The most natural question a sceptical reader asks of any valuation publication is whether the methodology, applied honestly, would have worked. Not whether it sounds correct in the abstract. Not whether it cites the right Berkshire letter. Whether — if you had taken its recommendations seriously, over a meaningful window — you would have made money. This is, fairly, the question Will Rogers would have asked, and we wanted to answer it before subscribers started asking it of us.

So we built The Five. It is a simulated portfolio that, each year, takes the five businesses Mr. Market is offering most generously below their intrinsic value, holds them equal-weighted for twelve months, and then rebalances to the new five. We let it run for five years. We compared what it earned to what the S&P 500 earned over the same window.

The most natural question a sceptical reader asks of any valuation publication is whether the methodology, applied honestly, would have worked.

The page itself tells you the numbers — open it and look. We will not, here, claim the result is an oracle. The simulation has the limitations every such simulation has: it doesn't pay taxes, it doesn't pay commissions, it can't see businesses that have since been delisted. The universe is the thirty businesses we presently follow, not the full S&P 500. The intrinsic values are computed at point-in-time from filings available at each rebalance date, but the growth input uses trailing earnings — we don't have access to the forward analyst consensus that was current on each historical day, because no commercial data vendor exposes that on a tier we presently pay for. Will's caveats are listed under the chart in plain English.

What we will claim is this: the portfolio is the same portfolio every time, picked from the same universe, by the same protocol, with no discretion. The protocol that produced the picks is the one published openly in The House Style. The data that fed it is the same data that feeds the live page. The procedure is repeatable. If the result is good, the protocol — applied with the discipline of an automaton — has worked. If the result is bad, the protocol has not, and we will tell you which.

We are also publishing The Ten and The Twenty, with toggles on the page. The Ten holds ten names; the Twenty holds twenty. The trade-off is what you would expect: a more concentrated portfolio swings harder both ways. A more diversified one approaches the index. Which is “better” depends entirely on what you are trying to do. We do not have a strong view. We are showing the data.

A note on what The Five is not: it is not a recommendation that you buy these five names today. The protocol publishes a fresh list every weekday — that's what The Tape is. The Five is a historical exercise: a way of asking whether, if a different version of you had followed the rules for the past five years, you would have ended up ahead. It tells you nothing about the next five.

We will update The Five every January from now on, automatically, alongside the live screen. Subscribers who want quarterly rebalance simulations — and we have built the math for it; the page just doesn't expose the toggle yet — will get those when the subscriber pages are wired up. For now, the page is free. Read it carefully. Let us know if the numbers look wrong.

— The editors