owner performance.
two distinct records, never conflated. the calibration backtest below is the signed, independently-verified aggregate across the six q7 aegis ai strategies. the live trading record reads straight from the connected broker account — and stays honestly empty until real fills land.
all calibration numbers net of modelled commissions and slippage. live numbers are realized fills, refreshed from canon-api.
Trade attribution by algo
Signed calibration backtest · net of modelled commissions & slippage
Attribution reflects the signed calibration backtest, net of modelled commissions and slippage. Past backtested performance does not guarantee future results. Live attribution appears here once real fills land.
risk governance
Realized equity read against the program’s hard −24% drawdown ceiling — the AI-governed risk floor, rendered rather than described. The governor is armed from day one, at $0 of P&L.
Each chip is a strategy’s live hybrid-AI gate state. No per-strategy gate feed is connected yet, so every chip is shown awaiting — the honest state, not a fabricated pass/fail.
data table (accessible equivalent)
| # | realized equity Δ | drawdown | headroom to ceiling |
|---|---|---|---|
| No closed live trades recorded yet. The 24% ceiling is armed and the frame is presented; the curve populates from real fills. | |||
source · equity & drawdown built from canon-api /api/connections/trades + /api/connections/nt/status (real closed-trade nets over the connected account). The −24% line is the program’s hard drawdown ceiling enforced by the daily circuit breaker — a governance rule, not a performance claim.
audit window 2025-05-14 → 2026-05-13
14-day rolling · public verification gate
activates when founding cohort signs on · owner account routes real trades
audit window 2025-05-14 → 2026-05-13·calibration data: tick-replay·forward test & live: launching with founding cohort·backtest aggregate — pending independent signed calibration artifact
calibration basis — each configuration is backtested on its own $10,000 of starting capital over a 1-year (2 years for the long-history strategy) window, net of modeled commissions and slippage. Aggregate figures sum across 16 configurations (16 × $10K = $160K deployed). Past backtested performance does not guarantee future results.
calibration headline.
6 strategies × 16 configs each. every single one passed the production gate. zero engine × timeframe combos retired.
net P&L is backtest result on the calibrated parameter set. live forward-test is the next gate before subscriber routing. whitelist retirement = engine × timeframe combo dropped after 2 consecutive Optuna rounds with PF<0.90 ∨ WR<45%.
source · calibration figures from the signed AEGIS AI strategy contract; live figures from the connected NinjaTrader account via canon-api /api/connections/nt/status + /api/connections/trades. Administered by Quant7 Alpha, LLC. Hypothetical and simulated results have inherent limitations and are not indicative of future performance.