Regime Detection: How HMM Classifies Market States

Breakout logic and mean-reversion logic can both be correct, but rarely at the same time. Hidden Markov regime detection gives the strategy layer a compact context: trending, mean-reverting, volatile, or transitional.

Chimera combines regime probability with Hurst persistence and live order-flow checks. The result is not a prediction of the future; it is a current-state filter that prevents the wrong strategy family from dominating.

The public site exposes only high-level regime state. The desktop app shows deeper model and confidence details after login, where the data is useful for operators rather than competitors scraping public pages.