strategy-pivot-designer

To detect stagnation during trading strategy backtesting, you append backtest evaluation data to an iteration history file and run a script to analyze the loop for stalled progress or signs of overfitting.

A strategy pivot proposal in algorithmic trading is a new, structurally different architecture generated to break out of a local optimum, using techniques like assumption inversion, archetype switch, and objective reframe.

You generate new strategy architectures to escape local optima by running a Python script that analyzes your iteration history and applies structural transformations to propose alternative strategy designs.

To run the stagnation detection scripts for strategy design, you need Python 3.9 or higher and the PyYAML dependency installed to process your backtest iteration history and trigger pivot proposals.

You should use a strategy pivot proposal when your backtesting iteration loop encounters a plateau, stalls, or shows signs of overfitting, indicating that incremental changes are no longer effective.

Yes, the strategy pivot designer works with existing backtest evaluation files by appending them to a JSON iteration history and passing the strategy ID and recent parameter changes into the detection script.