The Atomic Moat: The Titan Test

Would Peter Lynch Buy Jack In The Box ($JACK) Today?

Rob H. | Atomic Moat — Tue, 10 Feb 2026 12:19:00 GMT

There is a seductive simplicity to the restaurant industry. You see a line of cars wrapping around a drive-thru at 11:00 PM, and your intuition whispers, “This business is printing money.” You buy a burger, you enjoy the burger, and you wonder if you should buy the stock.

This is the siren song that Peter Lynch, the legendary manager of the Fidelity Magellan Fund, famously encouraged us to listen to.

But Lynch also warned that observation is only the start of the process, not the end.

If you buy a stock solely because you like the curly fries, you aren’t investing; you are gambling with a condiment preference.

Today, we apply the rigorous, common-sense lens of Peter Lynch to Jack in the Box (JACK). The brand is ubiquitous, the tacos are legendary, but does the financial engine under the hood meet the standards of history’s greatest mutual fund manager?

Let us begin.

Subscribe now

The Titan: Peter Lynch

The Strategy: “Growth at a Reasonable Price” (GARP) & “Invest in What You Know.”

Before we dissect the stock, we must understand the man. Peter Lynch didn’t rely on complex algorithms or insider whispers. Between 1977 and 1990, he steered the Fidelity Magellan Fund to an average annual return of 29.2%, crushing the S&P 500’s ~15%. He turned $18 million into $14 billion, earning him a place on the Mount Rushmore of capital allocators.

Lynch’s philosophy was bifurcated into two steps:

Observation: Use your eyes. What is popular? What is trending?
The “2-Minute Drill”: Verify the story with hard math. Check the earnings, the debt, and the category.

He famously categorized companies into six “stories”:

Slow Growers
Stalwarts
Fast Growers
Cyclicals
Turnarounds
Asset Plays

He treated each differently. A “Fast Grower” (his favorite) required explosive expansion; a “Slow Grower” required a fat dividend.

So, where does Jack in the Box fit? And would the Sage of Magellan add it to his portfolio today?

Pillar I: The Categorization

To apply the Lynch Test, we must first answer his most fundamental question: What kind of stock is this?

Lynch would look at the historical data provided in our files and pause. A “Fast Grower”—the coveted “tenbagger”—is characterized by 20–25% annual expansion. Jack in the Box is decidedly not that.

According to the financial records:

2006 Net Sales: $2.38 Billion
2015 Net Sales: $1.54 Billion
2024 Net Sales: $1.46 Billion

A cursory glance suggests a company in decline.

However, Lynch was a master of nuance. He would recognize that this revenue drop isn’t necessarily a failure of popularity, but a shift in business model—likely a move toward heavy refranchising (selling company-owned stores to franchisees). This reduces top-line revenue but often improves margins.

However, Lynch preferred “Fast Growers” where the top line was expanding alongside the bottom line. Jack in the Box has the revenue profile of a Slow Grower or perhaps a Turnaround.

Lynch’s rule for Slow Growers is strict: “If you aren’t growing, you better be paying me.” He would demand a high, safe dividend yield to justify holding a company with shrinking sales volume. Without 20% growth, the “story” relies entirely on efficiency and capital return.

The Lynch Takeaway: He would strip away the “growth” label immediately. This is a mature, potentially saturated business. He would proceed with extreme caution, looking for a fortress balance sheet to protect the downside.

Pillar II: The Balance Sheet “Minefield”

This is where the thesis faces its stiffest challenge. Peter Lynch famously said, “It’s very hard to go bankrupt if you don’t have any debt.”

Lynch loved companies with “Net Cash”—meaning their cash on hand exceeded their long-term debt. He viewed debt as a math problem that works against the shareholder, especially in cyclical industries like dining.

Let us look at the data for JACK:

Price-to-Tangible-Book Ratio (P/B-tang): The data shows consistent negatives.
- 2021: -3.25x
- 2022: -5.62x
- 2023: -4.65x
- Recent (R12): Negative values persist.

A negative Price-to-Tangible-Book ratio implies that the company’s liabilities exceed its tangible assets. In Lynch’s era, this was often a disqualifier.

While modern finance tolerates higher leverage for franchise models (since they require less capital to run), Lynch was “old school.” He wanted a margin of safety.

He would look at the Equity Ratio and the Net Debt. If he saw a company buying back stock (which boosts EPS) while the balance sheet deteriorated, he called it “diworsification” of capital strength.

While Jack in the Box has maintained profitability (Gross Income has remained relatively stable around $400M-$500M despite revenue drops), the quality of the balance sheet would likely offend Lynch’s sensibilities.

He avoided companies where the bankers owned more of the company than the shareholders did.

The Lynch Takeaway: The “Balance Sheet Check” fails. The negative tangible book value is a red flag that would likely cause Lynch to stop his “2-minute drill” right here.

Pillar III: The Valuation (The PEG Ratio)

Let us assume, for the sake of argument, that Lynch looked past the debt. He would then look at the price.

Lynch popularized the PEG Ratio (Price/Earnings to Growth).

PEG < 1.0: Fair.
PEG < 0.5: Undervalued.
PEG > 2.0: Expensive.

With reported earnings per share (EPS) recently diving into negative territory (approx. -$4.24 TTM), a standard P/E ratio is mathematically undefined or meaningless. You cannot calculate a PEG ratio on negative earnings. Lynch would stop reading right here.

However, let’s play devil’s advocate and look at the Forward P/E—what analysts expect the company to earn next year.

Stock Price: ~$22.77
Forward P/E: ~5.8x
Projected Earnings Growth: Analysts are forecasting a rebound in earnings (some estimates sit near 80-90% growth simply because they are bouncing back from near-zero or negative lows).

The “Mathematical Mirage”: On paper, a Forward P/E of 5.8x with high projected growth creates a PEG ratio well under 1.0.

But this is a classic “Value Trap” that Lynch warned about.

The low PEG isn’t driven by a booming business; it’s driven by a collapsed stock price and a statistical quirk of recovering from losses. Lynch preferred companies with consistent, durable growth (10-20% annually), not companies zigzagging between losses and profits.

Furthermore, revenue is forecast to decline by roughly 8% per year. Lynch hated shrinking sales. You can cut costs to boost earnings for a while (which lowers the P/E), but you cannot cut your way to prosperity forever.

The Lynch Takeaway: The “cheap” valuation is a mirage. A low P/E on a company with shrinking sales and negative momentum isn’t a bargain; it’s a “falling knife.” Lynch would see the negative historical earnings and the shrinking top line as immediate disqualifiers.

The Verdict: The Titan Speaks

If Peter Lynch were sitting at his desk at Fidelity today, reviewing the file on Jack in the Box, what would be the move?

He would likely appreciate the brand. He might even enjoy a Jumbo Jack. He would acknowledge that the company has successfully transitioned to a franchise model, insulating it somewhat from operational costs.

However, the Titan Test is binary.

The Verdict: Avoid

Why?

