The position you're holding right now, the one that's red, you're not holding it because you believe in it.

You're holding it because selling would make the loss real.

But the loss is already real. Your brokerage statement updates every second. The market priced it in the moment the stock moved. The only thing that changes when you sell is your willingness to look at what already happened.

In 1998, Terrance Odean at UC Berkeley got access to 10,000 brokerage accounts. Every transaction from 1987 to 1993. He tracked when people chose to sell and when they chose to hold.

The pattern was systematic.

Investors were 1.5 to 2 times more likely to sell a stock that was up than one that was down. Not sometimes. Across thousands of accounts. Across seven years. Regardless of stock, sector, or market regime.

They sold their winners. Kept their losers.

Odean tested every rational explanation. Tax optimization, no, the behavior was tax-suboptimal, costing them money they could have saved harvesting losses. Informed conviction, no, the stocks they sold went on to outperform the stocks they held by 3.4% over the following twelve months.

They were systematically cutting their strongest positions and preserving their weakest ones. Not because they had bad judgment about stocks. Because they had a reference point distorting every decision they made.

Their entry price.

Kahneman and Tversky mapped this in prospect theory. The brain doesn't evaluate a position based on where it is. It evaluates based on where it is relative to where you got in.

Above entry registers as gain. Below entry registers as loss. And the function is asymmetric, losses produce roughly twice the neural response of equivalent gains.

A position up 20% generates moderate satisfaction and an impulse to secure it. Lock it in. Don't let it evaporate.

A position down 20% generates acute discomfort and an impulse to wait. Hold. Don't crystallize. It'll recover.

Neither response has anything to do with where the stock is going. Both are entirely products of a number that stopped being relevant the moment after you bought.

{{first_name}} two investors can hold the same stock, same day, same price, and reach opposite conclusions about whether to sell. Not because they see different futures. Because they have different pasts. One entered at $40 and sees a gain to protect. The other entered at $80 and sees a wound to endure. Same equity. Same forward thesis. Opposite behavior governed entirely by a number the market never stored.

The cost of this isn't the bad hold. It's everything the bad hold displaces.

Every dollar anchored in a position you wouldn't buy today is a dollar absent from a position you would. Odean's 3.4% annual gap between sold winners and held losers doesn't sound catastrophic in isolation. Compounded across a decade of portfolio decisions, it's the difference between building wealth and organizing regret.

The investors in his dataset weren't picking bad stocks. They were allocating capital based on where they'd been instead of where the opportunity was now. The entry price, a number with zero predictive value, was quietly governing every sell decision, every hold decision, every rebalancing choice they made. And they didn't know it was happening.

That's the part that should bother you. This isn't a mistake you make once and learn from. Odean's data spans seven years. The bias didn't weaken with experience. Didn't improve with market knowledge. The disposition effect operated just as powerfully in year seven as year one because the mechanism lives below the level where experience accumulates.

The inversion is identical.

You're up 40% on something. The impulse to lock it in feels like discipline. Responsible risk management. Secure the gain before it disappears.

But the stock doesn't know you're up 40%. It doesn't remember your entry. It's either a good investment at today's price or it isn't. And the stocks that run 40% are often running because something structural shifted, earnings inflected, the market re-rated the sector, a growth driver materialized that didn't exist when you bought.

Selling because you've "made enough" is measuring from a number the market discarded months ago. The gain feels like yours to lose. It isn't. It's current price minus current price. There's nothing to give back that isn't already being repriced in real time.

The investors who compound across decades aren't wired differently. They just learned to ask a different question. Not "how much am I up" but "would I buy this today at this price with what I know right now."

One question measures from memory. The other measures from structure. Only one of them allocates capital forward.

The audit is simple.

Open your portfolio. Every position. For each one, if you had cash instead of that position, would you buy it today at today's price with today's thesis?

The ones you'd buy again are your real portfolio.

Everything else is emotional inventory, positions held not by conviction but by a reference point that expired the moment after execution. The thesis that justified the entry may have degraded. The structure that made it attractive may have dissolved. But the entry price persists in your memory like Kahneman's wheel, spinning a number that has nothing to do with the question in front of you and distorting your answer anyway.

Your portfolio isn't what you own. It's what you'd rebuild from cash. The distance between those two things is what the disposition effect costs you, silently, annually, and without any intention of stopping.

Talk soon,

Atif

P.S. The position you thought of while reading this, the one you've been waiting on for months, you already know the answer to the question. The only thing left is whether you measure from memory or from where the stock actually is.