A useful input survives a failed prediction rule
Present a plausible explanation or expectation, show the observation or evidence gap it cannot account for, and return to the same question with a revised account that changes the later interpretation or application.
How I Made $2,000,000 in the Stock Market
Nicolas Darvas · 27 weeksA useful input survives a failed prediction rule
SHORT SOURCE EXCERPT“Soon after, I found out that this was not true.”
Darvas adds the Dow-Jones average to his cables after noticing movements his individual quotations do not explain. He first expects his stocks to follow the average mechanically, reports repeated errors, and keeps the added input for the narrower purpose of assessing market conditions.
The sequence separates acquiring useful information from making an overconfident inference from it. The revision has an observable consequence: the same cable field remains, but the narrator changes the question he asks of it.
The source in context.
A dancing tour requires a remote workflow of weekly reading and coded cables. The narrator adjusts what the cables contain, narrows an inference about the market average and records his own error diagnoses; the closing account still reports a small net loss.
Open this chapterWhat this example supports
The sequence separates acquiring useful information from making an overconfident inference from it. The revision has an observable consequence: the same cable field remains, but the narrator changes the question he asks of it.
Its limit
No counts or complete set of predictions are supplied. The account neither measures the revised rule’s accuracy nor establishes the direction of causation between an index and its component or other stocks.
Source & edition
Supplied Evergreen Bestsellers text; all ten main chapters and appendix prose, with referenced cable and chart illustrations absent.
Supplied text, lines 1541–1563. The short quotation is checked against the source.
Read the complete whole-source AI synthesis and the summaries, section maps, qualifications and gaps of both ordered source segments. Checked all main chapter and part headings against the contents and body, inspected chapter openings and endings, and directly checked every retained example with its preceding setup and subsequent qualification. Inspected the interview and appendix prose to distinguish the narrator’s verification claims from accessible evidence. This is a source-checked editorial pass, not a second complete reading or an independent-model review.
Supplied Evergreen Bestsellers text, ISBN 9789387550469; publication date unverified. The source filename names 1970, and the referenced appendix illustrations are absent.