These stocks where my model and the market disagree

2026-08-15 · By Lubin Danilo, founder of Lubin Investment

Analyze a stock on Lubin Investment

My model projects each stock's future cash to derive a fair buy price, regardless of its quality score. Today's widest gap: Afya and Yelp, priced at a fraction of what my model expects. But a huge gap is not always reliable, as one case I deliberately exclude below shows.

Undervalued: compared to what, exactly?

I have already published two different ways to spot a cheap stock on this site. The first compares its price-to-free-cash-flow (P/FCF), the stock's price relative to the cash it actually generates once its bills are paid, to the market as a whole: a P/FCF of 10 means you are paying 10 years of that cash. The second compares that same P/FCF to the stock's own history, to say whether it is cheap or expensive for itself, regardless of the market.

There is a third method, one I had never published on its own: my model projects the company's future cash trajectory and derives a fair buy price from it, a dollar or euro amount, not a ratio. The gap between that price and the actual market price, expressed as a percentage, tells me how much safety margin is left on the table if my estimate is right. This ranking is not limited to perfect scores: it compares everything that passes at least 8 of my 10 criteria, across every sector.

How this fair buy price is calculated

The mechanics come in three steps, which I detailed in my article on discounted cash flow (DCF): I project cash per share five years out, with growth capped at 20% a year even if the company's own history has run faster (nobody sustains hyper-growth forever). I then apply to that future cash a multiple that never exceeds the stock's own highest point over the past 5 years, capped at 40 times. Finally, I discount that future amount back to today using a fixed required return of 15% a year: the further out the payoff, the harder I discount the dollar promised, because a dollar in 5 years is worth less than a dollar today.

This price looks at neither the raw P/FCF from my study on the 50 most undervalued stocks, nor the historical percentile from my ranking of cheap perfect-score stocks: it compares today's price to what the company should be worth in 5 years, brought back to today. Two stocks with the same P/FCF can therefore show a completely different gap, depending on whether their cash is growing fast or stalling, and on whether the market has historically paid up for them or not.

The flip side: when one of the three steps is itself extreme, drawn from an unusual history, the final result can be extreme too. I show a case further below where that is clearly what happened, and why I prefer to set it aside rather than present it as a hidden gem.

The 10 stocks with the biggest gap today

I checked, live, on the day of writing, the fair buy price of every stock that passes at least 8 of my 10 quality criteria and already looks cheap on its raw P/FCF (33 candidates among the 500 top-rated stocks in my screen). Here are the 10 with the biggest gap to the actual market price, plus one case I exclude separately below.

In first place, Afya, the company that trains a large share of Brazil's doctors, already covered in detail on this site: scarce university licenses, recurring cash from students enrolled over several years. What's new since that article: cash-per-share growth, historically very fast, has clearly slowed (+2.6% year over year in the first quarter of 2026, against a double-digit average in prior years). The gap stays huge because my model still projects part of that fast historical pace, but it should narrow if this slowdown holds quarter after quarter.

Second place, Yelp: the market prices it at barely more than 7 times its annual cash, a level that reflects fear that generative AI (search summaries that answer the question directly instead of sending users to a third-party site) will dry up its free traffic. But Yelp's second-quarter 2026 release shows a company building a second leg: its non-advertising revenue (Yelp Assistant, Yelp Host, data licensing through Hatch) nearly doubled year over year, while traditional advertising fell 3%. If this shift holds, today's valuation underprices a company changing its growth engine, not just a declining ad network.

GoDaddy, already covered here, owes a good part of its cash to a simple but powerful mechanism: once a domain name is registered somewhere, switching providers is a real technical risk (broken redirects, lost emails), so renewal rates stay very high year after year, almost without any sales effort. The second-quarter 2026 release shows free cash flow up 13.3% year over year, while the market stays skeptical of its Airo bet (an AI tool that builds a website for you), whose bookings run rate jumped 5-fold in a single quarter, but from a still tiny base. It is that skepticism about the growth bet, not doubt about the core business, that explains much of the gap.

