memory gap
the before-state disappears
A market call gets judged later, but the original sources, assumptions, and confidence often vanish into the feed.
@evapredicts campaign · prediction memory
prediction markets price the moment. Eva remembers the thesis.
The public prediction feed is full of odds screenshots and confident takes. Eva turns a market call into an inspectable record: claim, cited signals, revision triggers, and an author trail people can check after the timeline moves on.
Campaign hypothesis
prediction-memory framing should pull higher-intent builders than another launch announcement.
Prediction-market-native readers do not need Eva to explain that odds exist. The sharper pain is memory: what did this person think, why did they think it, and what changed later? The campaign asks them to inspect one concrete thesis record before following the public loop.
memory gap
revision logic gets lost
Good predictors change their minds. Readers need to see what fact or market signal forced the update.
memory gap
agent outputs need receipts
Forecast automation only becomes useful when the output carries sources, runtime identity, and an audit trail.
target audience
people whose predictions need to survive the feed.
public predictors who already explain their market calls on X
Send them to the proof record first, then measure whether they inspect sources or start a thesis.
prediction-market operators trying to improve distribution quality
Send them to the proof record first, then measure whether they inspect sources or start a thesis.
crypto analysts whose broad theses need a record readers can revisit
Send them to the proof record first, then measure whether they inspect sources or start a thesis.
agent builders who need forecast outputs with receipts before automation
Send them to the proof record first, then measure whether they inspect sources or start a thesis.
Campaign sequence
one memory claim, one proof object, one measurable decision.
- Send approved @evapredicts traffic to this page with utm_campaign=prediction_memory.
- Route the first click into the SpaceX proof thesis so readers inspect a real record, not abstract positioning.
- Keep the angle only if it beats generic launch copy on proof-thesis reads, source-quality page visits, compose starts, and follow clicks.
@evapredicts copy to approve
make memory the public wedge.
External posting still needs explicit approval. Until then, this page is the live destination and the copy below is approval-ready for @evapredicts.
prediction markets are good at showing what people think now.
but they are bad at remembering why anyone thought it.
that is the wedge for Eva: public market theses with cited signals, revision history, and an author trail.
not louder calls. better memory.
first proof object: https://eva.jaack.me/campaigns/prediction-memory?utm_source=x&utm_medium=social&utm_campaign=prediction_memory&utm_content=memory_post
Metric to watch: sessions with utm_campaign=prediction_memory, proof-thesis reads, source-quality page visits, compose starts, and @evapredicts follow clicks. Do not claim traction until those are measured.