Eight experiments. Five rungs. One engine demolished, one audit framework
earned. Run any experiment live, then read the writeups. Everything runs
against a frozen 9-symbol OHLCV snapshot bundled with the study โ
no API calls, no credentials, so the results are
reproducible bit-for-bit.
Data: 9 symbols ยท
frozen at 2026-06-18 ยท
engine = binary_exhaustion_engine ยท
verdict = z โ +0.6 vs matched-bias null (null).
The 5-rung audit ladder
Rung 1 โ Wrong payoff. The original strategy "won" in shares-only PnL but lost in realistic option-premium PnL. (EXP1, EXP3)
Rung 2 โ Sampling noise. A single coin-flip control arm has its own variance. Sweep 200 seeds to characterize the null. (EXP4)
Rung 3 โ Wrong null. A long-biased strategy beats a fair-coin null structurally. Match the null's bias to the strategy's. (EXP5)
Rung 4 โ Autocorrelation. 494 fires of overlapping 10-day windows aren't 494 independent observations. Filter to non-overlap. (EXP6, EXP7)
Rung 5 โ Untested detector. A framework that has only ever said "no edge" might be too conservative. Feed it a known oracle. (EXP8)
Verdict on the engine
z = +0.6 to +1.0 vs matched-bias null at every honest test.
Apparent z = +3.83 at T+10 collapsed to z = +0.60 once autocorrelation was removed.
Worst 20-bar drawdown windows: engine z = โ1.25.
No detectable directional skill in any framing or regime.
Verdict on the framework
Detects 55% synthetic oracle at z = +2.50 on n = 675 โ
Nulls a 50% oracle at z = โ0.16 โ (no false positive)
Subtracts long-bias from a no-skill biased oracle โ
Two-way meter. Bedrock.
Experiments
EXP1 v2 โ Shares vs Options (premium-correct)
Premium-correct backtest with a random control arm. Establishes that the raw 'shares-only' framing was an artifact of mispriced options.