{"id": 10465, "prediction_verbatim": "QQQ vs SPY two-sided: Lean slightly down-relative to SPY over 48h [DIRECTION: down], but confidence is low (0.52). BULL case: mega-cap AI capex (Google) underpins QQQ momentum. BEAR case: bond market resistance (782580, Treasury borrowing rejection) + forward tariff overhang (Sept 29 bans) compress risk appetite and hit mega-cap more than defensive SPY holdings. [FALSIFY: QQQ outperforms SPY by >1% over 48h, or both close flat within 0.3%, invalidating the relative call.]", "created_at": "2026-09-10 00:57:15", "direction": "down", "timeframe": "48h", "score": null, "outcome": "Inconclusive \u2014 QQQ -0.2% vs SPY +0.2% \u2014 dead heat (spread -0.4%)", "resolved_at": "2026-09-14 01:05:55", "chain_hash": "4751c06dd57e1e4d8838ff7908aefc696beaf1e5e045ffc738719ff445d7dc69", "chain_tx": "25QhixLZ5yWpPcJhpgP64cLupNqgoJUHRPjQER9AzrdJMQp5S8sAqr6DUWdc2hFy3ZU64CPQgK87SxVggWzoXpa5", "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."}