The Disposition Effect In Your Own Trading
Traders tend to sell winning positions too early and hold losing positions too long, a well-documented pattern called the disposition effect, driven by the different way gains and losses feel once a position is already open.
Given two open positions of equal quality by any forward-looking measure — one currently up, one currently down — a trader is measurably more likely to sell the winner and hold the loser than the reverse, even when neither position's actual future prospects differ from the other's. This pattern, documented across retail accounts, professional funds, and controlled experiments alike, is called the disposition effect, and it runs directly opposite to what a purely forward-looking process would produce, since nothing about whether a position happened to be bought below or above the current price should change what it's worth going forward.
Why gain-status distorts a decision that shouldn't depend on it
The pull comes from treating each position's outcome relative to its own entry price as a separate, self-contained gamble rather than folding it into an overall portfolio view. A position that's up feels like a "sure gain" sitting there to be locked in, and taking a sure gain feels good in a way that outweighs the marginal expected value of holding for more upside. A position that's down feels like a choice between a certain loss (selling now) and a gamble that might get back to even (holding), and that framing pushes toward holding, since realizing the loss feels worse than the chance, however statistically thin, of avoiding it. Neither of these framings has anything to do with where either position is actually headed from here — they're both anchored to the accident of the entry price.
An analyst running two positions in similar-quality names, one up 8% and one down 8% with no material change in either company's fundamentals since entry, closed the winner to "bank the gain" and kept the loser open, reasoning it would "come back." A forward-looking review of both positions' prospects at that moment gave no reason to treat them differently — if anything, the losing position had gotten objectively cheaper relative to the same fundamentals. The decision to sell one and hold the other tracked entry price, not expected value.
The corrective is to force the same question onto both positions independent of their current gain or loss: given only what I know now, would I put this position on today, at today's price? If the answer is no for the loser, it should be closed regardless of the fact that closing it locks in a loss — and if the answer is yes for the winner, holding it shouldn't feel like refusing free money.
The disposition effect is the tendency to sell winning positions too early and hold losing ones too long, driven by treating each position's gain or loss relative to its own entry price as a separate emotional event rather than judging it on forward-looking merits. The fix is asking the same question of every position regardless of its current P&L: would I open this position today, at today's price, knowing what I know now?
Further reading
- Shefrin and Statman, The Disposition to Sell Winners Too Early and Ride Losers Too Long