All Posts (Page 27)

An adaptive trading system continuously updates its assessment of which models are suited to current market conditions, rather than deploying a fixed strategy based on historical optimisation. In darwintIQ, this adaptation is achieved through continuous model evaluation on a rolling window — meaning model rankings reflect what is working right now, not what worked at some point in the past.

Mean reversion and trend following are the two dominant systematic trading approaches. They succeed in almost opposite market conditions, which means neither has a permanent advantage. Understanding how market regimes create and destroy edge for each approach is the foundation of building models that remain functional across changing conditions.

Trending and ranging markets represent the two fundamental states that price can occupy, and they call for entirely different trading approaches. Understanding the distinction — and recognising when a market transitions between the two — is foundational to evaluating why trading models behave as they do and when their edge is most likely to hold.

The Stability Score in darwintIQ assesses how evenly distributed a trading model's returns are across the evaluation period, rather than whether returns exist in aggregate. A high Stability Score indicates consistent contribution throughout the window; a low score reveals that performance may be concentrated in a narrow portion of the period — a form of fragility that headline metrics can conceal.

The SupRes Position Manager is one of the position management components available in darwintIQ trading models. Rather than applying fixed pip distances or volatility-scaled rules, it uses identified support and resistance levels to determine where trades should exit — aligning trade management with the structural logic of the market itself.