Grading My 2021 Predictions
Read the original on Astral Codex Ten →
Summary
The annual prediction-grading post: 108 predictions binned by confidence with the calibration graph (slightly underconfident this year), plus the Simon-M comparison finding Zvi beat Scott and the markets beat both. Calibration-in-action / accountability, but content-light — mostly the list + brief calibration analysis, without a big analytical payoff.
Why this score
Quality 59 · Solid. Solid-Strong edge: good calibration practice/habit, but a list-plus-calibration-graph post rather than a substantive essay.
Claude’s paradigm shift 38 · Slight. Slight: routine annual scoring; little novelty.
Real-world impact 1 · Negligible. Personal calibration exercise; negligible material reach. =1.