Well, I could be more opinionated, but even in very specific situations, reasonable people disagree about the best way to model data, and I don't really know a lot about your specific problem-space or situation.

My personal preference is to think of almost any changing measurement or event stream as a time series. See also the reply to a sibling comment.