Regime Detection,
Autonomous Execution,
Published in Real‑Time
The System
→OVRWCH monitors a universe of liquid ETFs across equities, fixed income, commodities, currencies, and real estate. Every trading day, a Hidden Markov Model classifies the market regime. A walk-forward ML ensemble scores each asset. A three-layer defense system protects capital. Trades are executed automatically and published with full reasoning.
Detect
A Hidden Markov Model identifies which of four regime states the market currently occupies. The classification is based on a proprietary set of macro features and is updated daily. The model identifies present conditions — it does not forecast future returns.
Position
A walk-forward machine learning ensemble scores the full ETF universe using quantitative signals spanning momentum, volatility, macro conditions, and proprietary alternative features. The model retrains monthly and has never scored data it was trained on. Allocation is regime-conditional — the portfolio composition changes with the market environment.
Defend
Three independent defense layers protect capital simultaneously: position-level circuit breakers, a portfolio-level drawdown controller, and adaptive volatility targeting. The defense system operates independently from the alpha engine. It cannot be overridden.
Publish
Every position, every trade, and every regime classification is published in real-time. Performance numbers are walk-forward and out-of-sample exclusively. The system's track record is public, auditable, and honest — including its drawdowns and failures.
Performance
→OVRWCH vs S&P 500 — rebased to 100 at inception (April 10, 2026)
| Metric | OVRWCH | S&P 500 |
|---|---|---|
| Inception | April 10, 2026 | — |
| NAV | $100,033 | — |
| Return (Live) | +0.03% | — |
| Max Drawdown (Backtest OOS) | −13.2% | −34% |
| Sharpe (Backtest OOS) | 0.38 | ~0.45 |
| Days Live | 1 | — |
Paper trading since April 10, 2026. Backtest metrics are walk-forward out-of-sample (2013–2026). Updated daily at 4:00 PM ET.
OVRWCH publishes system performance for informational and educational purposes only. Not investment advice. Past performance does not guarantee future results. Paper trading since April 10, 2026. Full terms.