Why score predictions at all?
A foresight archive is more useful when it preserves both the calls that aged well and the ones that did not.
The purpose of a scorecard is learning, not self-congratulation. It asks whether a dated statement was specific enough to test, what happened afterward, and which part of the reasoning proved durable.
Preserve the receipt
Every assessment begins with the original wording, publication date and source URL. The historical statement is never silently rewritten to fit later events.
Five evaluation dimensions
1. Direction
Did the underlying change move in the direction anticipated?
2. Timing
Did material evidence appear inside the stated or reasonably implied time horizon?
3. Magnitude
Was the scale of change broadly consistent with the call, or was the outcome far smaller/larger?
4. Mechanism
Did the change occur for approximately the reasons anticipated? A correct destination reached through a materially different mechanism should be noted.
5. Specificity
Was the original statement falsifiable? Broad statements such as “technology will change business” carry less evidentiary weight than a dated, concrete claim.
Verdicts
HIT — the material outcome occurred substantially as predicted, including direction and a reasonable match on timing.
PARTIAL — important elements were correct but timing, magnitude or mechanism differed materially.
EARLY — the stated horizon has arrived or is approaching; evidence supports the direction, but the structural outcome is not yet mature.
MISS — material evidence contradicts the call or the anticipated transition did not occur within a defensible horizon.
OPEN — insufficient elapsed time or evidence to judge.
Evidence standard
Assessments should prefer primary statistics, regulators, international agencies, peer-reviewed work and direct industry data. Contemporary articles may establish what was known at the time but should not substitute for outcome evidence.
No numerical accuracy theater
Futurist.info does not reduce foresight to a single percentage accuracy score unless the underlying sample and scoring rules make that number meaningful. A collection of vague predictions can produce a misleadingly high hit rate.
Instead, report the distribution of verdicts and explain the reasoning.
Minimum scorecard record
Each evaluated prediction should show:
- original statement;
- original publication date;
- target date or horizon;
- assessment date;
- verdict;
- evidence considered;
- direction assessment;
- timing assessment;
- magnitude assessment;
- mechanism assessment;
- specificity note;
- rationale;
- what the miss or hit teaches us.
The learning question
What did the original reasoning see correctly, what did it misunderstand, and how should that change the next forecast?