Track Record & Calibration Analytics

Real consensus performance metrics, Brier calibration, and category distribution across executed queries.

Forecasts Logged0Immutable payload hashes
Resolved Outcomes0Scored against venue result
Pending Resolution0Not scored prematurely
Average Brierโ€”Lower is better ยท resolved only

Agent Accuracy Ranking

Ranked by average Brier Score from resolved forecasts. Lower scores indicate better calibration.

Agent rankings will appear after the first markets resolve.

Consensus-vs-Market TrajectoryLive Log

Probability progression across executed analyses.

Brier Score: awaiting resolved outcomes
Run #1Run #1Run #0

Category Split

Execute queries to generate category calibration statistics.