{"id": 2059, "prediction_verbatim": "Semi-conductor and EV-adjacent ETFs (e.g., SMH, XLI) outperform SPY on 2026-04-03", "created_at": "2026-04-02 10:03:39", "direction": "up", "timeframe": "24h", "score": 0.2, "outcome": "WRONG \u2014 Prediction: Semiconductors and EV-adjacent ETFs (SMH, XLI) outperform SPY on 2026-04-03. Data shows: SPY +0.1%. We lack direct SMH and XLI prices, but proxy indicators suggest underperformance: NVDA +0.9% (beats SPY marginally), but TSLA -5.4% (severe EV sector underperformance), MSFT +1.1% (semiconductor-adjacent but broader cloud/AI play). The EV component collapsed; thesis about China EV demand rebound and diplomatic restraint did not translate to outperformance. Cannot score 0.5 due to TSLA's explicit weakness in the EV thesis.", "resolved_at": "2026-04-03 10:06:46", "chain_hash": "36284b2e3879a3c520ae4661fade42811e364b0b130e92c81f63d61f11478130", "chain_tx": "2z7ECB4PAS5JXasfYEu3XMYAdMYYFdqS9uHytKamBL8YxG1H3go1DFydURom6u9iwCMi3RpcNg4hvxXdu48qKFTw", "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."}