Identity Crisis: It is no longer a “Fast Grower,” yet it carries the risks of a cyclical business.
The Debt Load: Lynch despised negative equity. The negative Price-to-Tangible-Book ratio is the antithesis of the “fortress balance sheets” he sought in the 1980s.
No “Tenbagger” Potential: The math simply doesn’t support a 10x return from these levels without a fundamental change in the business economics.

Lynch often said, “There are 60,000 stocks listed on the exchanges. You only need to find a few good ones.”

For Lynch, Jack in the Box is a company to watch from the drive-thru window, not from the portfolio. It lacks the explosive growth of a start-up and the financial safety of a true Stalwart.

In the eyes of the master, there are easier games to play.

Closing Thought from me

It is often tempting to feel that research is “wasted” if it does not end with a buy order. We are wired to crave the action of the transaction. However, I want to leave you with a personal truth I have learned after years of dissecting balance sheets: The most profitable decision you make this year might be the stock you didn’t buy.

Capital preservation is the silent engine of long-term compounding, and the ability to say “no” is the investor’s most important muscle.

Think of this analysis not as a failed search for a treasure, but as a successful defense of your fortress. By applying the Lynch framework today, we stripped away the nostalgia of a familiar brand to reveal the mathematical reality beneath. Every hour spent disqualifying a weak company is an hour invested in protecting your future returns. The burger may be tasty, but the discipline to walk away when the numbers don’t add up? That is the true taste of “old money” wisdom.

Stay prudent, and happy hunting.

— Rob

I keep this newsletter free for everyone. The only thing I ask in return is that you share this post with one friend who would find it useful.

Disclaimer

Not Financial Advice: The content provided in “The Atomic Moat” and “The Titan Test” is for informational and educational purposes only. It represents the opinions of the author and should not be construed as professional financial, legal, or tax advice. We are analyzing businesses, not providing personal investment recommendations.

Do Your Own Research: Financial markets are inherently risky. The strategies and frameworks discussed (including those of Peter Lynch) may not be suitable for your specific risk tolerance or time horizon. Always conduct your own due diligence or consult with a licensed financial advisor before making any investment decisions.

No Guarantees: Past performance is not indicative of future results. The numbers, figures, and “falsifiers” presented are based on current data and management guidance, which are subject to change without notice.

The Titan Test: Would Dev Kantesaria Buy dLocal ($DLO) Today?

Rob H. | Atomic Moat — Tue, 27 Jan 2026 13:54:27 GMT

New to The Atomic Moat? This Titan Test of dLocal ($DLO) is a prime example of how we dissect high-quality compounders. If you want these ‘Titan Tests’ sent to your inbox, join 900+ other investors below.

Join for Free

I have long kept a collection of various Dev Kantesaria’s interviews in my notepad. His intellectual discipline is something I deeply admire; he doesn’t just beat the market, he ignores 95% of it, focusing exclusively on “inevitable” monopolies like Visa or Moody’s.

But you and I possess a distinct advantage over the giants of Valley Forge: agility.

While funds like Kantesaria’s must deploy billions, forcing them to hunt only among the largest, most efficient mega-caps, we are not so constrained. As retail investors, we can navigate the shallow waters where the whales cannot swim. We can buy the illiquid, the misunderstood, and the small.

This begs a lucrative question: Can we apply Kantesaria’s rigid “Toll Booth” framework to a small-cap stock to find a monopoly before it becomes famous?

Today, we run that experiment. We are taking the rigorous lens of Valley Forge and focusing it on a $3 billion emerging market fintech player: dLocal (DLO).

Does it pass the test? Let’s find out.

Subscribe now

Pillar 1: The “Toll Booth” Test (Market Structure)

Kantesaria’s favorite holding is often the credit rating duopoly (S&P Global / Moody’s) or the payment rails (Visa / Mastercard). He loves them because they are Natural Monopolies. If you want to issue a bond or swipe a card, you must pay the toll.

Is dLocal a toll booth?

The Bull Case: dLocal describes itself as the “One dLocal” solution. They abstract away the messy complexity of emerging markets—managing hundreds of local payment methods like Pix in Brazil or Fawry in Egypt—so that a global merchant like Google or Bolt only needs one API.

The evidence of “lock-in” is strong. Kantesaria looks for Net Revenue Retention (NRR) as a proxy for how indispensable a product is. dLocal reported a staggering 149% NRR in Q3 2025. This means that even without adding new clients, their existing customers (cohorts) are spending 49% more with them year-over-year. That is the kind of “sticky” compounding Kantesaria craves.

The Kantesaria Critique: However, dLocal is not a monopoly. It is an Aggregator, not a Network. In Brazil, dLocal processes Pix payments, but they don’t own Pix. In Egypt, they process cards, but they don’t own the rails. They are a layer on top.

Furthermore, the “toll” is under pressure. dLocal’s management admitted that their net take rate (the % they keep of every dollar) was down sequentially in Q3. This was driven by mix shifts and “temporary margin pressure” in Mexico.

Kantesaria defines pricing power as the ability to raise prices 3–4% annually. If dLocal’s take rate is compressing due to “aggressive discounting by competitors” (a risk explicitly flagged in their guidance ), it fails the primary Kantesaria test. A true monopoly doesn’t discount.

Pillar 2: The Capital Intensity Filter

Kantesaria avoids businesses that must burn cash to grow (like airlines or factories). He demands High Return on Invested Capital (ROIC) and 100% Free Cash Flow (FCF) conversion. He wants a business that grows purely on intellectual property, not steel and concrete.

The Financials: dLocal shines here initially. It is an asset-light software business.

Gross Profit: $103 million (up 32% YoY).
Adjusted EBITDA Margin: A robust 70% of Gross Profit.
Revenue per Employee: This metric improved sequentially, showing operating leverage.

Kantesaria loves when a company grows revenue ($282M, up 52% YoY ) faster than its headcount costs.

The Cash Flow “Leak”: However, the “100% FCF Conversion” metric hit a snag. In Q3 2025, dLocal’s Adjusted Free Cash Flow conversion to Net Income dropped to 73%.

Why? Because of the reality of operating in volatile emerging markets. The company faced a $13M impact related to “structuring used to expatriate flows from Argentina”.

Kantesaria views “expatriation costs” and “inflation adjustments” as unnecessary friction. He wants a clean machine. A company that has to construct complex financial vehicles just to get its own cash out of a country (Argentina) introduces a layer of operational risk that Mastercard (mostly) avoids.

Pillar 3: Predictability vs. The “Macro” Excuse

Perhaps the most rigid filter in the Valley Forge philosophy is Predictability. Kantesaria avoids cyclical industries because he wants to hold a stock for 20 years without sweating the headlines.

dLocal’s Q3 2025 report is filled with what Kantesaria would call “noise”:

Currency Devaluation: Revenues in Egypt were hit by a devaluation.
Political Risk: New tariffs in Mexico slowed down TPV (Total Payment Volume).
Inflation Accounting: The company has to apply “IAS 29” inflation adjustments for Argentina, which distorts the P&L.

Management reiterated guidance, expecting TPV to exceed the high end of their range. The growth is undeniable—Total Payment Volume hit a record $10.4 billion, up 59% YoY.

