Disposition Effect in Trading: Stop Selling Winners and Excusing Losers

Disposition Effect in Trading: Stop Selling Winners and Excusing Losers

বাংলায় পড়ুন

Learning objective

By the end of this guide, you will be able to recognise the disposition effect, separate a position's purchase price from its current investment case, and build exit rules that apply the same evidence standard to winners and losers.

The disposition effect is the tendency to realise gains too readily while holding losing positions too long. A small profit feels worth protecting; selling a loss makes the mistake feel final. The result can be an inconsistent portfolio: strong positions are removed because they are green, while weak positions survive because they are red.

The idea was formalised by Hersh Shefrin and Meir Statman in 1985. Terrance Odean later studied 10,000 brokerage accounts and found that investors realised winning shares at a substantially higher rate than losing shares, except around tax-motivated year-end selling. Historical research does not prove that every losing asset should be sold or every winner held. It identifies a behavioural pattern that a trading process should test.

What the disposition effect is—and is not

A position showing a loss is not automatically bad. A winner is not automatically good. Price can move temporarily against a valid thesis, and a profitable trade can become overvalued or violate its exit rule.

The bias appears when the gain or loss relative to the entry price becomes the main reason for the decision:

• “I will sell this winner before the profit disappears.”
• “I will hold this loser until I get back to break-even.”
• “It is not a real loss until I sell.”
• “The winner has already gone too far, but the loser is now cheap.”

None of these statements evaluates expected return, risk, opportunity cost or the original thesis. The market does not know the trader's purchase price.

Step 1: define the exit before entry

Record four separate exit paths:

1. Thesis invalidation: evidence contradicts the reason for the trade.
2. Price-risk exit: a level or volatility condition limits loss.
3. Profit-management exit: a target, trailing rule or valuation condition applies.
4. Time exit: the expected catalyst or move fails to occur by a deadline.

Apply these rules whether the position is green or red. A thesis invalidation does not become less important because selling would realise a loss. A valid trend-following rule does not become less valid because the open profit feels large.

Step 2: hide the purchase price during review

Run a “clean-sheet” test:

“If I had no position, would I open this trade now at the current price, using the same thesis and risk rules?”

This question is not a command to sell whenever the answer is no. Existing positions may have different transaction costs, taxes or portfolio roles. It exposes whether the entry price is controlling the analysis.

During the first review, examine current price structure, fundamentals, catalysts and portfolio exposure without looking at unrealised P&L. Add P&L, tax and cost considerations afterwards.

Step 3: use identical evidence standards

Create a review table:

Question	Winner	Loser
Is the original thesis intact?	Yes / No	Yes / No
Has invalidation occurred?	Yes / No	Yes / No
Is expected reward still worth the risk?	Estimate	Estimate
Is a better use of capital available?	Yes / No	Yes / No
Has the planned exit triggered?	Yes / No	Yes / No

If a 10% decline would invalidate a winner, it should also invalidate a loser under the same setup. If a winner is held because fundamentals improved, document that evidence; do not hold merely because momentum feels good.

Step 4: measure opportunity cost

Holding a losing position is an active choice. Capital remains tied up, portfolio risk remains allocated and attention remains consumed.

Compare three choices from today—not from the entry date:

• keep the position;
• reduce or exit it;
• reallocate to cash or another qualified setup.

Use expected scenarios rather than the hope of break-even. Recovering to the purchase price may require a move that has no special economic meaning.

Worked example

Assume a hypothetical trader owns two positions, each currently worth £4,000:

• Share A: bought for £3,600, now showing an £400 gain;
• Share B: bought for £5,000, now showing a £1,000 loss.

The trader needs to free £4,000 and instinctively wants to sell A because it is profitable.

The rules-based review shows:

Evidence	Share A	Share B
Earnings trend	Improving	Deteriorating
Guidance	Raised	Cut
Price structure	Above trailing exit	Below invalidation level
Next catalyst	Defined in four weeks	No dated catalyst
Thesis	Intact	Invalidated

The original purchase prices are irrelevant to this comparison. Under the stated rules, B is the exit candidate because its thesis and price invalidation have failed. Selling B realises a loss, but holding it would be a new £4,000 decision today.

Now change the facts: suppose B's thesis remains intact, the decline is within tested volatility and A has reached a pre-defined valuation target. Selling A may then be correct. The lesson is not “always sell losers”; it is “use current evidence and pre-written rules rather than the colour of P&L”.

Step 5: audit realised gains and losses

Once a month or quarter, calculate:

• average holding period for winners and losers;
• percentage of winning positions sold;
• percentage of losing positions sold;
• rule compliance for each exit;
• subsequent performance after the exit, using a fixed review horizon.

Odean's research compared the proportion of gains realised with the proportion of losses realised. A personal journal can use a simpler version: when an exit opportunity occurred, did the trader sell winners more readily than comparable losers?

Do not optimise behaviour from a tiny sample. A strategy may intentionally have many small losses and a few long winners, or it may be mean-reverting. Judge exits against the documented strategy, not a universal ratio.

Step 6: separate tax decisions from trading decisions

Taxes can legitimately change which position is sold. Rules differ by country, account type and holding period. Record tax logic explicitly and obtain qualified advice where needed.

Do not use vague tax language to protect a failed trade. First state the pre-tax investment decision, then show how tax and transaction costs change implementation.

Common mistakes

Moving the stop to avoid a realised loss

Changing risk after the thesis fails converts a planned trade into an unplanned holding.

Taking profit only because it is visible

An unrealised gain can reverse, but that alone is not an exit rule. Use the strategy's target, trail or thesis condition.

Averaging down to reach break-even faster

Adding capital increases exposure. It requires a fresh thesis, total-risk calculation and pre-defined rule.

Calling every loser a long-term investment

A timeframe change after entry is a new decision. Reassess from zero.

Ignoring portfolio concentration

Keeping several correlated losers may create one large hidden bet.

Treating research as a command

The disposition effect is a population-level pattern, not proof that any specific holding should be sold.

Practical checklist

• Write thesis, invalidation, profit and time exits before entry.
• Review current evidence before looking at unrealised P&L.
• Ask whether you would open the position today.
• Apply the same evidence standards to winners and losers.
• Compare holding with cash and other qualified setups.
• Document tax and transaction-cost effects separately.
• Never widen risk solely to avoid realising a loss.
• Audit exit-rule compliance across both winners and losers.

Sources

The concept was developed in Shefrin and Statman's 1985 paper, The Disposition to Sell Winners Too Early and Ride Losers Too Long. Brokerage-account evidence comes from Terrance Odean's 1998 study, Are Investors Reluctant to Realize Their Losses?. The US SEC's Investor Bulletin on behavioural patterns provides broader investor context.

This tutorial was prepared with AI-assisted research and writing under The Money Man editorial workflow. It is educational information, not personalised investment or tax advice. The example is hypothetical, and no exit rule guarantees a return.