And, of course, this whole discussion assumes Trump cares about crime in D.C., when, in fact, he just wanted to punish D.C. for being a Democratic stronghold. But if we’re willing to pretend that Trump was acting in good faith, we can ask what is the effect of deploying hundreds of federal agents (some for law enforcement, others for immigration enforcement) along with a de facto regiment of 2,200 soldiers to a city of 700,000 people with a land area of 61 square miles.

As before, the raw data can be found in a very nice interactive GUI, but you also can download them here. Here’s the overview, comparing an equivalent length of time immediately before and during the occupation, along with the same dates in 2024:

There might have been a marginal decline in homicides, though they are rare enough and the data are variable enough, such it’s hard to say (there is usually a large decline in crime overall from August to September). It’s also worth noting that homicides dropped in places where there was a presence but also places where there was no presence to speak of, so it’s not entirely clear what happened.

Two categories that dropped dramatically were theft from auto and theft of auto. They declined dramatically in Wards 1, 2, and 6, where there was a large presence of internal security forces. Violent robberies also declined citywide; these seemed declines appear, on the whole, to be in areas very close to where the presence of large internal security forces was reported. That said, assault with deadly weapon really didn’t budge. Burglaries didn’t really change either.

And if you’re wondering why Trump said the stupid things he said about sexual assault, it’s not just that he’s a sexist pig, he also was trying to explain away the dramatic rise in sexual assault. Unfortunately, for Trump, he is a very stupid man.

A final critical point: the first half of the occupation showed much stronger declines than the second half, suggesting criminals were figuring out where they could commit crimes, and this occurred across all classes of offenses.

So did Trump’s occupation work? Well, it worked very well for car-related crimes and worked well for violent robberies, but it did little for other classes of offenses. And the cost to businesses, especially bars and restaurants*, has been high, with much lower traffic. Another cost is far more internal security forces personnel–something Mayor Bowser has wanted for a long time. That certainly doesn’t seem sustainable at the local level, and it also appears to be dependent on density and geographic footprint.

Anyway, those are the data. But Trump certainly didn’t end crime as he put it.

*Bars and restaurants were sacrosanct during the height of the COVID epidemic, not so much now, it seems…