But Kantesaria doesn’t buy growth; he buys certainty. He would look at the “Key Risks” slide (which lists currency devaluations in Egypt and Bolivia, fiscal regime changes in Brazil, and tariff wars) and likely conclude that the external variables are too high.

He prefers businesses where the only variable is “how many people used the product,” not “what is the inflation rate in Buenos Aires today?”

The Verdict

Applying the strict, “old money” discipline of Dev Kantesaria to dLocal (DLO):

Status: Pass (Keep Watching)

The Titan’s Reasoning: dLocal is an exceptional Growth company, but it is not yet a Valley Forge company.

Quality: The NRR of 149% is world-class. It proves the service is essential. Kantesaria would respect the “stickiness.”
Valuation/Capital: It is capital-light and profitable, fitting the financial model.
The Dealbreaker: It lacks Pricing Power and Predictability. Kantesaria invests in “monopolies in plain sight” that act as toll booths on the global economy. dLocal is currently fighting a war on multiple fronts—battling competitors on price (take rate compression) and battling governments on macro-economics (Argentina/Egypt).

Kantesaria is willing to pay a premium for a company that controls its own destiny (like FICO). dLocal, despite its brilliance, is still at the mercy of Emerging Market central banks and tariffs.

Final Thought for the Long-Term Investor: If you have a higher risk tolerance than Kantesaria, dLocal’s 59% volume growth is intoxicating. But if you are building a “sleep well at night” portfolio, dLocal has not yet dug a wide enough moat to keep the macro-economic storms at bay.

Compounding requires a smooth runway. Currently, dLocal is taking off on a gravel road.

I keep this newsletter free for everyone. The only thing I ask in return is that you share this post with one friend who would find it useful.

Disclaimer

Do Your Own Research: Financial markets are inherently risky. The strategies and frameworks discussed (including those of Dev Kantesaria) may not be suitable for your specific risk tolerance or time horizon. Always conduct your own due diligence or consult with a licensed financial advisor before making any investment decisions.

The Titan Test: Would Peter Lynch Buy Duolingo Today?

Rob H. | Atomic Moat — Mon, 12 Jan 2026 20:02:32 GMT

New to The Atomic Moat? This Titan Test of Duolingo ($DUOL) is a prime example of how we dissect high-quality compounders. If you want these ‘Titan Tests’ sent to your inbox, join 900+ other investors below.

Join for Free

If you have ridden a subway in New York, a tube in London, or a tram in Oslo recently, you have likely heard the “ding.” It is a distinct, Pavlovian sound—a chirpy, gamified affirmation that someone, somewhere, just correctly translated “The cat eats the bread.”

Peter Lynch, the legendary manager of the Fidelity Magellan Fund, famously said that the best investment ideas aren’t found on Bloomberg terminals. They are found at the mall, in the grocery store, or in the hands of your commute neighbor.

Lynch called this “Investing in What You Know.”

But Lynch was also a ruthless mathematician. He didn’t buy Dunkin’ Donuts just because he liked the coffee; he bought it because the store economics were elite and the balance sheet was pristine. He combined the observational skills of a sociologist with the skepticism of an accountant.

Today, we apply the “Lynch Test” to Duolingo ($DUOL). We are looking at a company that has permeated global culture, but does it possess the financial “story” to back up the ubiquity?

Let’s walk through the Magellan lens.

Disclaimer: I own shares in Duolingo at the time of writing this. The Titan Test” is for informational and educational purposes only. It represents my opinions and should not be construed as professional financial, legal, or tax advice. I am analyzing businesses, not providing personal investment recommendations.

The Titan: Peter Lynch

The Strategy: Growth at a Reasonable Price (GARP)

From 1977 to 1990, Peter Lynch achieved a 29.2% average annual return, crushing the S&P 500.His philosophy was deceptively simple: categorize the stock, check the “story” (the qualitative moat), and then run the math to ensure you aren’t overpaying for growth.

Lynch avoided complex tech he didn’t understand. He preferred businesses with:

A Niche: A product people have to keep buying (or using).
Growth: Specifically, earnings growth.
A Clean Balance Sheet: “Net Cash” was his safety net.
A Low PEG Ratio: He wanted growth to be higher than the P/E ratio.

He divided stocks into categories like “Slow Growers,” “Stalwarts,” and “Fast Growers.” Duolingo, growing revenue at 41% (Q3 2025), squarely fits Lynch’s favorite category: The Fast Grower.

The question is: Is it a Fast Grower that will eventually hit a wall, or is it a “Tenbagger” in disguise?

Subscribe now

Pillar 1: The “Mall Test” and The Story

Lynch’s Question: Does the product have a recurring habit loop, and can you explain the business to a fifth grader in two minutes?

If you tried to explain a semiconductor lithography machine to a child, you would fail. Duolingo passes the Lynch simplicity test with flying colors:

“It’s an app that teaches you languages like a game. People pay to remove ads and get extra lives.”

Lynch loved companies that didn’t rely on the economy to succeed. People learn languages regardless of interest rates. But what would truly pique Lynch’s interest here is the addiction metrics.

In Q3 2025, Duolingo reported 50.5 million Daily Active Users (DAUs). More importantly, the ratio of daily users to monthly users is ~37%. This is not an app people download and forget; it is a digital habit.

The “Unhinged” Moat

Lynch often looked for a company with a “niche” that was hard to dislodge. Duolingo’s moat is partially brand-driven. The “unhinged” social media strategy, where the owl mascot threatens users to keep their streaks alive, has created a cultural phenomenon that traditional advertising cannot buy.

However, Lynch would spot a risk here. He was wary of companies that relied heavily on a single supplier or customer. Duolingo lives on “rented land.”

The Landlords: Apple and Google process ~85% of revenues combined.
The Risk: If the landlord raises the rent (app store fees), the profit margin suffers.

The Lynch Verdict on “The Story”: Strong. It is a consumer staple in the digital age with high visibility. He would check this box but immediately pivot to the balance sheet to see if they can afford their rent.

Pillar 2: The Balance Sheet (The “Net Cash” Rule)

Lynch’s Question: How much cash is hidden in the share price?

Peter Lynch hated debt. He famously said, “It’s very hard to go bankrupt if you don’t have any debt.” He preferred to look at the “Net Cash” per share to see what he was really paying for the business.

Let’s look at Duolingo’s Q3 2025 numbers through Lynch’s eyes:

Cash & Investments: $1.12 Billion.
Debt: Negligible (mostly operating leases for offices).
Shares Outstanding: ~49.9 million.

The Calculation: $1,120,000,000 Cash ÷ 49,900,000 Shares = ~$22.44 per share in Cash

Subscribe now

With the stock trading at around $173, Lynch would mentally strip out that cash. And adjusting for cash, the market is effectively valuing the operating business at ~$151 per share.

Lynch viewed a high cash position as two things:

Insurance: Protection against bad times.
A War Chest: Ability to expand without borrowing at high rates.

