Disclosures

What this is

Educational content only. I'm not a registered investment adviser. Nothing on this site recommends buying or selling any security. Everything here describes rules I follow with my own money, not advice about what you should do with yours.

Questions about what you should buy or sell get no answer from me, wherever they're asked. Answering would turn education into advice, and I stay on the education side.

What is published

Every month, I publish the portfolio as an index that started at 100 at inception. It's time-weighted, so money going in or out doesn't move it, and it sits beside the world's stock markets as represented by VT. Each month's contribution shows up as a percentage of my gross income, never in dollars. Under the rules it can't drop below 30%. If I make a withdrawal, it's published that month with its reason and its size as a share of the portfolio.

Each month gets three other records next to the index. The Breach Count is every time I chose to act, or not to act, against the written rules. It's counted from the start and never reset. The Error count tracks unintended deviations, like selling the wrong lot or missing a transfer, and it's kept apart from breaches. The temptation log lists every temptation to act outside the rules that got as far as a concrete step, and whether I gave in. Each month's entry also names the principle that was tested and whether it held, and gives the number of companies I hold as a count.

Every quarter, I publish the portfolio's factor loadings, measured against published academic data.

One month later, I publish every holding by name with its weight, the sector exposure those holdings add up to against the cap, and the screen's full ranked list of eligible companies. They go up a month late, so by the time you see them I've already traded. It's a record of what I did, and nobody should read it as an instruction.

What is never published

I never publish my account balances, my net worth, or my income in dollars. Contributions and withdrawals appear only as percentages, and the portfolio only as an index. I don't name my broker or share any account identifier, and nothing here is published in real time.

I don't run a trade-alert service or a signals group, and I don't sell picks.

Nothing is edited

Ledger entries are never edited or deleted after publication, including the months that went badly. Rules are never changed in place: a change produces a new dated version, and every earlier version stays published. Both are kept in a public repository, so the history can be checked independently of this site.

Money

I'll never accept any arrangement that pays according to trading activity. If this site ever carries sponsorship or sells anything, it will be labeled where it appears.

AI assistance

I drafted the rules with help from AI, and AI wrote much of the code that runs them: the screener that ranks the companies each month, and the tracker that plans the trades and keeps the record. The strategy is mine. I decided what it is and what every rule says, I told the AI what to build, and I made every final call, including on the changes it suggested. Every number in the rules is one I chose and can defend without help, and the reasoning is published alongside them. Routine help with drafting and research isn't separately labeled.

The programs themselves contain no AI. They're ordinary code that applies the written rules the same way every month, and I check what they produce before anything trades.

The standing commitment: no AI system makes, recommends, or influences a live position decision. The rules are written by me and carried out by me. Assistance ends where the portfolio begins.

Foreign exposure

The universe is US-listed, not US-economy. Companies listed through depositary receipts, such as Shell, Sanofi, and Unilever, make up a meaningful share of the eligible list. The rules cap the US at its share of world stock markets (62% for now) and any other country at 10%, so a third or more of the portfolio will usually be based abroad. That brings currency risk, and dividends from those holdings are subject to foreign withholding tax, which is partly recoverable through the foreign tax credit. A foreign company that says it's a passive foreign investment company, which US tax law treats harshly, isn't bought.

Tilts I didn't choose

A few characteristics of the portfolio emerge from the rules rather than being chosen. They're worth being aware of.

Buybacks. Companies shrinking their share count are markedly over-represented. Nothing in the rules selects for this; the profitability and investment characteristics produce it.
Negative or zero book value. A company with zero or negative book equity can't get a value score, so it's left out completely. On the evidence so far, this isn't a hidden bias against heavy buyback companies.
Recent listings wait. A company needs three years of annual statements before it can be bought, so new listings and spin-offs stay out until they have a record.
Sector lean. The top of the ranking can bunch in one sector (in October 2026 it was technology, with 14 of the top 40), so I cap any single sector at 25% of the portfolio.
Smaller companies. Each month's money brings the companies bought toward the same position size, whatever their market value, and the eligible list reaches down to the NYSE 20th percentile of market value (about $1.2 billion in 2026). So the portfolio leans toward small and mid-size companies compared with the world market (VT), and the quarterly factor loadings show this as a positive size loading. The lean costs the Core in years when the largest companies lead: in the backtest from 2016 the Core trailed VT in three of the six years they led, one of them by a fifth of a point.
Bigger swings than the market. Small, cheap companies tend to swing harder than the market does. In the backtest from 2016 the portfolio moved about 1.27 times as much as VT, its volatility was 22% a year against VT's 14.6%, and its worst fall was -51% against VT's -34%.

The options account

An options account, selling defined-risk put spreads on the S&P 500, has been designed but hasn't started, and nothing is traded in it. If I start it, I'll publish its rules as a new version at least a month before its first trade. If it begins, it will be reported separately from the stock portfolio and never blended with it.