Huh. TEVUS submission went through. Need to poke around a bit to find out exactly what I'm not supposed to mention sans permission, but, uh. Even in like three minutes of use this thing is pretty neat. Hopefully it's general enough to mention that apparently a full half of left-wing domestic terror incidents since... '74, if I'm parsing this right. Occurred in three states. Over a quarter in one. Right wing is rather notably more spread out, though two of its most hit areas are shared with that top three, if not in the same order.
Left wing incidents also seem to be notably better... understood, I guess. Right has unknown for location in its top three, whereas it's down a bit for the left. That's somewhat surprisingly unsurprising, I think. Common sentiment is that stateside enforcement is notably more lackadaisic about right wing extremism, even when it's significantly (like, around twice vis a vis what tevus's sources cover) more likely, and at least on the face of it that kind of lack of information is pretty much exactly what you would expect if that were accurate.
Now we poke around the time related stuff and... jeez, what the hell happened in '93? I'd actually guess it was some sort of change in tracking and enforcement, but incidents bloody doubled that year, while general crime was starting its downslope.
... anyway, at least so far as what these folks track, the answer to if extremist acts are trending towards an increase is "Definitely not." Shit spikes up and down, doubles and halves what appears to be fairly randomly (i.e. the trend is a lot less trend-y), and largely looks to be roughly on par with early (pre-93) 90s numbers in terms of events in a few years leading up to 2015 (haven't actually finished poking things to get a full per-year count, 'cause I'm damn hungry and about to go feed myself, but I've tallied the bits from the start of what I'm first checking (period from 1990 to their latest, which is 2015) and the end), which is, uh. Our 2015 population was about 150% of what it was in 1990, up by a handful of millions short of a full 100 of 'em. Would suggest that to some degree the extremist ends of political violence are actually dropping, on the net.
Which would actually sorta' be what you'd expect, given the general crime trend, though parts of the extremist timeline doesn't even remotely mirror the general one. Interesting stuff. Entirely too specific for what I was curious about, heh, but still interesting.