Skip to content

Revisit a saved prediction and score it fairly

Find an old expectation, compare it with what happened and note what information you lacked.

reflection

A hand writing in an open notebook

Saves privately on this device. No account needed.

Read the full guide

How to review a prediction without rewriting your past confidence

Resolve a saved prediction against its original wording, record what evidence settled it, and learn from calibration instead of judging yourself by one outcome.

Plan the experience

Find an old expectation, compare it with what happened and note what information you lacked.

Planning estimates, not a price quote or a deadline. Your version may vary.

Time
Allow a short evidence-gathering session and an hour for a fair review.
Cost
Use a prediction already in your records; no betting or paid forecasting service is needed.
Where
A private workspace with the original dated prediction and relevant outcome evidence.

Before you start

Choose a low-stakes prediction whose outcome is now clear. Keep its original wording visible, read the forecasting resource’s scoring explanation and avoid selecting only a memorable success or rewriting a vague claim into something precise after the event.

How you would actually start

  1. Recover the original terms

    Find what you actually expected, by when and with what confidence. If the statement lacked a clear outcome or probability, label that limitation instead of manufacturing a precise score.

  2. Check what happened

    Gather reliable evidence that matches the original question and time period. Separate the result from whether your reasoning was sensible with the information available then, noting facts you could not have known.

  3. Improve the next forecast

    Record one lesson about question wording, evidence or confidence. If you make another prediction, state its resolution conditions in advance and set a date to review it rather than relying on memory.

What counts as done

You have compared a dated expectation with its outcome without changing the original claim.

Being right for a weak reason and wrong for a reasonable reason are different lessons; preserve both distinctions.

When you finish, mark it fulfilled in My list. You can remember it in your weekly journal.

Useful sources

  • Forecasting and keeping score

    Good Judgment Open · Read the forecasting organisation’s explanation of scoring, resolution and sharing reasoning.

Guide updated 2026-11-07

If this appeals, so might these