A paper-trading app and the earnings research behind it, in which twelve promising results died to their own controls and none survived.
| Claim | Raw | After its control |
|---|---|---|
| Intraday lead–lag | 71% accurate | 65% for random signals |
| Opening-range breakout | profitable | −31% vs QQQ +263% |
| Post-earnings drift | t = 3.58 | t = 1.20 beta-adjusted |
| Pre-event news | ρ = 0.432 | ρ = 0.077 within symbol |
| Straddle price gate | +30.6% modelled | t = 0.16 on real prices |
| Per-symbol cheapness | +62% | p = 0.080 vs best-of-8 |
Gap direction is unpredictable — six pre-event signals, 238 filing-anchored events, 45–53% against a 50% null. And range compression forecasts move size (ρ −0.261), though it predicts the median rather than the tail an option prices. Both were stated as nulls from the beginning.