Furthermore, he would love the Deferred Revenue of $441.7 million. Lynch understood that in subscription businesses, cash often comes in before the earnings are recognized. Duolingo is collecting cash faster than it reports income—a dynamic that often confuses novice investors but delights seasoned ones.

Stop & Think: Duolingo has more cash on hand than it has liabilities of any kind. This is exactly the kind of “Fortress Balance Sheet” that allows a Fast Grower to survive mistakes.

Pillar 3: The PEG Ratio (Valuation)

Lynch’s Question: Is the growth free, or are we paying too much for it?

This is the most critical part of the Titan Test. Lynch popularized the PEG Ratio (Price/Earnings-to-Growth).⁴ His rule of thumb:

PEG < 1.0: The stock is cheap (Growth rate > P/E ratio).
PEG = 1.0: Fairly valued.
PEG > 2.0: Expensive.

The Earnings Trap

Lynch would be very careful with Duolingo’s Q3 2025 Net Income. The reported GAAP net income was $292.2 million, but this included a one-time tax benefit of $222.7 million. Lynch warned repeatedly about “diworsification” and one-time earnings spikes. He would strip that tax benefit out immediately to find the real earnings power.

Instead of the distorted Net Income, Lynch would likely look at the Adjusted EBITDA or Operating Cash Flow to gauge the engine’s power.

Running the Numbers:

Current EV/EBITDA (FY2025 Guide): ~25.3x.
Growth Rate: Revenue grew 41%, and Bookings grew 33%. Adjusted EBITDA is scaling rapidly.

If we use the EBITDA multiple (25x) as a proxy for P/E in a high-growth phase, and compare it to the growth rate (~30-35% conservative sustainable growth):

PEG Proxy = 25 (Multiple) ÷ 33 (Growth Rate) = 0.75

The Lynch Interpretation:

A ratio of 0.75 suggests the stock is undervalued relative to its growth. Lynch would argue that the market is pricing Duolingo as a “Stalwart” (10-12% growth), but it is still behaving like a “Fast Grower” (30%+ growth).

However, he would have one major reservation: Margins vs. The “Next Thing.”

Gross margins declined slightly to 72.5% due to AI costs. Lynch was hawkish on inventory; for a software company, “inventory” is headcount and server costs. If the cost to service each user (AI compute) rises faster than the subscription price, the growth becomes “empty.”

Subscribe now

He would closely monitor if the company is spending too much on “diworsification”—trying to teach Math and Music—before fully monetizing the Language core.

The Verdict: Would Lynch Buy?

We have applied the lens of the Magellan Fund to the Owl of Pittsburgh.

The Positives:

Category: Classic “Fast Grower” with visible consumer habit (”Invest in what you know”).
Balance Sheet: $22+ per share in net cash; zero structural debt.
Valuation: Trading at a PEG proxy of < 1.0, assuming 30% growth persists.
Cash Flow: Free Cash Flow margin is elite at 28.5%.

The Negatives:

Platform Risk: Lynch disliked relying on a single buyer/supplier. Apple and Google hold the keys to the castle.
AI Costs: Rising “inventory” costs (compute) are pressuring gross margins.

The Titan’s Decision: Lynch would probably like it at todays price - based on the framework used in this analysis.

Peter Lynch would likely view Duolingo as a classic “Fast Grower” in the second inning. The company is profitable, cash-rich, and growing faster than its earnings multiple implies.

He would, however, apply his famous dictum: “You marry the girl, you don’t marry the stock.”

I do think he would watch the Daily Active User (DAU) count religiously. For Lynch, the moment the “story” (user addiction) breaks, or the balance sheet deteriorates (spending cash on foolish acquisitions), he would be out.

‘Buying what you know’ is a cornerstone of Peter Lynch’s framework, but it’s only the first step. After finding a company you understand, you must analyze its fundamentals. Duolingo is an excellent candidate for this approach. To see the framework in action, read my analysis of Jack in the Box here.

The Atomic Takeaway

We often overcomplicate investing with complex macro-theses. Lynch reminds us to keep it simple. Duolingo is a company with a product people use daily, a balance sheet that lets them sleep at night, and a price tag that—for the first time in years—offers a more attractive entry point relative to its growth rate

Compounding is the eighth wonder of the world.But it only works if you don’t overpay for the machine.

Rob H.

I keep this newsletter free for everyone. The only thing I ask in return is that you share this post with one friend who would find it useful.

Disclaimer

Disclosure: At the time of writing, the author (Rob H.) holds a long position in Duolingo ($DUOL). We believe in eating our own cooking, but this creates an inherent bias. Please read the analysis with that context in mind.

Would Peter Lynch Buy Evolution AB Today?

Rob H. | Atomic Moat — Mon, 12 Jan 2026 06:12:37 GMT

New to The Atomic Moat? This Titan Test of Evolution ($EVO $EVVTY) is a prime example of how we dissect high-quality compounders. If you want these ‘Titan Tests’ sent to your inbox, join 900+ other investors below.

Join for Free

There is a prevalent myth in modern finance that to beat the market, you need a Bloomberg terminal, an algorithm derived from quantum physics, and a direct line to a Swiss banker.

Peter Lynch, the man who ran the Fidelity Magellan Fund from 1977 to 1990, dispelled this myth with a simple, devastating track record. He didn’t use supercomputers. He used common sense, sixth-grade math, and a spiral-bound notebook.

During his tenure, Lynch compounded capital at 29.2% annually. To put that in perspective: if you had invested $10,000 with him on day one, you would have walked away with nearly $280,000 thirteen years later. He did this by ignoring macroeconomic noise and focusing entirely on the specific “story” of the company in front of him.

Today, we are taking the “Lynch Lens”—specifically his “Growth at a Reasonable Price” (GARP) framework—and applying it to Evolution AB ($EVO).

Evolution is a company that creates the digital infrastructure for online casinos. It is profitable, it is cash-rich, and it is currently suffering from a narrative crisis regarding growth in Asia.

The question we must answer is simple: Would Peter Lynch see a “Tenbagger” in disguise, or a “Fast Grower” that has run out of steam?

Disclaimer: I own shares in Evolution at the time of writing.

Part I: The Category Check

“Know what you own, and know why you own it.”

Before looking at a single number, Lynch insisted on categorizing a stock. He posited that you cannot judge a utility company by the same standards as a biotech startup. He offered six categories. Evolution AB sits precariously between two of them.

The “Fast Grower” Argument:

For years, Evolution was the quintessential Fast Grower—Lynch’s favorite category. These are small to mid-sized aggressive enterprises growing earnings at 20-25% a year. Lynch notes that you buy these for the explosion in earnings, but you must watch them like a hawk for the moment that growth decelerates.

The “Stalwart” Argument:

Looking at the Q3 2025 data, Evolution is showing signs of maturing into a “Stalwart.”

Net Revenue: €507.1m (Down 2.4% reported; +3.9% constant currency).
Adjusted EBITDA Margin: 66.4%.

Lynch defined Stalwarts as large, solid companies with 10-12% earnings growth. They are the “Old Faithfuls.” They won’t make you rich overnight, but they offer protection during recessions.

