Twelve principles, a small book of decisions, and the books that taught us.

If a strategy doesn't survive being multiplied by itself for thirty years, we don't begin it. The number we look at is not annual return — it's annual return, repeated, after fees, after tax, after panic. Most things that look like good investments fail one of those four filters.

The corollary is that nothing we do should require us to be cleverer next year than we are this year. The plan has to work even when we are tired.

Bullion preserves purchasing power. The house preserves shelter. Both are good. Neither is productive. The mistake we made for fifteen years was confusing the absence of loss with the presence of gain.

The ledger now names them separately. The "Storage" column is allowed to be sleepy. The "Growth" column has to be doing something — paying a coupon, paying a dividend, paying a salary, or being plausibly worth more in ten years because something happened, not because something else got worse.

A small thing that pays you every month is more educational, and more durable, than a large thing that might pay you once. The first hundred pounds of dividend income teaches more than the first hundred thousand pounds of paper gain.

This is also how the kids will learn: by being given a stake in something that posts a result every month, on a date we can mark in the calendar.

Most of our wealth will come from the four or five decisions we make right per decade. Activity is the enemy. Boredom is the precondition.

The Decision Log is, in part, a brake. If we have to write the thing down, defend it on paper, and revisit it next quarter — we make fewer of them, and the ones we make are usually better.

Every position above five percent of the ledger gets a thesis, a date, and a plan for being wrong. The plan for being wrong is the most important sentence on the page. If we cannot write it, we don't take the position.

The kids should be able to reconstruct, from this site alone, why we own what we own, what we sold and why, and what we believe about the next ten years. No private spreadsheet that only I know how to read. No tribal knowledge that dies with me.

Twelve property holdings is one bet. A house, an index fund, a private credit line, and a stake in your own son's SaaS is four bets. Always count the number of different reasons the portfolio could go up — that is the real diversification number.

I am 51. The youngest beneficiary of this office is 16. The shortest reasonable horizon for the slowest pound on the ledger is therefore the thirty-or-more years between today and the day she is my age. Anything we do should be defensible at that horizon.

From sixteen onwards, an explicit slice of any new long-term position is theirs. Small. Real. Voted on. The point is not the money. The point is the vote — and the experience of watching their own piece of a thing rise or fall over a year.

The mortgages on the rentals are fine — the tenants pay them. A loan against bullion to buy something speculative is not. The asset has to service the debt. If it doesn't, the debt is on us, and we have introduced a risk we can't model.

The long arc here is to be useful enough — with capital, with operating help, with introductions — that small private firms invite us into deals we would never have seen as outsiders. Every public-market position is, in part, dry powder for that. Every private operating skill the kids build is part of the rent we pay for the seat.

If the family ledger is not honest, none of the rest of this matters. We mark to a price we could actually transact at. We do not round up. We name our mistakes by their full names, in the Decision Log, with dates. The principle is in the name of the office.