Skip to content
eBook Charts
US
Before You Get Rich, Stop Losing

Before You Get Rich, Stop Losing

Noah Edge

$4.99

Buy on Amazon View on Apple Books

Prefer it in print? See every format on Amazon →

Price on Apple Books. Amazon’s price may differ.

Read a lot? Try Kindle+ * · Compare subscriptions

Published
Genres
FinanceBusiness & Personal FinanceInvesting

Description

A backtest showed 191%. After one line of bookkeeping code was fixed, it showed 8%, and the news and economic data we were testing turned out to add nothing. That happened in our own research, and it sums up this book. An individual trader and an AI research assistant numbered more than 400 experiments on Binance perpetual futures in about five weeks, with every cost counted, the plan written down before the results, and the failures kept on the record. Almost everything died. Only about one search round in 170 turned up a result worth using. This book is the record of those deaths, and of where the money leaked. Some of the stories inside: • The bot that went quiet. The AI assistant wrote "Root cause measured. Confirmed." Over the next two days, that "confirmed" collected five corrections. The bot was not broken: its owner had switched it off on purpose. The real lesson was in its old receipts. Before fees, it made less per trade than even the cheapest round-trip fee. • The coin flip that passed. A test declared that a hypothesis held because a score came out at 0.5068, where 0.5 is a coin toss. A test that checks only the direction lets a coin through. • The 191% that became 8%. A way of counting returns that is handy for training models made shorts look better than any account would record them. A short on a coin that halves in a day really gains 50%. The code booked 69.3%. • The ghost candles. A dead coin's chart kept printing the same price, with no trades at all, for three and a half years. More than two out of three of the delisted futures files we downloaded ended in candles like that. • One rule, two answers. On the exchange's 50-symbol rule, the FAQ said a restriction was not automatic. The support agent said it was. We designed for the stricter answer. • The withdrawal halt. On the very day we dated the first draft, the exchange we were studying paused withdrawals after a security incident. A backtest assumes you can always take your money out. Every chapter opens with a true story from the research records, asks the question you may be asking, and ends with a five-minute check you can run on your own trades or charts. The afterword gathers those checks into a one-page checklist. The appendices cover research methods, a map of the ideas that failed, the exchange's rules as checked in October 2026, a field guide to bugs and data traps, and a plain-English glossary. This book does not tell you what to buy. There are no signals, no coin picks and no strategy to copy. A few things did survive the tests, and the book tells that story, but it does not publish their numbers, and none of them has traded real money. For anyone who trades crypto futures or is thinking about it, builds backtests, or does research with an AI assistant. For education only. Not investment, legal or tax advice. A backtest is not a live trading record. Trading leveraged derivatives can lose all the money you put in. Fees and exchange rules are as checked in October 2026 and may have changed. Not endorsed by Binance, Anthropic or OpenAI. AI-generated content: the text of this book was written by an AI research assistant, Anthropic's Claude, which also ran most of the research. Some research tasks were done with OpenAI's Codex. The author set the project's direction and made every decision involving money, accounts or live trading; the drafts were revised under the author's direction.