When to Sell a Stock: Three Signals, One Trap

2026-10-22 · By Lubin Danilo, founder of Lubin Investment

I sell a stock when the thesis that made me buy it is broken, which shows up as a lasting deterioration in the numbers. I trim when the price runs far beyond what the business justifies. I never sell just because the price drops: a drop alone says nothing about quality.

Almost everything written about the stock market is about buying. Which stock to pick, at what price, in which sector. Selling gets little attention, yet that is where a good part of individual investors' money is lost. They hold bad companies out of pride, or they sell very good ones out of fear at the worst moment.

The question "when to sell a stock" has no universal answer, but it has a method. Mine fits in one sentence: I don't sell a price, I sell (or keep) a thesis. When I buy, I can write in three lines why: the company earns money durably, it reinvests it well, and the price paid leaves a margin of safety. Selling means noticing that one of those three lines is no longer true. Nothing else.

I will show you what that looks like with three real companies, with figures from my tool as of October 22, 2026. One whose thesis is crumbling (Nike), one whose business is superb but whose price has run too far ahead (Shopify), and one that shows the opposite case, where you definitely do not sell (Autodesk).

First signal: the thesis is broken

This is the only signal that justifies a full sale, and the hardest to accept, because it usually hits a company you know well and like. Nike is a textbook case. For years it was the perfect machine: a global brand, high margins, generous cash. Today my ten-criteria grid gives it 4 out of 10.

Look at the trajectory, not the snapshot. According to my tool's series, Nike's free cash flow (the money that really stays in the till after paying salaries, inventory, factories and taxes) was $6.6 billion for the fiscal year ended May 2024. It fell to $3.3 billion a year later, then to $2.2 billion for the year ended May 2026. Over the same period revenue went from $51.4 billion to $46.4 billion. On $100 of sales, Nike kept nearly $13 of free cash two years ago. Today it keeps a little under $5.

The mechanism is classic. When sales stall, fixed costs (stores, marketing, logistics) do not shrink as fast: the margin gets squeezed, and the cash with it. My tool sums it up: free cash flow per share is down about 16% a year over five years, return on invested capital (what each dollar put into the business earns) has fallen to 7.4% when I look for more than 15%, and costs are growing faster than revenue.

Should you sell everything at the first bad quarter? No, and this is where nuance matters. On October 1, 2026, Nike published quarterly results with real positives: the press release for the first quarter of fiscal 2027 shows gross margin up 0.6 points to 42.8%, earnings per share of $0.48 (above expectations) and the launch of a cost program called Pace, targeting about $2.5 billion over several years. But the same release shows sales down 4%, Converse falling 28% and the direct channel down 8%.

That is the right way to treat a break signal: I don't ask "was this quarter good?", I ask "has the underlying trend changed?". A gross margin that recovers thanks to lower logistics costs is a relief, not a turnaround. As long as sales do not pick up and free cash flow does not recover durably, the company worth holding is no longer the one I bought. For a stock already in the portfolio, this is the time to exit, or at least cut sharply, without waiting for the "confirmation quarter" that never arrives on time.

A simple test I apply: I write down on a sheet why I bought, then reread it with today's numbers. If two of the three lines of my thesis are false (profitability, cash growth, debt), I sell. If only one is false, I watch, read the results, and set a date by which I decide. This discipline avoids the worst error, rewriting the thesis to justify the position ("I keep it because the brand is still strong") when the numbers are what count.

Second signal: the price has run too far ahead

This second case is subtler, because the company is doing very well. Shopify scores 10 out of 10 in my grid, the maximum. Its revenue has grown about 26% a year over five years, its free cash flow went from $905 million in 2023 to $1.6 billion in 2024, then $2.0 billion in 2025, and it has no net debt. In its second quarter 2026 release, published August 5, Shopify reports revenue of $3.58 billion (+34%) and a free cash flow margin of 18%, versus 16% a year earlier. A business in excellent health.

And yet the question of selling arises, because quality is never the price. At $174.29, Shopify is valued at about 121 times its annual free cash flow. The P/FCF is the share price divided by free cash flow per share: 121 means you pay for 121 years of today's cash. For comparison, my reasonable valuation criterion is below 25. My model, which projects cash five years out with prudent assumptions and brings it back to today, gives a reasonable buy price of $71.30: the price exceeds it by nearly 60%. Against Shopify's own history, the valuation sits at the 68th percentile, meaning it has been lower about two times out of three.

Should you sell it all? I don't think so, and this is a nuance many investors miss. When quality stays intact, a high price is not proof that the thesis is wrong: it is a warning about what the market has already priced in. In that case I think in terms of position size. If Shopify weighs too much in the portfolio, I trim, take part of the gains and keep the rest, because I have no reason to sell a business that is accelerating. If I don't own it, I wait for my price and don't chase.

