Skip to main content

Holding Losers Too Long

Updated Jun 23, 2026 at 8:47pm

Research Draft Medium 1,041 words

Holding losers too long is the "ride losers" half of the disposition effect — the documented tendency to keep losing positions open in the hope they return to breakeven, while the same trader readily takes profits on winners. It is arguably the single most common account-killer in discretionary trading because the payoff is asymmetric: a winner closed early caps an opportunity, but a loser held without a stop has no floor — one un-cut position can erase the gains from many disciplined wins. The behavior is irrational from a forward-looking standpoint (the entry price is a sunk cost the market does not care about), yet it is deeply natural, driven by how human beings experience loss and the threat of admitting a mistake.

The behavioral drivers

  • Loss aversion (prospect theory). Kahneman & Tversky's prospect theory holds that the pain of a loss is felt more intensely than the pleasure of an equivalent gain. Their 1992 cumulative-prospect-theory estimate puts the loss-aversion coefficient at λ ≈ 2.25 — losses loom roughly twice as large as same-size gains. The trader is risk-averse over gains (grabs the sure profit) but risk-seeking over losses (gambles to avoid the sure loss), which is exactly the recipe for cutting winners and riding losers.
  • Avoiding realizing the loss / admitting being wrong. An open loser is a "paper" loss; closing it converts it into a realized loss and a documented mistake. Regret and the threat to self-image make traders defer the close — "it's not a loss until I sell."
  • Get-evenitis. Fixation on returning to the entry price ("I'll sell when it gets back to breakeven"), which substitutes the trader's cost basis for any actual judgment about the stock's prospects.
  • Anchoring / sunk-cost. The entry price becomes a psychological anchor and the money already lost feels like an investment to be "protected," even though it is gone and irrelevant to the next dollar's expected value.

How it shows up

Held losers tend to grow rather than resolve: the stop that "felt too tight" is widened, then ignored; the thesis is quietly re-written to fit the price ("now it's a long-term hold"); attention drifts from is this position still valid to how far underwater am I. The dangerous escalation is averaging down — buying more of the loser to lower the average cost. This is the cousin of holding too long and is more lethal: it turns a small, survivable loss into a large, position-threatening one and concentrates capital into the very position the market is rejecting. (Averaging into a planned scaled entry is different and intentional; averaging down to rescue a broken thesis is the disposition effect with leverage.)

The evidence

  • Shefrin & Statman (1985) coined the term disposition effect in "The Disposition to Sell Winners Too Early and Ride Losers Too Long," framing it via prospect theory plus mental accounting, regret aversion, and self-control. They showed tax considerations alone cannot explain the pattern (taxes argue for the opposite — realizing losses).
  • Odean (1998), "Are Investors Reluctant to Realize Their Losses?" measured it directly across ~10,000 accounts at a large U.S. discount brokerage (1987–1993). Investors realized gains at a meaningfully higher rate than losses — the proportion of gains realized (PGR ≈ 14.8%) ran well above the proportion of losses realized (PLR ≈ 9.8%) outside December, i.e. winners were roughly 50% more likely to be sold than losers (about 1.5×). Crucially, the behavior was not justified by performance: the winners investors sold subsequently outperformed the losers they kept by about 3.4 percentage points over the following year. They sold the wrong stocks. (The effect weakened in December, when tax-loss selling pulls the other way.)

The disposition effect is one of the most robustly replicated findings in behavioral finance — confirmed across retail brokerage data, experiments, and international markets — though its precise cause (pure prospect theory vs. realization utility, beliefs about mean-reversion, or rational rebalancing in some cases) is still debated in the literature. What is not seriously contested is the pattern and its cost to returns.

The fix (decision-useful for swing trading)

The remedy is to remove the decision from the moment of pain and pre-commit to it:

1. A pre-committed, mechanical stop, placed at entry. Define the price that proves the thesis wrong before you put the trade on, and place a hard stop there. A hard resting stop cannot be negotiated with; a "mental stop" is the disposition effect's favorite loophole. 2. Position-size so any single stop-out is survivable. Size the trade off the entry-to-stop distance and a fixed per-trade risk budget so that being stopped out is a non-event, not a catastrophe you're tempted to avoid. This is what makes taking the loss psychologically affordable. 3. Judge the position on its thesis, not your entry price. The honest test is "would I open this position here, today, at this price?" If no, the unrealized loss is irrelevant — close it. The market does not know or care what you paid. 4. Never average down to rescue a thesis. A broken setup is reason to exit, not to add. Reserve scaling-in for entries you planned in advance.

System relevance

This node is the loss-side mirror of Cutting Winners Early (same disposition-effect bias, opposite tail) — the two together describe a P&L distribution with truncated upside and a fat left tail, the exact opposite of the favorable skew swing trading depends on. Operationally it pairs with Stop-Loss Strategies (the mechanical fix) and position sizing / risk-per-trade (what makes the stop bearable). For the Augustus trade-setup agent the hard caveat is: a setup's expectancy is only valid if the stop is honored as written — any logic that widens, removes, or averages-down against a defined stop invalidates the trade's risk model and should be treated as a disqualifying behavior, not a discretionary call.

Sources