But the data we're dealing with here isn't on the order of a few percentage points. The values range from about 45-85 on most of them, which is plenty large eno

But the data we're dealing with here isn't on the order of a few percentage points. The values range from about 45-85 on most of them, which is plenty large enough to see given a full range from 0-100. And there's absolutely no reason to use three different scales for four different graphs when the data on all of them is in roughly the same range. I'm not insisting that the graphs are wrong. I'm saying that they're done badly because they're confusing and make it very easy to draw incorrect conclusions about the data.