Watch how most people manage a portfolio and you'll see the same pattern. A position goes green and they sell it fast to bank the win. A position goes red and they hold, waiting for it to come back. Hersh Shefrin and Meir Statman put a name on this in 1985, the disposition effect, and it has shown up in basically every market and every type of participant studied since. Retail traders have it worst, but even professional fund managers do it, just a bit less.
The engine underneath is prospect theory, Daniel Kahneman and Amos Tversky's model of how people weigh gains against losses. The short version is that we get risk-averse once we're in profit, preferring a sure gain over a bet with a higher expected payoff, and risk-seeking once we're in the red, preferring the gamble over booking a certain loss. Point that at a trading account and you get exactly what you'd expect: winners get cut early, losers get nursed in the hope they recover.
What it actually costs
The bill is real. Studies keep finding that the stocks people sell go on to beat the stocks they keep by several percentage points over the next year. Momentum is a genuine factor, and someone selling winners while holding losers is systematically betting against it. They're trimming their strongest positions and stockpiling their weakest.
Taxes make it worse. In a taxable account the smart move is to realize losses so you get the deduction and defer gains so you push the tax bill out. The disposition effect flips that. You crystallize gains, which triggers tax now, and you sit on losses, which throws away the write-off. The after-tax drag adds up over a few years.
Crypto shows the same thing, except you can watch it happen on-chain. Wallets that sell in profit and freeze up when underwater are everywhere. That creates predictable selling pressure at levels where a cluster of buyers just went green, and stubborn holding at levels where a cluster is still underwater. When we look at cohort data on Blockcircle those price zones are some of the more readable parts of a chart, because the behavior behind them is so consistent.
Rules that override the instinct
You don't fix this with willpower, you fix it with rules that fire before your emotions get a vote. A few that work:
- Trailing stops, so a winner keeps running and only gets cut when the trend actually reverses by a set amount, not the second you're up.
- Minimum holding periods, so you can't dump a position the day it turns green.
- Hard stop losses, so a loser gets closed before it grows big enough to flip you into that risk-seeking, hope-it-comes-back mode.
One mental reframe does a lot of the work too. Look at every position as if you held none of it. A stock is down 30% from where you bought, and the real question isn't "should I hold and hope." It's "if I had cash right now, would I buy this at today's price knowing what I know today." If the answer is no, you sell, and what you paid for it doesn't enter into it.
Portfolio-level rules help for the same reason. Something like "close any position that's been underwater 60 days with no improvement" gives you an automatic override for the urge to hold losers forever. It won't be right every time, and it doesn't have to be. It just has to beat the default, which is holding way too long, and that's a low bar to clear.