In this (simplistic) example, are a handful of threats (A, B, and C) that are causing far more damage than the rest. They are the current top threats which our theoretical company needs to focus on mitigating the most. Although C seems to be projecting downward and D seems to have an upward trajectory. Perhaps now, or if those trends continue to track identically over the next measuring period, D gets added as a top threat and C gets removed. The general idea is that we care about the top current damage sources and the most likely future top damage sources. It’s up to the defenders about how much effort goes into mitigating the rest of the threats (E, F, G, and H).
Of course, in real life, threat cycles, like stock market gains, don’t always go up or down in a straight-line, like the simplistic lines shown above. But like a longer-term investor, what a datadriven defense mind is interested in is the general trend. Although just like a stock market investor cares, if a trend line turns against its previous trend too sharply or too often, then it needs to be investigated for cause.