Every dealer-positioning product on the market publishes the same thing: a picture of where gamma sits. None of them publishes whether acting on it worked. This one takes the last step — dated calls with levels drawn from the mechanics rather than round numbers, and a forward-only public record of every one. The stop sits beyond the accelerant band, because a stop inside it is placed exactly where forced hedging will chase it. The target is scaled to how much of a session's volume a 1% move forces through dealer hands. And when the regime that justified a call inverts, the call closes on its own terms — reported in its own bucket, returns published, never quietly dropped.
Price exits — target, stop, or the 30-day horizon — are graded by the same rule that grades analysts, agents and user forecasts on this platform, so this record is comparable to every other ledger rather than scored on a private curve. Calls whose regime inverted before price resolved them are counted separately, with their returns, and the all-in average includes them: no exit path can flatter the headline.
The question an allocator asks after seeing a track record: given the edge you have actually demonstrated, what does the next thirty calls look like? This resamples the resolved returns with replacement — no normality assumption, no fitted alpha, no compounding — and draws the percentile envelope of cumulative outcomes. It is deliberately refused below a minimum sample, because a fan chart drawn from a handful of trades is a picture of noise wearing the costume of analysis.
A conviction score exists to answer exactly one question, and almost nobody publishes the answer: do the calls rated highly actually win more often than the ones rated low? This is that chart. A bucket with no resolved calls reports nothing rather than a flattering guess, and if the curve comes out flat — or inverted — it stays on the page. A score that does not discriminate is worth knowing about.
The same record cut by the mechanism that produced each call. Momentum calls ride a short-gamma tape; grinds fight a long-gamma one — if only one of those carries the record, that is the finding, and it should change what the engine is allowed to fire.
The positions dealer mechanics justify right now, with the reasoning that produced each one.
The discipline that makes the record mean anything. This engine only fires when dealer mechanics genuinely condition the outcome — when the book is immaterial against ADV, when the flip is close enough to invert intraday, or when the two sign conventions disagree about the regime, it declines and says so. A positioning product that has an opinion on every name is the 5-factor signal wearing a costume.
Nothing is backfilled and nothing is removed. Regime-closed rows show the return they carried at the close.
| Opened | Underlying | Setup | Entry | Exit | Return | Result |
|---|