2015 Predictions: Calibration Results

The annual calibration-grading post: Scott scores his 2015 predictions (35 items) by confidence level (50%/60%/70%/80%/90%/95%), finds good calibration except at the 50% level (an artifact of only three items there), and calls it a successful year. Includes a gentle probe of Scott Adams's claimed 9/9 prediction success. A bare calibration list with a graph.

Quality 49 · Solid. 49 — Solid floor. As an artifact it's essentially a list + calibration graph, with almost no argument; scores low like the annual-predictions post. Kept just in Solid rather than Minor because the self-grading practice is genuinely worthwhile. Firm.

Claude’s paradigm shift 32 · Slight. 32 — Slight. A predictions-scoring list; no novel idea.

Real-world impact 2 · Minor. 2 — minor. One year's calibration list; the broader annual-self-grading practice has some niche-community influence, but this artifact's reach is contained.