Short answer: the sunk cost fallacy in trading is holding, adding to, or widening a losing position because of what it has already cost you. The market no longer supports the trade; the money already spent is doing the arguing. Whatever you paid in pips, fees and hours is gone whether you close now or later. The only question that matters is whether you would open this exact position if you were flat right now. If the honest answer is no, the spent amount is a reason to leave, not a reason to stay.
What the sunk cost fallacy is, in plain trading terms
A sunk cost is money or effort that no future action can recover. The 40 pips already given up on a long EUR/USD position are sunk. So are the hours spent building the thesis and the swap fees and costs paid overnight. None of that changes what price does next. The fallacy is treating those spent amounts as a stake that must be defended. Some readers search for "sink cost" instead. The idea is the same: the cost has sunk, and you cannot pull it back.
Who is this page for?
Anyone trading forex, indices, metals or crypto as CFDs who has stared at a red position and thought "I have come too far to close now". It is also for anyone who adds to losers for a better average and then wonders why one trade ate a month of ordinary gains.
How spent money hijacks the next decision
The mechanism is simple and stubbornly human. A closed loss feels final, while an open loss keeps its ambiguity and its hope. Closing converts that ambiguity into a blunt number, and the brain treats the conversion as the moment the money is lost. So the trader waits. The longer the wait, the larger the sunk amount and the stronger the pull to justify it. Effort compounds the problem. A position you researched for days feels more deserving than one you clicked on in a minute. Price does not know the difference.
Costs quietly deepen the trap. Every extra night in a losing CFD position adds financing charges, and every averaging-down order pays spread and commission again. A trader who refuses to accept a 1 percent loss may pay a further 0.3 percent in fees and costs just to postpone it. Every held loser also eats a growing share of the daily risk budget. The link to the wider process matters too. The hesitancy that keeps one loser open tends to show up as a disorderly response to a losing streak. Several small sunk costs then merge into one precarious drawdown.
Three scenarios where traders hold losers
The swing trader and the stale thesis. A trader shorts gold at 2,400 expecting resistance to hold. Price pushes through to 2,430, so the original reason is now plainly false. The trader holds anyway, because the thesis took a week to build and closing would make that week feel wasted. The stop moves to 2,460. The position is now defended for the wrong reason.
The scalper who averages down. A scalper buys a US index CFD at 5,200 with a 10-point stop. Price drops 10 points. Instead of closing, the scalper adds a second unit at 5,190 to halve the distance to breakeven, then a third at 5,180. A trade meant to risk 10 points on one unit now risks 30 points across three. The reluctance to leave grew with each add.
The weekend crypto hold. A trader is long a BTC CFD at 68,000, down roughly 3 percent on Friday afternoon. Closing feels like locking in a lopsided outcome after a good week, so the position stays open and pays financing every night. The Monday gap takes it to 64,000. The weekend hold was not a market view; it was an attempt to avoid making the loss real.
Common mistakes that keep a sunk cost alive
| Mistake | What it sounds like | What it actually does |
|---|---|---|
| Moving the stop wider | "Just a bit more room" | Converts a defined loss into an undefined one |
| Averaging down without a plan | "Better average entry" | Multiplies size exactly where the idea is failing |
| Turning a day trade into a swing | "I will hold until it comes back" | Replaces a tested timeframe with hope |
| Counting hours of analysis as money | "I worked too hard on this to close" | Defends effort that price cannot see |
| Waiting for breakeven to exit | "I will close when I get back to flat" | Ties the exit to your entry price instead of to the market |
The breakeven wait is the most seductive, because it feels modest. You are not asking for a profit, just to get back what you spent. But the market does not owe you a return trip, and capital tied up waiting for one cannot work elsewhere.
A worked number example: hold or close
Picture an account of 10,000 with a written limit of 1 percent risk per trade. A long index CFD position is sitting at the stop, 100 in the red, exactly the planned 1 percent.
Close now: the loss is 100, the account is 9,900, and the next trade starts clean with a fresh 1 percent.
Hold and widen the stop to 2 percent: if price keeps falling, the loss becomes 200 plus roughly 5 in extra spread and overnight charges. The account is 9,795. The next winning trade at 1.5R returns about 147 rather than making you whole. It now takes two clean wins instead of one to climb back above 10,000. Holding did not change the odds of the trade. It only made the hole needlessly deeper and the recovery slower. Widen the stop on two trades out of ten in a month and the extra 200 wipes out more than one full winner.
A checklist to cut a position without flinching
The fix is not willpower. It is a short process that removes the spent amount from the decision.
- Write the invalidation level before entry, and treat it as the point where the idea is wrong, not where the pain is too much.
- Ask the flat question: if you had no position, would you buy here at this size? A no means close.
- Set a maximum of one add per trade, sized in advance, or ban adds for a month and compare results.
- Record the hours of analysis in your journal, then note that they are worth zero at exit time. Seeing it written down is awkward and useful.
- Price the delay: estimate the spread, commission and overnight charge of one more day. Would you pay that to open the trade fresh?
- If you catch yourself widening a stop, close half immediately. Halving size breaks the all-or-nothing bargaining loop.
This is tedious at first and seldom feels heroic. That is the point. Cutting a loser on schedule is the humble, thankless core of trading discipline, and it beats any clever entry.
Sunk cost inside a prop-firm evaluation
Evaluation accounts punish the sunk cost fallacy harder than a personal account does, because the limits are unforgiving and fixed. A daily loss cap that resets each session is scarcely a threat to a trader who closes losers on time. It is lethal to one who widens stops. A single held position can squander the day's allowance and part of the total drawdown in one move. Pivex builds its one-step evaluation around a total drawdown that stays fixed rather than trailing your gains, plus a daily cap that starts over each morning. Nothing forces you to finish by a date or to trade a set number of days. That removes a common excuse for holding: "I must get back to green before the clock runs out." The built-in assistant tracks your rules and limits as you trade, so the remaining room is visible rather than imaginary.
The other evaluation-specific trap is the fee. The one-time charge you paid for the attempt is itself a sunk cost, and traders sometimes hold a loser to "protect the fee". That gets the logic backwards. At Pivex the one-time charge is refundable, so finishing cleanly protects it and rescuing one trade does not. If the fallacy escalates into revenge-sized positions, read how revenge trading starts and how to interrupt it before the next session.
Next step: open your journal and mark every trade from the last month where the stop was widened or a unit was added after a loss. Total the extra loss those decisions caused, including fees and costs, and compare it with your best winner. If the number is larger, the sunk cost fallacy is already your most expensive habit. Build the checklist above into your pre-trade routine first. Then test it inside a rulebook that makes limits visible, such as the Pivex one-step evaluation on Match-Trader.
Related guides: how to handle a losing streak in trading and revenge trading.
Before participating, review Pivex's current program rules and decide whether a simulated proprietary trading evaluation fits your experience and risk limits. Participation does not guarantee a reward, income, or any particular trading result.