My trend-following basket had a great 2024 in trending markets. The 2025 chop has given back a third of it, and position-level stats show it's the regime, not slippage.
How are people detecting regime shifts systematically — rather than by vibes and a painful equity curve?
ADX percentile plus a realized-vol ratio as a regime classifier. The trend system only runs when ADX is in the top 40% of its one-year range. Simple, and it keeps me out of the worst chop. Not perfect — it lags, like everything — but it turned my 2025 from red to flat.
I tried Hidden Markov Models on returns: 3 regimes — trend up, trend down, range. Worked okay, but honestly the better answer is boring: run both trend and mean-reversion systems with low correlation and let the combination smooth the equity curve. Ensemble over prediction.
The ensemble answer keeps coming up and I keep resisting it because it's unglamorous. Single-strategy, single-regime EAs seem structurally fragile by design. Maybe the real regime filter is a second strategy.