Sunk Cost Fallacy

The sunk cost fallacy occurs when you factor past, unrecoverable costs into future decisions. In crypto, it manifests as: "I already lost 60% on this token, so I might as well hold and hope it recovers." The rational approach ignores what you have already lost and asks only: "Given what I know now, is this the best use of this remaining capital?"

Behavioral economics research by Kahneman and Tversky (1979) established that humans feel losses roughly 2.2 times more intensely than equivalent gains — a phenomenon called loss aversion. This makes selling a losing position psychologically painful, even when it is the mathematically optimal decision. The sunk cost fallacy adds a second layer: the feeling that selling "wastes" the money already lost, when in fact that money is gone regardless.

In the 2022 crypto bear market, many investors held Luna, Celsius, and FTX-related tokens well below any rational exit point because of sunk cost thinking. CoinGecko data shows that 53% of tokens that dropped 90% from their all-time high between 2018-2023 continued to lose additional value rather than recovering. Holding because "I've already lost so much" frequently made losses worse.

The antidote is the "clean slate" test: if you woke up tomorrow with your current portfolio value in cash and no memory of your purchase prices, would you buy the same positions at current prices? If not, the only rational action is to sell and reallocate. Your entry price is irrelevant to the asset's future prospects.

Having pre-set stop-losses removes the sunk cost decision entirely. When your stop triggers, you exit automatically — no deliberation, no rationalization. This is why mechanical trading rules consistently outperform discretionary decisions influenced by cognitive biases.