Most traders know losing is part of the game. What they underestimate is how hard the brain fights to avoid acknowledging a loss — and how that fight costs far more than the original losing trade ever would have. The sunk cost fallacy is the mechanism behind that fight, and it quietly destroys more trading accounts than any single bad setup.
What Sunk Cost Actually Means (and Why the Market Doesn’t Care)
A sunk cost is any resource — money, time, effort — that has already been spent and cannot be recovered. The defining feature is that it is irreversible. In trading, the moment you enter a position, the capital you risked becomes a sunk cost. Whether NVDA is at $115 or $98, your purchase price is historical data. The market has no memory of what you paid.
Rational decision-making looks only at forward expected value: given where price is right now, with the information available right now, does holding this position give me a positive expected outcome? The sunk cost fallacy corrupts that question by injecting a backward-looking variable — “but I’m already down $1,400” — that is completely irrelevant to the forward calculation.
Kahneman and Tversky’s Prospect Theory (1979) explains why this is so hard to resist: losses feel approximately 2x more painful than equivalent gains feel pleasurable. A $1,000 loss registers neurologically as twice as bad as a $1,000 gain feels good. The brain therefore does everything it can to avoid converting an unrealized loss into a realized one — including constructing elaborate justifications for holding a trade that has already failed its thesis.
The Disposition Effect: Proof This Is Widespread
This is not a rare personality quirk. In 1998, economist Terrance Odean analyzed 10,000 retail brokerage accounts and published what became one of the most cited papers in behavioral finance. His finding: retail investors sell winning positions 50% more readily than losing ones. They were systematically locking in gains early and riding losses far too long — the exact opposite of “cut losses short, let winners run.”
The label for this pattern is the disposition effect, and it is the sunk cost fallacy in aggregate. Across tens of thousands of trades, the data showed traders making holding decisions based on where they entered, not where the trade was going. The practical consequence is a portfolio that accumulates losers and prematurely exits winners — a recipe for long-run underperformance even when the underlying trade selection is sound.
Three Patterns Where Sunk Cost Appears in Trading
Recognizing the bias means identifying how it shows up in practice, not just in theory.
1. Refusing to exit a broken setup. A trader buys 200 shares of NVDA at $115 after a breakout above that level. Their journal rule says: “Exit if price closes below $110 — max loss $1,000, or 2% of a $50K account.” NVDA gaps down to $108 the next day on sector-wide news. The sunk cost voice says: “I’m already down $1,400 — I can’t sell now, I’ll wait for it to bounce back to $115.” The rational voice says: “The thesis was a breakout above $115. Price is now below $110. The original reason to hold is gone.” Selling at $108 locks in a $1,400 loss. Waiting and selling at $98 — a common outcome when a thesis breaks on fundamental news — locks in a $3,400 loss. The extra $2,000 damage came entirely from sunk cost reasoning.
2. Averaging down as a compounding trap. Some traders respond to a losing position by buying more shares at lower prices to reduce their average cost. If NVDA falls to $105, buying 200 more shares lowers the average to $110 — but it doubles the position size in a trade that is already proving the original thesis wrong. The risk increases precisely when the evidence against the trade is strongest. Averaging down is only rational when it follows a fresh, forward-looking analysis that confirms the trade still has positive expected value with fully defined risk. When the motivation is “I need to lower my cost basis to get out at breakeven,” it is a sunk cost trap.
3. Holding a deteriorating options contract. Options add a dimension of urgency that equities do not have. An at-the-money 30-day option loses roughly 1/30th of its remaining time value per day. With 5 days left, it loses roughly 1/5th per day — decay accelerates as expiration approaches. A trader who buys 5 SPY $540 calls expiring in 21 days for $3.20 each ($1,600 total) and watches them fall to $1.10 ($550 remaining) with 10 days left faces a stark math problem. Theta is stripping approximately $0.08 per contract per day — $4 total per day — from a position that is already directionally wrong. Holding for recovery means paying an accelerating daily decay tax on a trade that has not gone in the intended direction. A pre-entry journal rule — “exit if the position loses 50% ($800)” — would have triggered an exit when contracts hit $1.60, saving $250 in additional decay loss versus the position at $1.10.
The Critical Distinction: Sunk Cost vs. Thesis Still Intact
Not every decision to hold a losing position is a sunk cost error. Sometimes holding is correct. The test is simple: is your reason for holding forward-looking or backward-looking?