The trap here would be to confuse two errors. The first is to sell everything as soon as the price passes my threshold: market history is full of exceptional companies that stayed "too expensive" for ten years. The second is to say "it's a great company, the price doesn't matter". At 121 times free cash flow, growth has to last a very long time for today's investor to be paid for the risk. The right reflex sits in between: trim when the gap becomes extreme, keep a core, and accept not having perfect timing.

Third signal: the cash has a better home elsewhere

The last reason to sell is less spectacular but rational: opportunity cost. Every euro tied up in a "decent" stock is a euro that does not fund an "excellent and cheap" one. If I find a company that ticks all my quality criteria, is valued clearly below its reasonable buy price, and my portfolio is full, I can sell the weakest of my holdings to fund the best. Not out of panic, not out of boredom: by comparison.

Take Autodesk, the opposite of Shopify. My tool also gives it 10 out of 10, and its price of $244.84 sits well below the reasonable buy price of $470.58 computed by my model, with a valuation of about 26 times free cash flow, at the 23rd percentile of its own history, so lower than it has been in more than three quarters of the time. Nothing pushes me to sell here, quite the opposite. If I owned Autodesk and it had fallen 20% without its margins or cash moving, the right reaction would be to hold, or even add.

That is the line I want to draw between signal and noise. A falling price is never, on its own, a sell signal. What matters is what happened inside the company. If the numbers are stable and the price falls, the stock becomes more attractive, not less. If the numbers deteriorate and the price falls, that is not a bargain, it is what is called a value trap, as I explain in my article on value traps. In both cases the price teaches you nothing: read the cash.

The costliest trap: selling on fear or on gains

The three signals above have one thing in common: they start from the company's numbers. The two bad reasons to sell start from you. The first is fear: the market drops 10%, red headlines pile up, and you sell an excellent business because the screen is red. The second is the urge to "lock in a gain": the stock is up 40%, you sell so as not to watch the gain vanish, although nothing in the company's quality has changed. In both cases you make a financial decision based on an emotion.

The math is simple and brutal. A stock that loses 30% must then gain about 43% to get back to its starting point. Whoever sells at the bottom locks in the loss, and then has to be right twice: on the exit and on the re-entry. Almost nobody manages it. Keeping a company whose cash has not moved means refusing to pay that toll.

To protect myself from myself, I set a protocol. Before any sale I reread the buying thesis and check three numbers: is the free cash flow margin lower than two years ago, is the share count rising, has debt swollen. If the answer is no on all three, I don't sell, whatever the price says. If it is yes on two, I sell, even if the stock rose the day before. This protocol is not perfect. It is simply stable, which is what you need against emotions that never are.

One last point on taxes and fees. Selling has a cost: tax on the gain, brokerage fees, and sometimes a gap between buy and sell prices. That should not stop you from leaving a broken thesis, but it reinforces the idea that you don't sell to "do something". The cost of doing nothing on a good company is often lower than that of a pointless round trip. To go deeper on price logic, reread my method on the margin of safety and, to understand how I assess quality, the full methodology.

This is exactly the follow-up, checking a thesis with the same criteria on the day of purchase and the day of doubt, that I wanted to be able to do for any stock. You can test it on the Nike page or the Shopify one: you will see whether their thesis still holds, criterion by criterion. To build your own decision rule, the valuation work of Aswath Damodaran is an excellent reading base.

FAQ

When should you sell a stock?

When the buying thesis is broken for the long term: margins and free cash flow falling over several years, debt swelling, share count rising. A price that is too high is a reason to trim rather than sell everything. A simple price drop is not a reason to sell.

Should you sell a stock that has risen a lot?

Not automatically. If quality and cash growth stay intact, a rise does not invalidate the thesis. But if the price clearly exceeds the value the business justifies, it is reasonable to trim to reduce the weight of the position and keep a core.

Should you sell a stock that is falling?

Only if its numbers are deteriorating too. If free cash flow, margins and debt are stable, a drop makes the stock more attractive. If the numbers worsen at the same time, it is a value trap and selling is justified.

About the author

Written by Lubin Danilo, founder of Lubin Investment. A self-taught individual investor, I find fundamental analysis fascinating, and it has delivered excellent results. For three years now, my performance has beaten the S&P 500. But analyzing every stock took too much time: sites with incomplete data, calculation methods and criteria never aligned with mine. And spotting the best stocks was just as time-consuming, even with my own well-defined checklist. So I put my software development background to work to build this software, base my investment strategy on its results, and share it with people who share the same passion as me. It judges a company's quality and its price separately, using criteria drawn from the financial literature (Warren Buffett, Michael Mauboussin, Aswath Damodaran).