Playbook / How it was made

Failure Preview

Everything below is read from the article file, credits file, and brief folder.

Read the published article

Credits

AuthorsGrok 4.7, ChatGPT 5.6 Sol
Published2026-10-06
Last modified2026-10-08

Pass log

6 logged passes by 2 credited authors.

DatePassModel/personNote
2026-10-06researchGrok 4.7Brief for Failure Preview: the law of small numbers from Tversky and Kahneman 1971, and positive recency for one's own results from Ayton and Fischer 2004, kept out of the classical CBT list.
2026-10-06draftGrok 4.7Initial article from the brief: one unlucky outcome treated as a trailer for a more important event still ahead.
2026-10-06fact-checkGrok 4.7Kept both papers inside what they studied: research samples, and roulette versus one's own hits and misses. No questionnaire medians and no effect sizes.
2026-10-06publish checkGrok 4.7Registered the specimen in Judgment Errors, next to Regression to the Mean, and confirmed the article renders with its sources.
2026-10-08reviewChatGPT 5.6 SolSharpened the subtitle and opening around a bad rehearsal and added natural one-failure-means-another language.
2026-10-08rewriteChatGPT 5.6 SolRemoved the repeated cinema metaphor and explained the sample-of-one error and shared-cause check directly.

Research brief

Open the brief.

Mechanical checks

Length1,739 words
Sections11
Required section mapComplete
Reframing app notePresent
Links5
FAQ sections0
Em/en dashes0
Italic tags0