Questions it answers

What breaks a signal.

The point is not to make a model look good. The point is to find out whether the idea survives the first real questions.

  • Does the model beat a simple baseline? If not, stop there.
  • Does the edge survive walk-forward validation? That is the real test.
  • Do costs and slippage erase the gain? Usually faster than people expect.
  • Is the threshold just decorating noise? The simulator makes that visible.