A share's purchase price works like a peg driven into the ground on the day it was bought. Every later price gets measured against that peg, whether or not the peg has anything to do with where the price goes next. Psychologists call this leaning on a fixed number anchoring, and it quietly shapes what happens after.
A profit that has not been sold yet works like the score at half-time. It is real, in the sense that it is on the board, but the game is not finished and nothing has been paid out. Selling is the final whistle. It turns a number on screen into money in the bank, or into a loss that cannot be undone.
Research by psychologists Daniel Kahneman and Amos Tversky found that losing money hurts roughly twice as much as gaining the same amount feels good. A lost £20 note stings more than a found £20 note lifts the mood. This imbalance, known as loss aversion, is what pushes a rising position toward protection and a falling one toward denial.
Losses loom larger than gains.Daniel Kahneman and Amos Tversky
A study of 10,000 US brokerage accounts, covering trades placed between 1987 and 1993, measured this directly. On the days investors sold anything, they sold about 15 percent of the gains sitting in their accounts but only about 10 percent of the losses. A position showing a gain was roughly 50 percent more likely to be sold than one showing a loss. Researchers call this pattern the disposition effect.
The winners sold went on to outperform the losers kept by 3.4 percentage points over the following year. That gap comes from one study, one market, and one stretch of years. Prices that have risen do not always keep rising, and a losing position is not owed a recovery. The study does show a large group of real investors sorting decisions by the peg, not by the case for owning what they held.
Palantir, a US data and AI software company, shows how hard this is to resist. Anyone who bought its shares during the 2021 boom watched them fall by more than 80%, then waited over three years just to get their money back.
After a wait like that, getting back to even feels like relief, and every gain after it feels like something to bank before it disappears. Plenty of investors sold for a 20% to 50% profit. Those who held until their money had doubled faced the same temptation with an even better excuse: a 100% return, after years of pain.
Anyone who sold at that point missed a further rise of around 189% to the shares' peak. Even after the price roughly halved from that peak and then recovered, it still sits around 165% above that exit point at the time of writing.
On paper, selling looked sensible. All the way up, Palantir's valuation was extremely stretched by traditional measures, and plenty of professional analysts called it a clear sell. Holding on took more than patience. It took conviction in the business itself, built on understanding it in a way the price alone could not show. That is where an investor's edge comes from, and it cuts both ways: had the business disappointed, the same conviction would have been costly.
The strongest test of that conviction came at the bottom. Investors with real conviction in the business, who added to their holding during the fall, were buying at around $6 a share. By the peak, those shares were worth roughly 30 times what they paid. At the time, it looked reckless to most people. It only looks obvious now.
That only worked because the business delivered. Plenty of fast-growing companies fell just as far in 2021 and 2022 and never came back, and buying more of those turned a bad loss into a worse one. A lower price on its own is never a reason to add. Adding into a fall only makes sense when the reasons for owning the business still hold.
A purchase price becomes a fixed peg, and later prices get compared to that peg instead of to what has actually changed.
This is anchoring: leaning on the first number available rather than on the current case for owning something.
A position moving from a paper loss into a paper gain shifts from an opportunity into something to protect.
That shift pushes toward selling before the original reasons for holding it have changed.
A position moving from a paper gain into a paper loss shifts from an opportunity into a mistake waiting to be confirmed.
That shift pushes toward holding well past the point new information would justify it.
Holding a winner takes conviction in the business, not comfort with the price.
That conviction is only an edge if it rests on genuine understanding, and it can be costly when that understanding turns out to be wrong.
In short: what you paid is a fact about the past, not a signal about the future.
Once a purchase price takes hold, gains start to feel like something to protect and losses like something to avoid admitting. That pushes investors to sell winners too soon and hold losers too long, a pattern researchers call the disposition effect.
So next time you feel the urge to sell something because it is up, or keep something because it is down, you can ask: if I didn't already own this, would I buy it today at this price? That question won't tell you what happens next. It takes the purchase price out of the decision.