On the shifting paradigm of prediction: A Countinghouse primer