Research log

What we learn, written down

No tips, no signals, no hype. Just the methods, the measurements, and the failures behind how Mercurio is built and validated.

BacktestingJune 12, 2026 · 12 min read

Inside the Mercurio Backtest: How We Turn a Claim Into Evidence

A backtest is a hypothesis, not a result. Here is exactly how Mercurio is validated — the universe, the execution model, the regime filter, walk-forward testing, and the drawdowns we refuse to hide.

Read the deep dive →

Strategy8 min

Why Trend Following Was the Only Strategy That Survived

We built mean reversion, momentum, gap, pairs, and crypto strategies — then tested them across regimes. Only one survived. Here is why focus beat diversification.

Research9 min

We Plugged a State-of-the-Art AI Forecasting Model Into Our Strategy. It Failed.

An open-source market-forecasting model with strong academic metrics. We tested it as a trade filter on real data. It had zero predictive power for our entries — and we show exactly why.

Prefer the evidence?

The writing explains the method; the backtest shows the result.

See the backtest How it works