Someone is well-calibrated if the things they predict with X% chance of happening in fact occur X% of the time.

Interval scoring uses the formula developed by Greenberg, 2018. The numbers are chosen such that your expected score per question is roughly maximized when you give your true beliefs.

You can look at our charts to check your calibration curve. If you're well calibrated, the error bars should be overlapping with the red line and your overconfidence/underconfidence should be close to zero.

Please leave a comment at our Airtable or email us at hello@sage-future.org.

We're currently gauging whether our calibration game is a useful addition to the forecasting ecosystem (mostly by usage). If it is, we'll commit to spending more time improving the website and adding more features.

You can email us at hello@sage-future.org.