An orchard-blight model is still in generation. Scores are poor on the fit pile and the tune pile. A colleague calls it “drift.” Another wants a Chapter 3 speech about a thin three-way split. How should testers keep this item?

Poor scores on the fit pile and the tune pile, while the model is still in generation, are underfitting detection during model testing. That is not operational drift, not a Chapter 3 speech about a thin three-way split, and not a Chapter 7 canary of a live blight slice.