We obviously are operating from different data sources. Even the original covid rate had an average fatality rate of 1%. We don't have great numbers yet for Delta, but there are already clear signs that the risks are much higher for younger people now.

And I wouldn't get too caught up in the craziness associated with "car accidents skewing the numbers" theories that I've seen promulgated here. Just think about it: 38.6K people died from car accidents in 2020. Just statistically we can see that only a very few of them would have covid. So even assuming 10% of those who died were diagnosed as covid positive, that's not going to move the needle much even if every single one of that group was actually counted as a covid death (which I don't think is factually correct, at least for US numbers, as based on my reading of CDC reporting guidelines to hospitals at that time, a causal link showing covid was a contributing factor to classify them as a covid death: so someone who tested positive but was asymptomatic wouldn't be classified as a covid death, as a simple example).