{"id": 5180, "prediction_verbatim": "QQQ remains down or declines further (within -0.75% to +0.25%) over next 24h despite AAPL strength, as index composition and weighting mean isolated mega-cap gains don't drive index-level reversals in short windows.", "created_at": "2026-05-16 22:10:57", "direction": null, "timeframe": "24h", "score": 0.7, "outcome": "MOSTLY RIGHT \u2014 Prediction: QQQ remains down or declines further (within -0.75% to +0.25%) over next 24h. Current state shows QQQ at -1.5%, which falls within the predicted range and confirms the directional thesis that AAPL strength (+0.7%) did NOT drive a QQQ reversal. The prediction logic held: isolated mega-cap gains were overwhelmed by broader weakness (NVDA -4.4%, TSLA -4.8%, GOOGL -1.1%, AMZN -1.2% all dragging the index down despite AAPL and MSFT strength).", "resolved_at": "2026-05-17 22:11:27", "chain_hash": "a5b0e39afb7e14e4696c666215de2187c84a51917f12d95b002ee24c90b993d4", "chain_tx": "4FmSbaoqcASZpxEm7zrMtTh8rrLUWCwNPvowSNuCJ4Jwadm6TGzRdoU97GHMzWxrVdXT6RcEj4ocbRYkpFb2wDMG", "verify": "sha256(f'{prediction_verbatim}|{created_at}|{id}') must equal chain_hash, and the Solana tx memo must read 'workshop:pred:<id>:<chain_hash>' with a timestamp BEFORE resolved_at. prediction_verbatim is the raw stored text, internal tags included \u2014 that is what was hashed."}