Another use case is the probabilistic structures. Usually people assume that answering a question is deterministic. As I showed in Chapter 2, Data Pipelines and Modeling, in many cases, the true answer has some uncertainty associated with it. One of the most popular ways to encode uncertainty is probability, which is a frequentist approach, meaning that the simple count of when the answer does happen to be the true answer, divided by the total number of attempts—the probability also can encode our beliefs. I will touch on probabilistic analysis and models in the following chapters, but probabilistic analysis requires storing each possible outcome with some measure of probability, which happens to be a nested structure.