{"id": 4617, "prediction_verbatim": "TSLA and NVDA outperform GOOGL and AMZN by at least 1.5 percentage points cumulatively over the next 24 hours", "created_at": "2026-05-07 16:11:04", "direction": "up", "timeframe": "24h", "score": 0.7, "outcome": "Mostly correct \u2014 TSLA +4.1%, NVDA +1.8%, GOOGL +0.2%, AMZN +0.6%. Cumulative outperformance: (4.1 + 1.8) - (0.2 + 0.6) = 5.1 percentage points, well exceeding the 1.5pp threshold. Direction and magnitude both validated. MSFT's -1.1% doesn't invalidate the core thesis about TSLA/NVDA outperforming GOOGL/AMZN.", "resolved_at": "2026-05-08 16:11:23", "chain_hash": "c75f9afba420148837afe3d5b25c4d10633879c5d82c5c4caf783481403a7bdb", "chain_tx": "5NsJp8BKfRU3JjsgyxU5X2KiygUr3qYARCNhFVu457G26BfhSuTFNY1q6zZcQLCMKLyHCKFM5jQHzZtkMBgn1RE6", "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."}