The Lynch Analysis:

Lynch would look at the 3.9% constant currency growth and immediately strip Evolution of its “Fast Grower” badge. A company growing at barely 4% is not a growth stock; it is a mature industrial paying a dividend.

However, he would be intrigued by the margin. A 66.4% EBITDA margin is an anomaly. Lynch loved companies with a niche that was difficult to replicate. Evolution supplies the live dealers, the studios, and the tech to operators who cannot afford to build it themselves.

The Lynch Question: “Is this a Fast Grower having a bad quarter, or a Stalwart that is priced like a Fast Grower?”

If the market is still pricing $EVO as if it will grow 20% forever, Lynch would walk away. But if the market has punished the stock enough to price it as a “Slow Grower,” he would open the balance sheet.

Part II: The Balance Sheet Test

“The balance sheet is the only lie detector on Wall Street.”

Peter Lynch had a profound distaste for bank debt. He famously said, “It’s very hard to go bankrupt if you don’t have any debt.”

When analyzing a company, Lynch looked for a specific metric: Net Cash per Share. He wanted to see a pile of cash that exceeded the long-term debt, which he viewed as a “hidden rebate” on the stock price.

Let us run the “Geiger Test” on Evolution’s Q3 2025 balance sheet through Lynch’s eyes:

Cash & Equivalents: €656.4m
Bond Portfolio: €103.2m
Total Liquidity: ~€759.6m
Long-Term Debt: The analysis indicates a bond portfolio, but explicitly mentions the company is “cash-rich” with no alarming debt figures cited in the summary.

The Lynch Verdict on Solvency:

Evolution passes the solvency test with flying colors. Lynch would love the Equity/Assets ratio of 73.8%. This suggests a conservative management team that isn’t leveraging the farm to chase growth.

However, Lynch would pause at one line item: Goodwill.

Goodwill: €2,336.2m
Total Equity: €3,819.6m

Goodwill makes up over 60% of the company’s equity. Lynch was generally skeptical of serial acquirers who padded their books with “Goodwill” (the premium paid for an acquisition). He preferred organic growth. He would view this €2.3bn as “soft assets”—if the acquisitions don’t perform, that equity evaporates in a writedown.

The Capital Allocation Check:

Lynch loved companies that bought back their own shares—but only if they did it cheaply.

Buybacks (9M 2025): €406.5m
Dividends (9M 2025): €572.5m

Evolution is returning nearly €1 billion to shareholders in nine months. Lynch would classify this behavior as typical of a “Stalwart.” A Fast Grower should be reinvesting every penny into expansion. The fact that Evolution is handing cash back suggests they may not have enough high-return places to deploy capital internally. This reinforces the “maturing company” thesis.

Subscribe now

Part III: The PEG Ratio & The Valuation

“If you can’t explain it to a 10-year-old in two minutes, don’t own it.”

This is where the rubber meets the road. Lynch popularized the PEG Ratio (Price/Earnings to Growth) to find bargains.

The Numbers (as of January 2026):

Share Price: 620 SEK (Stockholm) or $67.45 (ADR).
Earnings (EPS): With Q3 EPS at €1.25, we can estimate an annualized EPS of roughly €5.00.
- Currency Check: At current exchange rates (~11.6 SEK/EUR), €5.00 is roughly 58 SEK.
The P/E Ratio: 10.7x.

The Lynch Verdict on Valuation:

A P/E of roughly 11x is undeniably cheap for a company with 66% margins. The market is pricing Evolution like a dying newspaper business.

However, the “G” (Growth) is the problem.

Current Growth: ~4% (Constant Currency).
The PEG Calculation: 2.6.

The Logic:

Even at a rock-bottom P/E of 11, the stock is still expensive if it only grows at 4%. A PEG of 2.6 is typically considered overvalued in Lynch’s book.

However, this creates the asymmetry Lynch loved. If Evolution solves the Asian cyber-issue and growth returns to just 15%:

New PEG Calculation if this happens: 0.7.

A PEG of 0.7 suggests the stock is undervalued relative to its growth. The entire investment thesis rests on this pivot.

The “Diworsification” or “Distraction” Factor:

Lynch often warned about “The Whisper Stock” or distractions. For Evolution, the distraction is Asia.

The Fact: Asia revenue is down (or flat at €189.1m) due to “cybercriminality.”
Lynch’s Take: He would be highly skeptical of an external excuse like “cybercrime” lasting more than a quarter. He preferred simple businesses. Selling casino games is simple. Fighting a cyber-war in unregulated Asian markets is complex.

Subscribe now

The Verdict:

So, would the sage of the Magellan Fund buy Evolution AB today at 620 SEK?

We must separate the company from the stock.

As a company, Evolution is a fortress. It has Lynch’s favorite traits: a high profit margin (66%), a strong net cash position, and a product that is habit-forming for its customers (casinos).

However, as a stock investment under the GARP framework, it sits in “Purgatory.”

The Decision: WAIT (Watchlist)

Lynch would likely not buy Evolution AB at 620 SEK today for the “Fast Grower” bucket, because the growth has temporarily evaporated. He would not buy it for the “Asset Play” bucket, because too much asset value is tied up in Goodwill.

He might, however, keep it at the very top of his watchlist for one specific trigger: The return of earnings growth to the 15%+ range.

Lynch was never afraid to pay up for quality, but he demanded the “G” in GARP. Until Evolution proves that the Asian “cyber drag” is solved and growth re-accelerates, the math of the PEG ratio simply doesn’t work—even at 11x earnings.

Lynch famously said, “People lose more money waiting for corrections than in the corrections themselves.” Conversely, in this case, waiting for clarity on the Asian market isn’t market timing—it’s verifying the business model.

My personal assessment aligns with the Lynch framework: Evolution is a Stalwart currently being audited by the market. If the growth engine reignites, it becomes a “Tenbagger” candidate again. Until then, enjoy the dividend, but do not expect a rocket ship.

The Editor’s Note: Why I Am Buying When Lynch Says “Wait”

You might notice a contradiction here. The Lynch Test concludes with a “Wait” rating because the growth rate has temporarily stalled, yet I disclosed that I own shares.

Why am I deviating from the master?

Because Lynch’s “PEG Ratio” is a backward-looking metric, but investing is a forward-looking art. The math says Evolution is expensive relative to its current 4% growth. But my thesis is that the “Asian cyber drag” is a solvable operational headache, not a terminal cancer.

I am betting that the 4% growth is the floor, not the ceiling. You can read more about my take on Evolution here.

When you can buy a monopoly-like business with 66% margins at a P/E of 10x, you don’t need “Tenbagger” growth to make money—you just need the business not to implode. I view the current price as a “mispriced option.” If growth stays low, I collect a healthy yield from a cash machine. If growth returns to 15%? I am holding a quality company bought at a garage sale price.

— Rob H.

I keep this newsletter free for everyone. The only thing I ask in return is that you share this post with one friend who would find it useful.

