As discussed in the bullet points, those incentives create a sort of natural selection which changes how the statistics are reported over time. It's not like w

As discussed in the bullet points, those incentives create a sort of natural selection which changes how the statistics are reported over time. It's not like we are measuring length or temperature. The data has a degree of human subjectivity when being recorded. It seems normal for that subjectivity to shift over time, especially when we are taking about different humans over timespans of decades.