Cryptolean

Were the Odds Calibrated?

Below the diagonal the market was overconfident. The top-end gap is the headline and it rests on a handful of contracts — worth publishing, not worth trusting.

As of 8 Aug 2026Metric calib-v1Points 30Market-days 4,747Settled contracts 36Mean error -4.3%Realised +72.4%Daily close · not real-time
Zoom

Scroll or drag the plot for time · drag the value axis to rescale · double-click it to reset

Chart description

calib-v1 · implied-v1 · flags-v2 · cat-v1 · A contract settling at $1 makes its price read as a probability, but it is a RISK-NEUTRAL one: it carries risk premium and hedging demand and it is not a forecast. Treating it as the market's true belief is the mistake this chart exists to make hard. Settled Kalshi contracts only. Each bucket carries two counts and the smaller is the one to read: market-days say how much quoting there was, distinct contracts say how many outcomes were actually observed. No confidence interval is drawn, because the interval correct for 36 events and the one correct for 4,747 market-days differ by more than the result does.

How to read it

Below the diagonal the market was overconfident. The top-end gap is the headline and it rests on a handful of contracts — worth publishing, not worth trusting.

Every value plotted for a past date could have been computed on that date. Normalised series expand their window rather than scaling against the full history, so nothing here is drawn using information that did not exist yet — which is also why the early years of a normalised series are dropped rather than shown against a range of one.

Metric version
calib-v1
Export format
web-v6
Computed
8 Aug 2026
Shape
timeseries
Category
Prediction markets
Data sources
Daily close. Free public endpoints, cached before processing; nothing here is real-time.