Disclaimer & Disclosure

The “Atomic Moat” Standard The content provided in “The Titan Test” is designed to be a bolt-on intelligence layer to your own due diligence process, not a replacement for it. This analysis uses historical frameworks (like Peter Lynch’s) to stress-test modern companies; it is not a crystal ball.

Positions At the time of writing, the author of this report holds a long position in Evolution AB ($EVO). This creates an inherent bias in favor of the company’s success, which readers should weigh against the arguments presented.

No Advice I am not a financial advisor, and this is not financial advice. All investment strategies and investments involve risk of loss. Nothing contained in this newsletter should be construed as investment advice. Any reference to an investment’s past or potential performance is not, and should not be construed as, a recommendation or as a guarantee of any specific outcome or profit.

Independence The Atomic Moat does not accept compensation from the companies we analyze. We answer only to the truth of the balance sheet.

The Nick Sleep Test: Would He Buy Coupang Today?

Rob H. | Atomic Moat — Sat, 10 Jan 2026 09:16:40 GMT

Join for Free

There is a quiet tragedy in modern investing: the obsession with “beating the quarter” often destroys the ability to “win the decade.”

I was recently revisiting the letters of the Nomad Partnership, written by Nick Sleep and Qais Zakaria. Between 2001 and 2014, they returned over 900% not by trading rapidly, but by doing almost nothing. They identified a specific business architecture, bought it, and went to sleep.

Their philosophy was simple yet radical. They believed most companies pursue Scale Economies Kept—they get big, achieve efficiency, and keep the extra profit for shareholders. This invites competition.

Sleep preferred Scale Economies Shared. These companies (Costco, Amazon) get big, achieve efficiency, and give the savings back to the customer. This builds a moat of loyalty so deep that competition becomes mathematically impossible.

Today, we apply this lens to Coupang (CPNG).

Often lazy analysts call Coupang “The Amazon of South Korea.” But titles are cheap. The real question is: Is CEO Bom Kim actually building a shared-scale engine, or is he simply running a cash-burning logistics firm with a growth multiple?

Let us walk through the fire.

The Titan: Nick Sleep

The Strategy: Scale Economies Shared.

In the office we have a whiteboard on which we have listed the very few investment models that work... The most important model is Scale Economies Shared. This explains the success of Costco, Amazon, and Geico.
— Nomad Investment Letters

Sleep looks for a “positive feedback loop.” The company lowers costs → lowers prices → drives volume → lowers costs further. It is a perpetual motion machine of value. To pass the Sleep Test, a company must demonstrate a willingness to look “optically expensive” today to secure dominance tomorrow.

Pillar I: The Core Engine (Is the Moat Widening?)

When Nick Sleep analyzed Amazon in the early 2000s, Wall Street was screaming about net income. Sleep ignored them. He looked at Revenue Growth and Cash Flow. He wanted to see if the “customer value proposition” was increasing.

Looking at Coupang’s Q3 2025 data, we see a machine that is undeniably working.

The Hard Numbers:

Total Net Revenues: $9.27 billion (+18% reported).
Product Commerce Gross Margin: 32.1%.
Operating Income: $162 million.

The Sleep Interpretation:

A traditional analyst cheers the $162 million in operating income. It proves Coupang is “grown up.”

Nick Sleep would view this differently. He would look at the 32.1% Gross Margin in the Product Commerce segment and ask a provocative question: Is this margin too high?

Costco operates on razor-thin margins (often 11-13%) intentionally. Amazon Retail ran at breakeven for decades. If Coupang is taking a 32% cut at the gross level, are they truly “sharing” the economies of scale? Or are they beginning to “keep” them?

However, the revenue growth (20% constant currency) suggests the value proposition is still resonating. The moat is the logistics network—the “Rocket Delivery” infrastructure that no competitor in Korea can replicate. The fact that they generated $1.27 billion in Trailing Free Cash Flow (FCF) while investing heavily ($1.1 billion in PPE purchases) suggests the core business is self-funding.

The Verdict on Pillar I:

The engine is robust. The cash flow generation ($7.2 billion cash on hand) is the “oxygen” that allows for long-term thinking. Sleep would approve of the logistics dominance, but he would watch that gross margin closely. If it creeps higher without a corresponding increase in customer utility, the “Shared” part of the thesis breaks.

Pillar II: The Reinvestment (The “Taiwan” Bet)

Sleep famously held Amazon while they burned cash building AWS and Prime. He understood that Capex today is the moat of tomorrow, provided the capital is allocated intelligently.

Coupang is currently running a “two-speed” economy:

Product Commerce (Korea): The Cash Cow.
Developing Offerings (Taiwan/Eats/Play): The Cash Bonfire.

The Hard Numbers:

Developing Offerings Revenue: $1.29 billion.
Segment Adjusted EBITDA: A loss of $(292) million.
Guidance: Full-year losses of $900M–$950M for this segment.

The Sleep Interpretation:

Wall Street hates this. They see a nearly $1 billion drag on profits.

Sleep would likely love it—if the unit economics in Taiwan mirror Korea. This is “deferred gratification.” By suffering a $(292) million loss this quarter to build infrastructure in Taiwan, Coupang is depressing current earnings to build a larger future addressable market.

The “Atomic Moat” view is that this is the correct use of the $7.2 billion cash pile. A company that hoards cash is stagnant; a company that deploys cash into high-probability adjacent markets is compounding.

However, Sleep requires intellectual honesty. The “Developing Offerings” segment includes fringe bets. If Taiwan is the next Korea, it is a masterstroke. If it is a vanity project, it is value destruction. The “Scale Economies Shared” model only works if the new volume actually lowers costs for the entire system. Does shipping boxes in Taipei make shipping boxes in Seoul cheaper? That link is tenuous compared to Amazon’s global model.

Pillar III: The Governance & Trust (The Destination Analysis)

This is where the thesis faces its stiffest test.

Nick Sleep and Qais Zakaria placed immense weight on “Destination Analysis”—where is this company going, and are the people running it admirable?¹ They invested in Jim Sinegal (Costco) and Jeff Bezos because they believed these founders were fanatically obsessed with customer trust.

The “Geiger Counter” Reading:

The Data Breach: Reuters reports a massive breach fallout, including a 1.69 trillion KRW (~$1.18 billion) customer compensation plan.
The Politics: CEO Bom Kim skipped a parliamentary hearing, leading to political escalation.
The Regulatory Risk: Potential fines rising to 3% (or even 10%) of revenue.

The Sleep Interpretation:

This is a siren-level red flag.

The “Scale Economies Shared” model is built entirely on reciprocity. The customer gives the company volume/loyalty, and the company gives the customer lower prices and protection.

A $1.18 billion compensation plan is not just a financial hit (though it is roughly equal to their entire trailing Free Cash Flow); it is a breach of the unspoken contract. If customers feel Coupang is careless with their data, the “flywheel” friction increases.

Furthermore, Sleep preferred “Nomad” founders—those who quietly built value. A CEO skipping hearings and inviting regulatory wrath suggests a hubris that is antithetical to the “humble operator” archetype of a Costco.

The financial strength is there:

“Cash & equivalents of $7,229M is oxygen; it buys patience for investment mistakes.”

