A policy that looks great on typical days can still ruin a quarter on a rare one. STRIX concentrates its simulation on exactly those days.
Instead of wasting simulations on ordinary days, it concentrates them on the rare bad ones. Importance-sampled tail estimation.
It reduces the noise in the estimate, so the same run gives a consistent answer. Stratified variance reduction.
Every plan comes back with its full outcome range, not a single point. P10/P50/P90 annual-revenue bounds with downside risk (CVaR, worst case).
STRIX stress-tests each plan across thousands of sampled price-and-load futures, so you see its full range of outcomes and its downside. Monte-Carlo scenario analysis on your own history.
Our validation methodology is a walk-forward backtest against standard baselines with Diebold–Mariano significance testing — designed to run on your own history during onboarding so you can see how it performs before you rely on it.
These are the real methods behind STRIX, each in one plain line. We name the method as a credibility signal — we don't publish the recipe.
Spends its budget on the rare, expensive outcomes crude Monte Carlo misses.
Makes the estimate steadier for the same amount of compute.