Systematic ideas. Inspectable evidence. Explore our algorithmic research across orderflow, options flow and price action—before you decide what deserves your trust.
Historical research. No promised returns. No live execution.
Inspect the research without an account.
5, 10 and 30 admitted sessions. No cherry-picking.
Gaps are disclosed, never silently filled.
Public evidence. Proprietary logic.
Start with a published research stream. Understand its record, its assumptions and its limits. Featured means reviewed for publication—not ranked by returns.
An individual research stream. Inspect its historical record, risk, and admitted-session coverage.
An individual research stream. Inspect its historical record, risk, and admitted-session coverage.
An individual research stream. Inspect its historical record, risk, and admitted-session coverage.
Explore Combined Research Portfolio. Its own published ledger, with a fixed cutoff of Sep 11, 2026. This is historical research, not live account equity.
Membership is fixed at this ledger’s cutoff, not chosen by performance. These windows are descriptive, not holdouts or promotion criteria. Unknown outcomes remain null; no-trade sessions are zero only when known. Fewer than the requested sessions are shown if history is shorter.
Four steps for researching an idea. Not a trading signal or an execution pipeline.
Explore published studies across orderflow, options flow and price action. Understand the market question and the scope of a record before looking at its outcomes.
Research evidence ≠ execution authorization
Read historical and hypothetical results against the study’s stated accounting. A backtest is evidence under assumptions, not a promise of achievable fills or future returns.
Research evidence ≠ execution authorization
Compare records with their limitations intact. A publication receipt establishes snapshot consistency, not an independent audit, equal risk or research-to-runtime parity.
Research evidence ≠ execution authorization
Review published availability and disclosed deliverables before considering access. Research, a packaged specification, an installed plugin and execution authorization are separate states.
Research evidence ≠ execution authorization
The interface is localized. Original research records remain in English to preserve the published evidence.
We publish what the evidence can support. We keep the strategy logic private. And we draw a clear line between research, products and execution.
Read our methodology ↗Public numbers come from sealed research ledgers, not a separate marketing spreadsheet.
Coverage, sizing, costs and cutoffs matter. A strong backtest is not a promise of achievable results.
A study, installed plugin or purchase never bypasses independent execution and account safety gates.