But does it buy forgiveness for a breach of trust?

The Verdict

If Nick Sleep were looking at Coupang today, he would see a business with the architectural advantage of Amazon (logistics dominance) but the governance risks of a frontier market operator.

He would admire the discipline to invest $1 billion into the future (Taiwan) while the market clamors for dividends. He would respect the $1.2 billion in Free Cash Flow.

But the “Scale Economies Shared” model is fragile. It requires the customer to believe the company is on their side. The massive data breach and the subsequent political fallout create a crack in the moat.

The Decision:

Based on the Nomad Partnership criteria, Nick Sleep would likely WAIT.

I think he would probably wait to see if the management team treats the data breach as a “checklist item” to be paid off, or a cultural failure to be fixed.

Atomic Moat Rating: STABLE ISOTOPE (3/5)

Great business, clouding governance.

The most dangerous competitor to Coupang is not another retailer. It is their own arrogance. If they treat the customer as a data point rather than a partner, the scale economies will stop sharing, and start eroding.

Disclaimer: This article is a theoretical application of Nick Sleep’s “Scale Economies Shared” framework to Coupang ($CPNG). This analysis was conducted by The Atomic Moat and does not represent an actual investment decision, endorsement, or comment by Nick Sleep or the Nomad Partnership. Investing is risky; please do your own due diligence.

Would Shelby Davis Buy Kinsale Today?

Rob H. | Atomic Moat — Wed, 07 Jan 2026 23:05:52 GMT

New to The Atomic Moat? This Titan Test of Kinsale ($KNSL) is a prime example of how we dissect high-quality compounders. If you want these ‘Titan Tests’ sent to your inbox, join 900+ other investors below.

Join for Free

I’ve always had a fascination with insurance. As far as I know, it is the only business model on earth where the customer hands you the cash before you even manufacture the product.

Everyone knows the Warren Buffett story: the textile mill that morphed into a float-eating dragon. But the real purists worship at the altar of Shelby Davis.

Davis wasn’t a conglomerate builder; he was a sniper. He took $50,000 in 1947 and turned it into $900 million by the time he died in 1994. His strategy wasn’t complex: buy boring, invisible insurers with disciplined underwriting, then let the float do the heavy lifting for 47 years.

Subscribe now

(If you want to learn more about Shelby Cullom Davis, read this in-depth article about him, where I go through his strategy, bio and philosophy)

He didn’t get rich chasing tech or timing oil. He got rich because he understood that a well-run insurer isn’t a casino: it’s a toll booth.

My question today is: Is Kinsale just another stock, or is it the kind of toll booth Shelby Davis would have mortgaged his house to buy?

Well, let’s check!

First, a little bit about the company:

Kinsale plays in the Excess & Surplus (E&S) sandbox. In plain English: they insure the risks the standard carriers are too scared to touch. The axe-throwing bars, the cannabis dispensaries, the small construction outfits with spotty records.

Most insurers treat these risks like toxic waste.

Kinsale are the only pure-play carrier that built its own tech stack to ruthlessly automate the “boring” back-office work. While the industry giants are busy feeding legacy mainframes and paying brokers to play golf, Kinsale uses its tech to strangle the expense ratio down to the low 20s (the industry average is often bloated in the 30s).

They focus on small accounts and they retain absolute control over the data. This isn’t just an insurer; it is a margin extraction machine operating in a muddy field.

The Setup

So, Kinsale deals in “weird risk.” They price the unlovable, take the premium, invest it, and wait for the claims. Sounds dull.

Until you check the mechanics.

A combined ratio under 100% means the engine generates profit before they even touch the investment portfolio.

This quarter wasn’t divine intervention. It was… routine. But there is a crack in the windshield: growth looks fine on the surface, but the heavy lifter, Commercial Propert, is fighting a much tougher tape.

[KNSL][REPORTED (GAAP)][Q3 2025] KEY STATS

GROSS WRITTEN PREMIUMS (USD)...........486,251,000
NET EARNED PREMIUMS (USD)..............410,957,000
UNDERWRITING INCOME (USD)..............105,672,000 (NON-GAAP)
NET INVESTMENT INCOME (USD)............49,604,000
NET INCOME (USD).......................141,646,000

Personal Stake: No position. Watching from the sidelines.

What the market thinks right now (Bulls vs Bears)

Market: “Great underwriter. Now prove you don’t melt when the property backdrop gets soft.”
Bull: “Pure compounding. The product isn’t insurance; it’s discipline.”
Bear: “You mistook luck for skill. Peak conditions plus a reinsurance tailwind.”

We are here to see if the underwriting gears grind or glide as commercial property cools off and retention rises.

Atomic Take: Kinsale prints underwriting profit. But the growth mix is shifting. The “easy” money is getting harder to keep.

What breaks this? The combined ratio cracks above 85% and stays there (vs 77.5% for 9M 2025, reported).

Balance Sheet: The Geiger Test

Insurers run on scary-looking liabilities. That’s the business model. Loss reserves are future claims; unearned premiums are cash they haven’t earned yet.

As of 30 Sep 2025, Kinsale sat on $2.77b of reserves for unpaid losses and LAE and $909.6m of unearned premiums.

The asset side backs those promises. Management listed cash and invested assets of $4.9b at 30 Sep 2025, anchored by $4.2b of fixed-maturity securities.

The number-free truth: This balance sheet is engineered to induce sleep.

[KNSL][REPORTED (GAAP)][AS OF 30 SEP 2025] KEY STATS

TOTAL ASSETS (USD).....................5,833,305,000
CASH & CASH EQUIVALENTS (USD)..........170,162,000
FIXED-MATURITY SECURITIES FV (USD).....4,171,238,000
EQUITY SECURITIES (USD)................571,100,000
TOTAL STOCKHOLDERS’ EQUITY (USD).......1,865,217,000

The Audit

Optical Illusion #1: Don’t let “low debt” fool you. The real leverage isn’t the $199.3m of debt (GAAP). It’s the massive promise embedded in $2.77b of loss reserves.
Optical Illusion #2: Accumulated other comprehensive loss isn’t a mysterious leak. It’s bond math. AOCI hit $(36.2m) at 30 Sep 2025 (GAAP), an improvement from year-end.

Atomic Take: The balance sheet signals an insurer expecting to keep the privilege of holding a mountain of other people’s money.

What breaks this? Reserve volatility spikes. Specifically, prior-year development turns ugly (vs $(45.9m) favorable in 9M 2025, reported).

Cash Flow: Energy Output

In this game, “cash from operations” is a mix of underwriting skill and timing—when the check clears vs. when the claim hits.

Kinsale reported $802.3m of net cash provided by operating activities for 9M ended 30 Sep 2025.

Ignore standard “investing cash flow” logic. Here, that’s just the portfolio breathing (securities trading). The real Capex is just property and equipment.

The number-free truth: Cash timing is a feature. It is also a trap.

