Trust the process.
Inspect the evidence.

One source of truth.

VisionAlgos is a separate, allowlisted projection of the same sealed research publication used by QuantNQ. It does not recalculate performance for sales copy or pull a live trading account feed.

The current public catalog contains 1,224 ledger views. The latest selected publication cutoff is Sep 11, 2026, but individual studies may end earlier. The cutoff printed on each record is authoritative for that record.

Sources and generated detail files are verified by byte count and SHA-256. These hashes prove snapshot consistency, not independent authenticity, an audit opinion, or research-to-runtime parity.

Unknown is not zero.

Calendar membership comes from admitted research sessions, not inferred weekdays. A missing complete-session outcome remains null. A resolved subtotal can exist on an otherwise unknown day without making that day complete.

Whole-window P&L is unavailable if any admitted session is unknown. The chart breaks and resets its cumulative path at every unknown session. Known no-trade or flat sessions remain in the record. Session shares use all admitted sessions, including unknowns.

No handpicked recent streaks.

Every detail record exposes full history and the fixed last 5, 10, and 30 admitted sessions ending at that record’s cutoff, including exact dates and retained daily rows. These windows are descriptive only: they do not select a strategy, promote a candidate, or establish untouched forward evidence.

Dollar economics use each study’s published accounting. Costs, contract sizing, exposure, calendars, and sample sizes vary. Public aggregates deliberately omit private parameterization, so they are not sufficient to recreate a strategy or independently model brokerage returns. They are not capital-normalized returns, a promise of achievable fills, or proof that two strategies share the same risk.

Trade profit factor and daily profit factor are different measures. Missing support stays null; a supported positive, loss-free profit factor is shown as infinity. Win rate uses wins divided by wins plus losses, excluding zero-P&L outcomes. Best-day contribution can exceed 100% or be negative.

Maximum known-segment drawdown resets at unknown sessions and is observed at the published daily-close granularity. It is not intraday drawdown, worst-case risk, or a funding requirement. Where event economics are not supplied, event metrics remain unavailable rather than being inferred from daily totals.

A component is not a contribution.

Each portfolio is a distinct accounting authority. Standalone results cannot be summed into an overlap-accounted parent. A linked component may have a separate cost variant from its featured standalone view; exact variants retain separate public identities.

Options-flow alternatives are not an options-inclusive joint portfolio. Unevaluated combinations are listed as unavailable. A comparison page neither sums nor optimizes its selected records. Featured status reflects an explicit publication selection, not a profitability ranking.

Evidence, not the recipe.

The public app receives neutral public identities, approved labels, daily aggregate outcomes, window metrics, and receipts. It does not receive execution rules, parameters, entry/exit timestamps or prices, stops and targets, raw trade ledgers, research source paths, broker credentials, Pine code, private reports, or provider ladders.

Aggregate economics still disclose historical behavior and could support inference. Publication requires owner review and appropriate redistribution rights; sanitization is not a guarantee that all strategy characteristics are unknowable.

Research does not place orders.

Packaged references, portfolio and monitoring specifications, installed execution plugins, and practice-authorized products are separate states. No catalog item on this site is practice-authorized or live-authorized. Historical results are not attributed to a related runtime plugin without exact parity.

Checkout, when enabled for a specifically approved research offer, buys only the disclosed research deliverable. It never activates the trading bot. Practice execution would require a separately reviewed immutable plugin, server entitlement, verified practice account, local enablement, reconciliation, and centralized risk approval. Funded and live execution remain prohibited.