Sample data for demonstration — not real holdings. Add your own portfolio in the MINE tab to make every computed tab use your book.
Where the AI’s +61% comes from — and why nobody actually earned it
This book is a rehearsal run backward: Graham's rules laid onto prices the market had already revealed. It is not a live robot, and this money was never earned by anyone. What is real is the rules — and the gap between this book and the human book is the price of not following them.
units are percentage points of the whole book's cost — they sum to the +61% exactly · valued 2026-05-16
Four rules produced this number: buy when the crowd is afraid (entries at low P/E, fear index red) · size the winners · always hold insurance (gold) · avoid the drags. The panels below measure each rule.
Same money. Same window. The bot used Graham's rules. Hindsight is honest — we said upfront these prices were known when the trades were scored.
Cheap P/E entry · diversified by geo · 20% defensive · 12 themes vs 5
Held the funds his bank recommended · Thai-equity heavy · F&G was 68 at peak
Why did the bot win? Not because the algorithm is smarter. Because Graham's rules — buy cheap, diversify, defensive sleeve, no leverage — turn out to be the boring thing that compounds.
AI +61.3% · me +12.6% — and the reason is a clock, not a stock
The ledger below is the finding: I eventually bought 8 of the same funds as the AI rehearsal. 7 of those entries came later; among those late entries, the average delay was 3.7 years. Across all shared funds, my average unit cost was 16.8% higher. I did not pick different things. I picked the same things after they stopped feeling scary — which is exactly when they had stopped being cheap.
Each of those entry dates has the same story behind it: I waited until the headlines stopped being frightening. Semiconductors after the boom was proven. Crypto after the recovery. Gold after the record. The market charges an entry fee for comfort, and the ledger above is the receipt.
Three mechanical reasons, no genius involved: it bought the identical funds years earlier (the whole table above); it never paid the comfort premium because hindsight does not feel fear; and its winners were let to run instead of being trimmed on scary weeks. The full decomposition of the +61 is below — it is hindsight rehearsal, and it says so.
The one theme only I own: China, bought near the 2022 bottom, +24.7% — proof the discipline exists; the ledger says it just needs to fire earlier, more often.
BOOK SNAPSHOTS · HUMAN 30/05/2026 · AI 16/05/2026 · AI BOOK IS HINDSIGHT REHEARSAL · THE +61 DECOMPOSED →