A cemetery plot-demand model recites last decade’s odd memorial-week spikes and then fails on next month’s unseen weeks. What should testers name that behavior?

Reciting last decade’s odd memorial-week spikes and then failing on next month’s unseen weeks is overfitting: the model captured noise rather than the underlying pattern and does not generalize. Underfitting is poor on both slices, concept drift is a changed input-to-label relationship, and a missing hardware note is a card-review finding.