The relative sizes, percent, and rate/100K are all making VT look worse than it really is, that's how.
Vermont looks "bad" in the rate per 100K because there's nobody in VT generally speaking (apologies to
@Heppenheimer). Assuming my math is correct, VT has about 600K people, total. While FL has 34 times more people. So, if a large family of 6 all catches COVID the VT rate/100K goes up 1. Say, a family that's back from a FL vacation. In FL, to move the rate up 1/100K that family would need to be 215 people.
It's the same problem looking at the percents. Each individual person in VT contributes to a much larger change in percent than a single person in FL. The huge difference in size between VT and FL makes even trying to use a normalized metric questionable.
Vermont did really well early on. They're doing about the same now. The vaccinated percent's ability to reduce transmission isn't linear. Until it hits the inflection point, it's simply not going to do much on the population scale.
That's how I explain the VT graph. That it's simply not as bad as it appears at first glance. That looking at the raw totals and the graph scale not just it's shape becomes important.
VT compared to NH, where NH is still bigger but only twice as big, looks different. That comparison shows that NH did way worse than VT in the earlier wave and they're doing about the same now. Which gives some perspective that it's not VT doing super bad now, it's that they did very good earlier.