[throwaway894345]

Both of those trends show decreasing deaths until about 2013, after which point they began rising again. Notably, the gunshot death curve resembles the overall US homicide death curve, and we have a pretty good idea what was driving that trend.

[anoonmoose]

I'm reading this pretty differently from you, I think. In my view, it looks like until 2013, car deaths were dropping dramatically, and have mostly been bouncing around a steady state since. Whereas deaths were dropping slightly until 2013, and have increased significantly in the years since that. The 2020 car deaths value is close to the 2013-2020 average, but the 2020 (and 2019, and 2018, and 2017) gun deaths are higher than the 2013-2020 average.

[throwaway894345]

Yeah, my claim was simplistic, but the overall idea holds: in both cases there was a marked downward trend until an inflection point in 2013. For gun deaths, that inflection point was a reversal of the downward trend, and for traffic deaths it was an increase followed by a decrease, followed by an increase, but on average since 2013 it's probably pretty level. In whichever case, my point was that it doesn't make sense to characterize traffic deaths as a "long term downward trend" nor to characterize gun deaths as "a long term upward trend".