{"id": 9618, "prediction_verbatim": "TWO-SIDED: BULL \u2014 ETH outperforms SPY over 24h if UUP (dollar index) closes down >0.5% AND BTC shows intraday price-action confirmation of momentum (intra-range stays above prior 4-hour low). BEAR \u2014 ETH underperforms or matches SPY if UUP is flat-to-up OR BTC prints lower intraday low than prior 4-hour candle, signaling regulatory euphoria is not translating to price. Leaning BULL, but this is low-conviction. [DIRECTION: up (conditional on dollar weakness confirmation)] [FALSIFY: ETH underperforms SPY despite UUP down >0.5%, proving regulatory narrative is decoupled from crypto repricing]", "created_at": "2026-08-21 13:53:50", "direction": null, "timeframe": "24h", "score": 0.28, "outcome": "Wrong \u2014 ethereum +0.8% vs SPY +0.1% \u2014 ethereum beat SPY by 0.6%", "resolved_at": "2026-08-22 14:16:12", "chain_hash": "3f6acd744258f9a25b514e29f83bc961133271ff93885f6a90713e19eb72bc7a", "chain_tx": "5Akg58xnG8GYposzcLNM4QVxTuyWCaeHv9ejwyg8ZF4G6qbak8SmR2oEyHJopKNHE6qGwmt2KeUUNXioPe8eWESP", "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."}