RenaissanceRe, already covered here, illustrates a mechanism specific to insurance that is worth understanding before judging its price: the combined ratio. It is the share of every premium dollar collected that goes back out in claims and expenses; below 100%, the insurer keeps the rest as underwriting profit, before even counting what it earns investing premium money while waiting to pay claims. The second-quarter 2026 release shows a combined ratio of 72.8%, close to 27 cents of underwriting profit per premium dollar before any investment income, plus $350 million in share buybacks in the quarter alone. The market stays cautious because premiums written are falling (down 12.5% year over year), a sector pulling back in good years to better withstand bad ones.

Gold Fields is the most clearly cyclical case in this list. Its first-quarter 2026 regulatory filing (Form 6-K) with the SEC reports production up 15%, but also rising costs: the diesel that fuels mining equipment has risen 30% to 70% since February because of tensions around Iran. For a gold miner, the cash generated depends almost entirely on two variables outside its control, the gold price and the cost of energy: that is what makes its cash historically volatile, and therefore my fair buy price, which rests on a projection of that cash, more uncertain than for other names in this ranking.

The one case I deliberately exclude, and why

PagSeguro (the parent of PagBank, a Brazilian digital bank) shows, on paper, the most spectacular gap in my entire screen: more than 19 times the current price. I exclude it from the ranking above because one number alone gives away the problem: the free cash flow per share reported over the recent period ($17.80) exceeds the stock's own price ($8.72). In other words, my model claims this company generates, in one year, more cash per share than what the market pays today to own the entire business. That is not impossible in theory, but it is a signal to check, never one to publish as is.

The most likely explanation lies in the nature of its business: PagBank is now a bank with a real loan and deposit book. In this type of business, cash from operations can jump in a given quarter simply because customer deposits grow faster than new loans issued, a balance-sheet swing, not a recurring profit shareholders can expect to see repeat. Without fully untangling this point (beyond the scope of this article), the simple fact that cash per share exceeds the stock price is enough to disqualify this case from a ranking meant to spot real opportunities, not calculation artifacts.

What this ranking does not tell you

A large gap between the price and my fair buy estimate is not an automatic buy signal, it is a starting point for digging further. Yelp and Gold Fields are on this list for opposite, equally real reasons: one faces a genuine structural threat to its business model (AI answering instead of redirecting to a site), the other depends on a commodity price and an energy cost it does not control. My model cannot judge whether these risks are already fully priced in or not: it only calculates the gap by assuming the company keeps following its recent cash trajectory.

Always check the full profile before forming an opinion, especially the detail of the 10 criteria and the stock's own valuation history: my full methodology explains how I weight each criterion, and my analysis tool lets you run the same calculation on any stock, including ones not on today's list.

Four takeaways

• Raw P/FCF, historical percentile, and my fair buy price measure three different things: the same stock can be 'expensive' on one and 'cheap' on another.
• Today's widest gap concerns Afya (Brazilian medical education) and Yelp (local search), followed by GoDaddy, RenaissanceRe, and Gold Fields.
• A huge gap is not always reliable: I deliberately exclude PagSeguro, whose reported cash per share exceeds the stock's own price, a signal to check, not to follow.
• This ranking is a starting point, not a buy list: Yelp and Gold Fields carry real, opposite risks that my model does not decide for you.

FAQ

What is my model's fair buy price?

It is the price I would be willing to pay today for a stock, calculated by projecting its cash per share five years out (growth capped at 20% a year), applying a multiple that never exceeds its own historical high, then discounting that amount at a required return of 15% a year.

Why is the gap so different from one stock to another?

Because it depends on two things specific to each stock: how fast its cash per share has actually grown, and the highest multiple the market has historically been willing to pay for it. Two stocks with the same price-to-free-cash-flow ratio can show a very different gap depending on these two factors.

Does a big gap mean I should buy the stock?

No. A big gap signals that my model, based on recent cash trajectory, estimates the stock as undervalued. It cannot judge whether that trajectory will continue, or whether the market has good reasons to stay cautious, as the Gold Fields case shows, exposed to an energy cost it does not control.

Why exclude PagSeguro when it shows the biggest gap?

Because one number reveals an inconsistency: its reported free cash flow per share exceeds the stock's own price, which makes no sense for a stable company. That points to a calculation artifact, likely tied to balance-sheet swings in a banking business, not a real opportunity.

Related reading

• Seven high-quality Canadian stocks, most of them pricey
• Which quality sector is cheapest in 2026?
• Are the top rated stocks cheap in 2026?

Analyze a stock on Lubin Investment

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).