Washington, DC · 2025 to 2026

DC crime fell after the National Guard arrived. It fell just as much where there were no troops.

Homicides and gunfire fell by close to half in the first Guard year. Areas near troop posts did no better than comparable areas, and the data can't say what drove the citywide drop.

On August 11, 2025, the National Guard deployed in Washington, DC. The same day, federal agents arrived and the federal government took control of the Metropolitan Police Department for a month. This report asks two questions: did crime fall, and did it fall more where the troops were?

Crime fell

In the first Guard year, August 11, 2025 to August 10, 2026, reported crime was 34% below the average for the same dates over the previous 10 years. Some of that was already under way: the year before, it was 11% below. The drop was largest for violent crime involving a gun, 54% below average against 19% the year before, and for homicide, 46% below against 5%.

Two measures don't depend on anyone calling the police. Homicides fell from 167 to 94, and gunshots picked up by sensors fell 44% from the year before. From September to June, DC's homicide drop was the second largest of 36 large police departments: 51%, against 15% nationally.

Not all of it was new. Modeled as a change at the deployment, homicide dropped by 48% and property crime by 25%, but overall violent crime shows a change of 12% that can't be told apart from no change, because it had already been falling.

It did not fall more where troops were

The Guard's posts were mostly downtown: at Metro stations, on the National Mall, at Union Station and Navy Yard, and along H Street, 14th Street, Dupont Circle and Georgetown. In fall 2025, gunshots fell 57% in the two wards holding nearly all the posts and 60% in the wards without them. When the Guard began patrolling Anacostia in 2026, gunshots there fell 35%, against 40% in neighboring Ward 7, which had no documented patrols.

Before running more detailed tests, the analysis published a plan that fixed the areas, dates, measures and what would count as an effect: a reduction of 15% or more near posts. Across about 800 half-kilometer squares, crime near posts fell no more than in comparable areas on any of six measures. For property crime, the results rule out a reduction of 15% or more. For gunshots and gun crime, the range is too wide to rule out a modest reduction, and it also allows an increase. Changing the radius, the start dates or the comparison areas doesn't change that.

Two results point the other way. When 12 Metro stations were drawn at random 1,000 times, 93% of the random sets saw gunshots fall more than at the real Guard stations; the Guard may have chosen stations whose problems were growing. And homicide, robbery and sex abuse near posts ran 43% above comparable areas with the planned list of posts, and 30% above with a fuller list the Guard filed in court. With the planned list, that rise holds up after correcting for testing six measures; with the fuller list it does not. About half of the gap was opening before the troops arrived.

The surge as a whole

Compared with a "synthetic DC" built from other large cities' trends, DC's property crime ran 22% lower from September 2025 to June 2026, a break only 3 of the 92 comparison agencies matched. Murder ran 27% lower, which could be chance (p = 0.10), and robbery showed no change. That measures everything that arrived on August 11 together, not the troops alone. When troop numbers fell from 5,148 in July 2026 to 2,864 in October, citywide crime did not change against normal.

What this does not show

• What caused the decline. The data shows violence fell and that the fall does not follow the troops. Federal agents, the police department's own work, a trend already under way, or some mix could explain it.
• Exact troop locations. There is no official map of Guard posts. Locations come from news reports, Guard releases, a Guard update filed in federal court and a court opinion. Records requests for the post list are drafted.
• Every crime. MPD's crime data is under investigation for downgraded reports, which is why this analysis leans on homicides and gunshots. People avoiding police during immigration enforcement, or federal agencies taking reports that never reach MPD's data, would also lower counts. The National Mall is mostly patrolled by the Park Police, whose records are not in this data.