The Atomic Portfolio
This is not a model portfolio. It represents my actual capital deployment.
The strategy is straightforward: identify a small number of exceptional businesses with structural competitive advantages, buy them at rational prices, and hold them for long enough that compounding does the work. The edge is patience, not cleverness.
I view holdings as fractional ownership in real businesses — the way a private buyer views an acquisition. The correct valuation anchor is what a rational owner of the entire business would pay, not what a momentum trader bid for the shares last Tuesday. That single reframe eliminates most of the noise that passes for investment analysis.
The Philosophy
The foundational principle comes from Charlie Munger’s correction of Graham: stop buying fair businesses at wonderful prices, and start buying wonderful businesses at fair prices. A cheap business with poor economics gives you one last puff. A franchise with a genuine structural moat compounds for decades.
The goal is 20%+ annualised returns over a ten-year horizon — not lottery-ticket outcomes, but the steady, boring compounding that comes from owning genuinely excellent businesses and having the temperament to hold them when the market disagrees.
In the words of Terry Smith:
Buy good companies. Don’t overpay. Do nothing.
Concentration Strategy
I do not believe in diworsification. Spreading capital thinly across many positions reduces risk on paper and guarantees mediocrity in practice. Munger: “There is less risk in owning three easy-to-identify wonderful businesses than in owning 50 well-known big businesses.”
My top three positions account for approximately 50% of total equity. Position sizing follows conviction, not an equal-weight formula. The highest-conviction holdings — the ones where I have studied the business most deeply and where the margin of safety is widest — carry the most weight.
The portfolio is behind the subscriber wall. Access is free with a subscription.
The Investment Framework
To enter the Atomic Portfolio, a company must pass five filters in sequence. A failure at any stage ends the analysis. I never skip to valuation because the numbers look attractive — the filters exist precisely to prevent that.
1. Circle of Competence: Can I describe, in plain language, how this business makes money, why customers stay, and what it will look like in ten years? If the answer requires more than thirty seconds to explain to an intelligent non-specialist, I pass. Simple businesses with predictable economics. Not boring — Visa is simple. Baltic Classifieds is simple. A biotech betting on a single clinical trial is not simple.
2. The Moat: What specifically prevents competitors from taking this company’s customers? I require a structural answer — regulatory approval that took 12 years to obtain, switching costs that trigger re-certification, a network that becomes more valuable as it grows. Not “customers really like us.” The moat must be tested, not merely asserted. I focus on small and mid-cap businesses below $5B where institutional neglect creates pricing inefficiency — the same structural advantage Buffett identified in Lou Simpson’s outperformance at GEICO: smaller capital means access to better businesses at better prices.
3. Management Quality: Are these people good operators and honest stewards of capital? I look for owner-operators with genuine skin in the game, a track record of reinvesting at high returns rather than empire-building, and shareholder letters that discuss failures honestly. One signal I weight heavily: a CEO who conducts a substantial buyback when the stock is ignored is demonstrating, with real capital, that they believe the market is wrong. That is a rare and meaningful signal.
4. Financial Quality: Do the financials prove the moat is real? I run 18 specific checks across the income statement, balance sheet, and cash flow statement. The North Star metric is Owner Earnings — the cash a buyer of the entire business could extract annually without impairing it. Gross margin ≥ 40% and stable over time. FCF conversion tracking reported earnings. Net cash preferred; debt only acceptable for highly predictable businesses. I target a minimum financial quality score of 14/18 before considering valuation.
5. Price and Margin of Safety: Only after the four qualitative filters pass. The valuation confirms whether the price is consistent with the thesis already formed — it does not generate the thesis. I build three scenarios, use owner earnings as the base, and require a genuine margin of safety proportionate to uncertainty. If I need the model to build conviction, the conviction was never there.
The Small Capital Advantage
I hunt where institutional capital cannot follow. Funds managing billions cannot build meaningful positions in companies with market caps below $500M — the position size would move the market and the regulatory disclosure requirements would undermine the thesis. This creates a systematic pricing inefficiency in small and mid-cap territory that disappears as capital scales up.
My portfolio size means I can take meaningful positions in sub-$500M businesses that institutions ignore entirely. The margin of safety that creates this edge is temporary — it compresses as the portfolio grows — so I am exploiting it aggressively while it lasts.
The Discipline of Inactivity
Wall Street competes on milliseconds and quarters. I compete on decades.
Most investment errors are errors of action — selling when the thesis is intact but the price has fallen, rotating into the next interesting idea before the current one has had time to work, confusing price movement with information about the business. Price action is never information about the business. The only things that are information about the business are gross margin trends, churn rates, free cash flow conversion, management changes, and competitive developments.
Munger: “We try to avoid compromise of these standards, although we find doing nothing the most difficult task of all.”
The pull toward activity is almost irresistible. Resisting it is the edge.
When I Sell
Four conditions. Only these four.
1. The Moat Has Narrowed. A specific, named mechanism has fired — not a general feeling that “the thesis has changed,” but a measurable event I identified before buying as the thing that would break the thesis permanently. Gross margin declining for three consecutive years. A credible competitor achieving meaningful share. A structural regulatory change. Named, specific, observable.
2. Management Has Demonstrated Capital Allocation Failure. A large dilutive acquisition at an inflated price. Compensation structures that extract value from shareholders rather than create it. A pattern of blaming macro for poor results.
3. The Price Has Reached 120–150% of Intrinsic Value — and a Clearly Better Alternative Exists. Price alone is not sufficient. The replacement must be meaningfully better, not marginally better, after accounting for the deferred tax cost of rotating out of an appreciated position.
4. A Post-Purchase Discovery That Would Have Caused Rejection Pre-Purchase. A material fact about the business or management that was not visible at the time of purchase and that, had it been known, would have prevented the investment. This is rare but it happens. When it does, the position is closed regardless of the current price.
I do not sell because the stock has fallen. A falling price with an intact thesis is not a reason to sell — it is frequently a reason to add.