[KNSL][REPORTED (GAAP)][9M 2025] KEY STATS

CASH FROM OPS (USD)....................802,344,000
PURCHASE OF PPE (USD)..................43,083,000
CASH-AFTER-INVESTMENTS (USD)...........759,261,000 (CFO - PPE)
CASH & CASH EQUIVALENTS END (USD)......170,162,000

The Audit

Management warns the plumbing is noisy: claim payments and reinsurance receipts shake the pipes from quarter to quarter.

Also: financing outflows aren’t just dividends. They include buybacks and payroll taxes on equity awards. “Cash returned” is messy.

Atomic Take: The engine hums, but the odometer twitches. Claims don’t respect office hours.

What breaks this?

Operating cash flow starts dragging behind underwriting profit for multiple periods.

Share Capital & Insiders: Nucleus Check

Kinsale returns capital while growing equity. That is the cleanest signal of confidence in the insurance game.

During 9M 2025, they shipped $0.51 per share in dividends and bought back $40.0m of stock.

Shares outstanding sat at 23.26m at quarter-end.

The number-free truth: Buybacks are easy when the core engine leaks cash in the right direction.

[KNSL][REPORTED (GAAP)][AS OF 30 SEP 2025] KEY STATS

SHARES OUTSTANDING (SHARES)............23,261,360
TREASURY STOCK COST (USD)..............49,999,000
DIVIDENDS DECLARED (USD/SHARE).........0.17 (Q3 2025)
DIVIDENDS PAID (USD/SHARE).............0.51 (9M 2025)
SHARE REPURCHASES (USD)................39,999,000 (9M 2025)

Atomic Take: Shareholder returns look calculated, not desperate. Small enough to remain optional.

What breaks this?

Payouts rise while the debt load climbs (baseline: total debt $199.3m as of 30 Sep 2025).

Subscribe now

Income Statement: Reaction Chamber

For 9M ended 30 Sep 2025, gross written premiums hit $1.53b. Net earned premiums landed at $1.16b.

The underwriting room stayed profitable. Combined ratio was 77.5%, driving underwriting income of $268.6m.

The growth plot has a twist. Commercial Property (the biggest division) saw gross written premiums shrink 15.2% in 9M 2025. The rest of the business grew 14.4%.

The number-free truth: Pricing power is the only scene that matters.

[KNSL][REPORTED (GAAP)][9M 2025] KEY STATS

GROSS WRITTEN PREMIUMS (USD)...........1,526,048,000
NET EARNED PREMIUMS (USD)..............1,160,360,000
UNDERWRITING INCOME (USD)..............268,586,000 (NON-GAAP)
COMBINED RATIO (PCT)...................77.5
NET INVESTMENT INCOME (USD)............139,896,000

Quick Hits

Prior-year reserve development was $(45.9m) favorable in 9M 2025.
Catastrophe losses hit $27.5m in 9M 2025.
Net retention ratio climbed to 81.6%.
Expense ratio clocked in at 20.6%.
Investment portfolio gross investment return ran at 4.3% annualized for 9M 2025.
Holding more risk (retention) boosts future profit but spikes the expense ratio as ceding commissions vanish. Management blames lower ceding commissions for the expense pressure.
Investment income is morphing into a tailwind: net investment income hit $139.9m for 9M 2025 (GAAP), up from the prior year.

The Scorecard

This is the filter. We separate the noise from the signal, then name the metric that proves the thesis right or wrong.

Signal 1: Commercial property competitiveness

🔒 FACT: Commercial Property gross written premiums dropped 15.2% in 9M 2025.
🧠 HYPOTHESIS: The broader book outruns the property drag without cutting corners.
📈 MONITOR: Gross written premiums excluding Commercial Property stays above 14.4% growth.

Signal 2: Reserve “good news” durability

🔒 FACT: Prior accident years developed favorably by $45.9m in 9M 2025
🧠 HYPOTHESIS: This is a tailwind, not a lottery ticket.
📈 MONITOR: Prior-year development doesn’t turn ugly relative to the $45.9m favorable baseline (9M 2025).

Signal 3: Retention and reinsurance economics

🔒 FACT: Net retention ratio jumped to 81.6% in 9M 2025.
🧠 HYPOTHESIS: Profit rises faster than volatility.
📈 MONITOR: Combined ratio holds the 77.5% line while retention stays high (9M 2025 baseline).

Atomic Take: The underwriting is elite. But the mix is rotating toward lines that demand you prove it every quarter.

What breaks this?

Reserve development flips and cat losses shove the combined ratio meaningfully above the 77.5% baseline.

Subscribe now

The Atomic Verdict

Quality: Underwriting profitability is exceptional at 77.5% combined ratio for 9M 2025.
Health: Balance sheet holds $1.87b stockholders’ equity as of 30 Sep 2025.
Overhang: Pricing pressure is visible. Commercial Property GWP fell 15.2% in 9M 2025.
What must happen next: Non-property lines keep scaling without losing the reserve discipline shown by $45.9m favorable development (9M 2025).

Status: STABLE ISOTOPE (Watchlist)

Stars: ★★★★☆ — underwriting machine stays cold; growth mix is getting harder.

Upgrade Triggers

Commercial Property returns to non-negative gross written premium growth (baseline: -15.2% in 9M 2025).
Combined ratio holds at or below the 77.5% 9M 2025 baseline while retention remains elevated.
Operating cash flow stays heavy versus the $802.3m 9M 2025 baseline.

Downgrade Triggers

Prior-year development turns adverse against the $45.9m favorable 9M 2025 baseline.
Catastrophe losses breach the $27.5m 9M 2025 level.
Credit Facility borrowing becomes a habit, not a bridge (baseline: $26.0m outstanding at 30 Sep 2025.

…So, what about Shelby?

Shelby Davis didn’t build a dynasty by betting on miracles. He bet on math. He bet on managers who refused to write a bad policy just to hit a quarterly target.

In my opinion, Kinsale is currently passing the Davis test, even though it does not currently fulfill the Davis Double Play. That 77.5% combined ratio is the kind of boring excellence he would have framed on his wall. The “alchemy” is working: the cash is coming in before the product goes out, and the float is building.

But even the best toll booth suffers when the traffic pattern changes. The machine is engineered correctly, but the road beneath it (commercial property) is getting rougher.

The question isn’t whether the engine works; we know it does. The question is whether you have the stomach to hold the “lab coat” while it gets splattered by a softening cycle.

Davis held for 47 years. If you have that kind of timeline, the current property drag is just noise. If you don’t, keep your eyes on the fenders.

What do you think? Is Kinsale a company Davis would have liked? or even more important, do YOU like it?

Comment below!

Disclaimer

Atomic Moat Research is a “bolt-on” intelligence layer. Use it to check your work, not replace it. I break down financial statements and models for the rational retail operator.

Independence: I take zero compensation from the companies I review. This research exists because I hunt high-quality compounders for myself.

Skin in the Game: Unless I say otherwise, assume I hold long positions in the assets discussed. I bet my own capital on my own analysis. That creates a bias. You should know that. I strive for institutional depth using the latest filings, but these are my opinions. I am not a financial advisor. Do your own homework before you deploy capital.