Comparison

Athena Index vs Metaculus

Metaculus measures forecasters who volunteered to be measured, on questions someone wrote for them. Athena measures public figures on claims they made on their own channels, in their own words.

Athena Index

Behavioral credibility of named public figures, extracted from what they actually published.

Metaculus

Community prediction. Aggregated forecasts from participants who opt in and answer structured questions.

The difference in one paragraph

The split is consent and provenance. A Metaculus forecaster chooses to participate, sees the question in advance, and states a probability in a structured field. That produces clean data and a genuinely skilled crowd.

Athena starts from the opposite end. The claims were made on YouTube to an audience, without a resolution criterion, often without a deadline, and usually by someone with no intention of being scored. Extracting a falsifiable claim from that is the hard part of the work, and it is why specificity is one of the six pillars: a vague claim scores badly because it was vague.

This also changes what the score means. A Metaculus track record tells you how good a forecaster is. An Athena score tells you how much weight to give a public figure who is already influencing decisions, including how they behave when they turn out to be wrong.

Feature comparison

CapabilityAthena IndexMetaculus
Subjects opted in to being scored
Scores named public figures
Claims extracted from public content
Produces forecasts of future events
Structured resolution criteria set in advance
Tracks deletions and walk-backs
Behavioral pillars beyond accuracy6 pillars
Source-linked evidence per claim

When Metaculus is the better choice

If you want a calibrated probability on a well-specified future event, a forecasting community will serve you far better. Athena does not produce forecasts and takes no position on what will happen next.

Common questions

Is Athena Index a prediction market or a forecasting platform?

No. Athena makes no forecasts and takes no position on future events. It documents claims other people already made and records what happened afterwards.

Why not just use a forecasting platform's track records?

Those records only cover people who chose to participate. The public figures actually moving retail decisions are largely not on those platforms, which is the gap Athena addresses.

How does Athena handle a claim with no deadline?

It stays open and unscored until the outcome is observable. A claim with no falsifiable target also scores lower on the specificity pillar, because vagueness is itself a behavior worth recording.