{"id": 10383, "prediction_verbatim": "TWO-SIDED: Bull case: QQQ outperforms SPY over 48h (rate-cut repricing, duration relief, capex resilience). Bear case: QQQ underperforms SPY over 48h (tariff headwind, layoff ripple, crisis regime persists). LEAN: Bull, but confidence = 0.50. [DIRECTION: up (QQQ vs SPY, slight lean)] [FALSIFY: QQQ underperforms SPY by >0.2% over 48h; equivalently, SPY closes +0.6% and QQQ closes +0.3% or flat]", "created_at": "2026-09-04 02:10:03", "direction": null, "timeframe": "48h", "score": 0.28, "outcome": "Wrong \u2014 QQQ +0.2% vs SPY -0.4% \u2014 QQQ beat SPY by 0.6%", "resolved_at": "2026-09-06 02:28:32", "chain_hash": "f273ed463a4368038402304420052ec654cf0073b0c8e57e4964be0f2c14b363", "chain_tx": "2a6LhhWHe99sry59J773GpptrMjUa7pUbptGK8KkRMyojZPwpGiokAg1HbCLt2UQGuCs4sdhWkeVqGwqQnKiVUdf", "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."}