Backward-looking (sunk cost): “I’m down $800 and need to recover it.” / “I can’t sell here, I’ll be locking in a loss.” / “It has to come back to where I bought it.”
Forward-looking (legitimate): “Price pulled back to my planned re-entry zone and volume is holding. The original thesis is intact and risk is still defined.” / “Sector news is sector-wide and my company’s fundamentals are unchanged. I’ll hold to my original stop.”
The distinction is not about optimism versus pessimism. It is about what information is actually doing the work in your decision. If removing your entry price from the analysis would change your decision, you are making a sunk cost-based choice.
How a Trading Journal Breaks the Cognitive Loop
The most effective structural fix for the sunk cost fallacy is pre-commitment: write your exit rule before you enter the trade, when you are calm, unemotional, and thinking clearly about expected value. A trading journal makes that pre-commitment binding.
When your journal says “exit if price closes below $110,” that rule was written by a version of you with no emotional stake in the outcome. At the moment the loss hits, you are not making a decision under duress — you are executing a decision already made. The journal effectively creates a second self that overrides panicked-you with rational-you.
This shifts the journal from a record-keeping tool into a real-time bias-breaking mechanism. The rule you wrote at entry — including the specific trigger, the maximum dollar loss, and the percentage of account at risk — becomes a firewall between the sunk cost impulse and the actual order. Reviewing losing trades against those pre-written rules also reveals whether sunk cost reasoning is a recurring pattern in your trading, which is the first step to eliminating it.
For options traders specifically, the pre-entry rule should also account for time: “exit if the position loses 50% OR if 10 days remain and the position is more than 30% underwater” captures both the dollar loss and the decay-acceleration risk that makes holding especially costly near expiry. The options trading journal guide covers additional rules worth logging at entry for contracts positions.
Understanding risk management through journaling and building a consistent stop-loss optimization process both depend on the same foundation: decisions made before emotion enters the picture.
Key Takeaways
- The market does not know or care what you paid. Past capital is irrelevant to forward expected value — only the future distribution of outcomes matters from the current price.
- The disposition effect (Odean, 1998) across 10,000 accounts proves this bias is universal: retail investors hold losers roughly 50% longer than winners, systematically.
- The test for sunk cost thinking is simple — if removing your entry price from the analysis would change your decision, you are making a backward-looking choice.
- Averaging down is only rational when it follows a fresh assessment of forward expected value with defined risk. When the motivation is cost basis reduction, it compounds risk at exactly the wrong moment.
- Options traders face an additional theta decay cost for holding losing positions near expiry — decay accelerates, making a bad hold geometrically worse over the final 30 days.
- Pre-defined exit rules logged at trade entry separate the decision from the emotion. Rational-you at entry overrides panicked-you mid-trade.
JournalPlus lets you log your exit rules, position sizing rationale, and maximum loss thresholds at trade entry — before the position moves against you. When your pre-written rule triggers, the decision is already made. At $159 one-time with lifetime access, it is a permanent structural fix for one of trading’s most expensive behavioral patterns. Learn more about how options traders use JournalPlus to manage bias-driven holding decisions.
People Also Ask
What is the sunk cost fallacy in trading?
The sunk cost fallacy is the tendency to hold a losing position because of money already invested, rather than evaluating whether the trade still has forward-looking merit. The market does not care what you paid — only future expected value matters.
How does the disposition effect affect retail traders?
Terrance Odean's 1998 study of 10,000 brokerage accounts found that retail investors sell winning positions 50% more readily than losing ones. This asymmetry — holding losers too long while cutting winners too early — directly damages long-run performance.
Why is the sunk cost fallacy worse for options traders?
Options lose time value daily through theta decay. An at-the-money 30-day option loses roughly 1/30th of its remaining time value per day, and that rate accelerates near expiry. Holding a losing options position while waiting to 'get back to even' compounds the loss geometrically.
How do you avoid the sunk cost fallacy in trading?
Set your exit rules before entering the trade — while you are calm and unemotional. Log them in a trading journal so the rule is binding. When the trigger is hit, the decision has already been made by a rational version of yourself.
Is averaging down always a sunk cost mistake?
Not always, but it usually is. Averaging down is a sunk cost trap when the reason for adding shares is to lower your cost basis rather than a fresh, forward-looking assessment that the trade still has positive expected value with defined risk.
