Every prediction is logged with its confidence band and graded against what actually happened at the stated horizon — including the bad ones. No headline accuracy number is published until the sample is meaningful.
60
predictions logged
0
graded so far
—
hit rate (graded)
Why nothing is graded yet
A prediction is only graded once its stated horizon has elapsed — grading early would just measure noise. So early on, nearly every call is still cooking. As horizons pass, grades appear here automatically.
60
Logged
Recorded the moment each call is made.
60
Awaiting horizon
Still inside their evaluation window.
0
Graded
Matured and scored against reality.
The next predictions become gradeable in about 363 days.
Calibration by confidence band
Calibration data is still accumulating — 0 resolved predictions on record. Per-band hit rates will appear here once each confidence band has at least 5 resolved calls, and confidence language across the platform will then be anchored to those demonstrated rates.
Recent predictions
Status shows when each call becomes gradeable — “in Nd” means that many days until its horizon elapses.