Is a Stock Expensive? Compare It to Itself

2026-10-13 · By Lubin Danilo, founder of Lubin Investment

A stock is expensive or cheap first relative to itself. I look at where its current valuation sits in its own history: Adobe has been valued this low in only 3% of past readings, while Visa is richer than 78% of its past. That marker does not decide alone, but it asks the right question.

The trap of comparing with the neighbor

The first reflex when asking whether a stock is expensive is to look at its peers. The sector trades at 25 times cash, this one at 18, so it must be cheap. The logic looks serious and is almost always misleading. Two companies in the same sector do not have the same margins, the same growth or the same resilience. Putting them side by side is like comparing the price per square meter of a Paris studio and a Breton farmhouse and calling the second a bargain.

I prefer to ask a different question: how much has the market agreed to pay for THIS company over the years? To answer, I use the P/FCF (price-to-free-cash-flow). Free cash flow is the money that really stays in the till once the bills are paid and the investments made. P/FCF divides the company's stock market value by that yearly cash: a P/FCF of 12 means you pay the equivalent of twelve years of cash. The lower it is, the less the stock is valued.

Then I take every past reading of that ratio for the same company and check what share of them was lower than or equal to today's value. That is the percentile. A percentile of 3 means that in only 3% of the available readings has the stock been valued as low as it is today. A percentile of 88 means it was pricier barely 12% of the time. The scale depends on how deep each company's history is, so I do not compare it across companies with scientific precision. It is for locating, not for calculating. And I always use it after judging the quality of the business, never instead of it: a mediocre company at the bottom of its history is still mediocre.

Adobe and Copart: the same signal, two stories

Take the two clearest examples from my screener. Adobe is valued today at 11.7 times its free cash flow, which puts it at the 3rd percentile of its own history. In other words, over the measured period, the maker of Photoshop has almost never been valued this low. Copart, the online auction platform for damaged vehicles, shows 19.6 times its cash and a percentile of 2. Both sit at the floor of their history. That does not mean the same floor.

Adobe has a free cash flow margin of 32.7%: out of 100 euros of sales, nearly 33 end up as available cash. Its cash per share grows around 13% a year, helped by a share count shrinking 3.1% a year through buybacks. Above all, the latest quarter does not look like a company in trouble: the September 10 release reports record revenue of $6.76 billion, up 13%, and a raised full-year outlook. It also says its AI-first revenue grew more than 150%, without disclosing the starting base, so I draw no numerical conclusion from it. What the percentile tells me here is that the market has punished the fear of AI far more than the results. Whether that fear is justified is a real debate: a new CEO arrives on December 1 and the finance chief role is held on an interim basis. A discount can also signal a transition that is not well prepared.

Copart tells something else. Over five years its cash per share grew around 17% a year, with a free cash flow margin of 28%. But its latest fiscal year, ended July 31, stalled: according to its September 10 release, revenue grew only 0.4% and earnings per share slipped from $1.59 to $1.55. In the fourth quarter, general expenses jumped 16.6% and net income fell 17.4%. The company attributes part of the U.S. decline to the prior-year base, which included one-off revenue from Hurricanes Helene and Milton. So Copart's low percentile is a signal of a different nature: the market is not panicking over a distant fear, it is noting that growth has stopped. The question is no longer "will AI destroy it?" but "is this stall temporary or lasting?"

That is the whole value of the method. The percentile never gives the answer, it forces the right question. In both cases I then have to explain why the multiple fell, and judge whether that reason holds. Without this tool I would have seen two "cheap stocks" and confused them.

Visa and Coca-Cola: when expensive has an explanation

The reasoning works the other way too. Visa is valued at 39.5 times its cash, above 78% of its past readings. Coca-Cola, at 31.5 times, is above about 88% of its own. Taken alone these are warning signs: at those levels the market already pays a lot for the future. But they do not tell the same story, and it is still business quality that sets them apart.

Visa produces a free cash flow margin of 47.1%, one of the highest I measure, cash per share up around 13% a year and buybacks shrinking the share count by 3.7% a year. Its payment network works like a toll road: the more merchants and cardholders, the harder it is to bypass. A high multiple for a machine like this can be explained. It simply leaves little room for error: if growth disappoints, the multiple can drift back toward its average and the stock can lose value without the company doing anything wrong. Coca-Cola grows more slowly, around 5% a year in revenue, with a free cash flow margin of 24.9%. To me, a multiple that high on growth that modest is harder to justify. That is a personal reading, not a fact: the appeal of a stable dividend may also explain the price.

A high percentile is therefore not an automatic "do not buy", any more than a low percentile is an automatic "buy". PayPal shows it: 11.5 times its cash, 8th percentile, with cash per share up around 12.5% a year. On paper that is the same profile as Adobe. Yet I still have to explain why the market does not grant it more, and its free cash flow margin of 13.3% is much lower than Adobe's. A low multiple on a less profitable business is less of an anomaly.

The rule in three questions

Here is how I proceed, in order. First, I check that the company is good, without looking at the price: profitability, cash growth, buybacks, debt. Then I locate its current valuation in its own history. Finally, whatever the result, I look for the explanation of the move and ask whether it holds. A low multiple because the market fears an exaggerated risk is a possible opportunity. A low multiple because growth has vanished is a possible trap. Both produce exactly the same number.

I do not claim this marker is infallible. The history can be short, a change of business model can make the past less relevant, and a percentile says nothing about the future. But it is a better starting point than the neighbor in the sector. It is precisely what I wanted to automate in my methodology, so I could run any stock through the filter in seconds, quality on one side and price on the other. You can see the result on the pages for Adobe, Copart or Visa. For the theory, Aswath Damodaran's work on valuation multiples remains a reference.

FAQ

How can I tell whether a stock is expensive?

Do not compare it only with its competitors. Look at where its current valuation sits in its own history, then find out why it moved. A stock is only expensive or cheap relative to what its business deserves.

What does a percentile of 3 mean for a stock's valuation?

In about 3% of past readings the stock was valued as low as today. It signals that the market pays very little compared with its habit, not a guarantee that it is a bargain.

Is a low multiple always good news?

No. It can reflect an exaggerated fear, or growth that has really stopped. The same number hides two opposite situations, which is why I judge the quality of the company first.

Why compare with history rather than the sector?

Because two companies in the same sector do not have the same profitability or growth. Its own past is the only comparison with the same kind of business.

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).