the tendency directly. He compared the proportion of gains realised (PGR) against the proportion of losses realised (PLR) — that is, of all the winning positions a trader could have sold, how many they did sell, versus the same figure for losers.
Across the full year, PGR was 0.148 and PLR was 0.098 (Odean, “Are Investors Reluctant to Realize Their Losses?”, Journal of Finance vol. 53 no. 5, 1998). Winners were sold roughly one and a half times as readily as losers. The gap was not marginal; the paper reports the null hypothesis being rejected with a t-statistic above 35.
The defence people offer is that they are holding losers because they expect them to turn around. Odean tested that too, and the same paper answers it plainly: “For winners that are sold, the average excess return over the following year is 3.4 percent more than it is for losers that are not sold.” The positions being cut were, on average, the better ones. The positions being nursed were the worse ones. The behaviour was not merely uncomfortable, it was pointed the wrong way.
Why the brain does this
The mechanism is that an open position feels provisional and a closed one feels final. While a trade is open, a loss is still hypothetical — it can still come back, and nothing has been conceded. Closing it converts a paper loss into a fact about you. Holding on is not really a forecast about the instrument; it is a way of deferring a verdict.
The same logic runs in reverse on the winning side. An unrealised gain feels fragile, and the fear is of watching it evaporate and having been right for nothing. Closing early converts it into something nobody can take back. In both cases the position is being managed for how it will feel to close, not for what price is likely to do next.
Notice what this shares with the wider family of biases in trading psychology: the last result, rather than the current chart, is setting the terms. It is the same failure that produces revenge trading, wearing calmer clothes.
What it does to your numbers
The reason this bias is so destructive is that it attacks reward-to-risk from both ends at once, while leaving your win rate looking healthy.
Suppose you plan trades at 1% risk targeting 3% — a 3-to-1 setup. Now apply the bias. Winners get closed around 1% because that is where holding starts to feel greedy. Losers get held past the stop, so instead of losing 1% they lose 2%. Your intended 3-to-1 has become roughly 0.5-to-1, and it happened without a single change to your analysis.
| As planned | As traded with the bias | |
|---|---|---|
| Average win | 3.0% | 1.0% |
| Average loss | 1.0% | 2.0% |
| Reward-to-risk | 3.0 to 1 | 0.5 to 1 |
| Win rate needed to break even | 25% | 67% |
The last row is the one that matters. A strategy that only needed to be right one time in four now needs to be right two times in three, and nothing about the strategy changed. This is why win rate is such a misleading number — a trader in the right-hand column can post a perfectly respectable win rate while losing money steadily, and will usually conclude they need better entries when what they actually need is better exits. Trading expectancy is the number that makes this visible.
The tells, in real time
- You are watching unrealised profit and loss rather than the chart.
- You moved a stop further away, and can name a feeling but not a level that justified it.
- You closed at a round number, or at “back to break even,” rather than at a level you had marked.
- You are looking for reasons the losing position will recover, having found none before you entered.
- You describe the trade as “still fine” while it sits below your written invalidation.
- You sold a winner and immediately started watching it, which suggests you did not think the trade was finished.
The rules that reverse it
You cannot fix an asymmetry in how outcomes feel by trying to feel differently about them. You fix it by deciding the exits before the position exists, when neither outcome has any emotional weight.
- Write the targets and the invalidation before the entry. This is the whole fix in one line. An exit chosen in advance is a plan; an exit chosen while staring at a moving number is a mood. See what invalidation means for how to define one properly.
- Take partial profits at the first target. This is the practical answer to the urge to bank something, because it satisfies it without ending the trade. How to take partial profits covers the mechanics and the common errors.
- Trail the stop behind the remainder. Once the runner is protected at or above break-even, holding stops being a test of nerve and becomes a mechanical process. What is a trailing stop explains the ways to set one.
- Never widen a stop. Not once. Moving a stop away from price is the exact behaviour Odean measured, executed by hand. If the invalidation was wrong, that is a lesson for the journal, not an adjustment for the live trade.
- Use alerts, and stop watching the balance. Set alerts at your levels and close the profit and loss window. Most premature exits are triggered by watching a number rather than a chart.
- Track average win against average loss in your journal. Not win rate. If your average win is not comfortably larger than your average loss, this bias is present regardless of what you believe about your discipline. Your trading journal settles the argument with evidence.
Frequently Asked Questions
What is the disposition effect?
The disposition effect is the documented tendency to sell winning positions too early and hold losing positions too long. It happens because closing a trade converts a paper result into a settled one, so taking a gain registers as being right while taking a loss registers as being wrong. The position is being judged on how it feels to close rather than on what it is likely to do next.
How much does cutting winners early actually cost?
Terrance Odean measured it across 10,000 brokerage accounts and found investors realised gains at a rate of 0.148 against a rate of 0.098 for losses, meaning winners were sold roughly one and a half times as readily as losers. He also found the winning positions that were sold went on to earn about 3.4 percent more over the following year than the losing positions that were kept. The behaviour was not just uncomfortable, it was backwards.
How do I stop selling my winners too early?
By deciding where you will exit before the position is open, so the exit is a plan rather than a reaction to an unrealised number. The rules that work are mechanical: write targets in advance, take partial profits at the first target so the urge to bank something is satisfied without closing the whole position, and trail the stop behind the remainder rather than watching profit and loss. Set alerts at your levels and stop watching the tick-by-tick balance.
Is it ever right to close a winner early?
Yes, when something in the market changes rather than something in your feelings changes. A structure break against your position, a scheduled news event you did not plan to hold through, or a target that has become unreachable are all legitimate reasons to exit ahead of plan. The test is whether you can name what changed on the chart. If the only thing that changed is that the position is now green, that is the disposition effect rather than analysis.
Bottom line
Cutting winners and holding losers is not impatience paired with stubbornness; it is one bias producing both symptoms, and it has been measured across tens of thousands of real accounts. It is dangerous because it wrecks your reward-to-risk while leaving your win rate intact, so the damage hides behind a statistic that looks fine. The correction is not willpower, it is sequence: write the targets and the invalidation before the entry, scale out at the first target, trail the rest, and never move a stop away from price. Read risk-to-reward ratio explained for the arithmetic this protects, or how to set a stop loss for placing the level you are then going to honour.