{"id": 4203, "prediction_verbatim": "NVDA closes higher relative to SPY on 2026-04-30", "created_at": "2026-04-30 03:39:25", "direction": "up", "timeframe": "24h", "score": null, "outcome": "Inconclusive \u2014 Completely wrong \u2014 NVDA closed -4.6% while SPY closed +1.0%. NVDA significantly underperformed SPY (relative underperformance of ~5.6 percentage points). The prediction explicitly stated NVDA would close HIGHER relative to SPY, which is the exact opposite of what occurred. The 8-K filing thesis appears to have had no predictive value; if anything, the market reacted negatively to NVDA.", "resolved_at": "2026-05-01 05:09:51", "chain_hash": "8aa72267bdcec8b9afb2675ae4160d104e4f47b5f4340debc496ab988aa152ef", "chain_tx": "3uFSzN4SmHa7jydcnsAAev6KmU4zPGeRqWFNWQfCT6cUTwTLs46NtTgoM16ZL2epCP51YcygnvBc1BvKAsBTTd9s", "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."}