Take the LEAP
A rules-based AI screener hunting rare, asymmetric LEAPS setups in quality companies caught in panic — and honest enough to tell you, most days, that the right trade is no trade.
One email per market day: the macro read, what got screened out and why, and the rare candidate that survives every gate.
Second consecutive day the RESTRICTED regime's raised threshold (80 vs 75 normal) is the binding constraint, not a lack of real dislocation: MDB's S11 total would need only 2 more points (any 2 of its 7 already-scored dimensions, or a mid-single-digit option_implementation read) to clear at the RESTRICTED bar, and would clear easily at the NORMAL bar (75). state/macro-latest.json's R=4/RESTRICTED reading is now 2 days stale (as_of 2026-09-02) -- a macro-refresh run (desktop-only) is needed before this throttle reading can be treated as current; until then it is being applied conservatively (fail-closed, not re-derived) rather than assumed to have eased.
See the full record →Why the restraint is the product
Every session runs a hard gauntlet — a macro throttle, event and quality gates, four entry-pattern checks, and an eight-dimension scoring bar — and most names die before anything gets priced. A candidate only reaches your inbox by surviving all of it, scoring above the bar, and pricing out to a positive expected value after real friction.
The record so far: 7 sessions, 0 forced trades, 70 gates fired. The methodology, framework versions, and every architecture decision are public — this is a working example of disciplined, AI-assisted research, not a tip service.
⚠️ Not financial advice — read this
Take the LEAP publishes the output of a rules-based AI research system. It is NOT financial advice, and nothing here creates an advisory or client relationship. The system can be wrong — often, and confidently. Signals are hypothetical research output, not recommendations to buy or sell any security. Options are leveraged instruments that can lose 100% of the premium paid. Hypothetical performance does not reflect real trading (no slippage, assignment, or taxes) and past results never guarantee future returns. Do your own research and consult a licensed professional before making any investment decision.