Shut Up And Guess

A real-world demonstration that rationality occasionally pays. On an exam scoring +1 correct, -0.5 wrong, 0 for 'don't know,' the expected value of a blind guess is +0.25, so one should always guess — yet Scott was the only person in his postgrad class (mostly hard-science degree-holders) who guessed on everything, and he couldn't convince a single classmate even though it could have raised their grades a full letter (loss aversion as a 'macro of absolute denial'). The capstone is his friend 'Larry,' who admitted guessing would help but refused on moral principle (tests should measure knowledge, not luck) — then guessed freely on a later multiple-choice exam once losing points no longer needed rationalizing. Diagnoses loss aversion generating an emotional inclination that the rational mind then rationalizes.

Quality 74 · Strong. Strong. A clear, entertaining, and pointed illustration of expected-value reasoning colliding with loss aversion and motivated rationalization, with a genuinely memorable anecdote. Held at 74 as a single-example illustration (a follow-up to 'Extreme Rationality'). 74.

Claude’s paradigm shift 42 · Moderate. Moderate(-low). EV-of-guessing is elementary and loss aversion is Kahneman's; the value is the vivid demonstration and the motivated-rationalization observation. 42.

Real-world impact 2 · Minor. A clear, entertaining illustration of expected-value reasoning colliding with loss aversion and motivated rationalization (the always-guess-on-the-exam demonstration; 'Larry' refusing on moral grounds). Conceptual/illustrative influence within rationalist discourse, single-example, no material change — low RWI.