District 3 (3D): stop representation
3.33×
In District 3, Black people were stopped at 3.33× their share of the district's own population (73.4% of known-race stops vs 22.0% of residents).
- Black stop share (known-race stops)
- 73.4%
- District's local Black population
- 22.0%
- Black stops this rests on (N)
- 23,511 of 32,037 known-race stops
Show citywide comparison
The citywide index compares each district's stop share to DC's overall 43.3% Black population. It runs high for majority-Black districts simply because those areas are majority-Black, which is why the local index (comparing to each district's own population) is the honest measure of disparity.
District 3's citywide baseline index is 1.70× (Black stop share vs DC's 43.3% Black population); citywide overall is 1.82×.
The disparity over time
District 3's local index, year by year: the same metric as the headline above, computed on each year's stops alone.
* 2019 is partial (from Jul 22), 2026 is partial (through Jan 1). MPD only began NEAR Act stop-data collection on Jul 22, 2019; no stop data exists before that date.
| Year | Black stop share | Black stops (N) | Local index |
|---|---|---|---|
| 2019* | 64.8% | 5,735 | 2.94× |
| 2020 | 62.8% | 7,515 | 2.85× |
| 2021 | 65.6% | 6,457 | 2.98× |
| 2022 | 69.5% | 7,906 | 3.15× |
| 2023 | 68.1% | 7,695 | 3.09× |
| 2024 | 73.0% | 7,363 | 3.31× |
| 2025 | 79.4% | 8,456 | 3.60× |
Excluded as UTC year-boundary slivers, far below any real year: 2026 (42 known-race stops). New Year's Eve stops cross midnight into the next UTC year; they are in the data, just not plotted as a year of their own.
One ACS population baseline (see acs_vintage) is applied to every year of the trend. Population shares move slowly, but the early years lean on a denominator measured later; year-over-year changes in the index reflect changes in stops, not population.
Search rates in this district
Black people stopped in District 3 were pat-down searched at 2.79× the rate of non-Black stopped people (4.8% vs 1.7%).
- Black: pat-downs / stops
- 1,124 of 23,511 (4.8%)
- Non-Black (known race): pat-downs / stops
- 146 of 8,526 (1.7%)
- Any search (consent, probable cause, or warrant)
- Black 2.7% · non-Black 1.3%
Rates are per stopped person within each group, so a district's stop volume does not inflate them. Hit rates (what searches found) are citywide, on the searches page, because per-district search outcomes are too thin to split honestly.
Stop outcomes by race Preliminary
Share of street stops that ended with no ticket (a stop that went nowhere). Harbor (marine-unit) stops are excluded: 0. This rests on STOP_TYPE alone, not a full enforcement measure, so it is marked preliminary.
| Group | No-ticket share | Street stops |
|---|---|---|
| Black | 53.1% | 23,511 |
| Non-Black (known race) | 33.4% | 8,526 |
How this number is built (and where it's soft)
- Local baseline. The headline index compares each district's Black stop share to that district's own Black population share, not the citywide 43.3%. It answers "relative to who lives here." The citywide figure is shown alongside for comparison; the two deliberately have opposite shapes.
- Grain: stops of persons. MPD stop rows are per person, not per stop-event (multi-person stops repeat). Counts are person-records, paired against ACS person population.
- Exclusions (reported, not dropped quietly). Stops with race recorded as Unknown (15,728, 6.87% of all stops) are excluded from the known-race denominator. Stops with no mappable district (3,057, 1.33%) are excluded from any per-district figure. The figures on this page rest on 91.95% of all stops.
- Whole-tract population caveat. District populations are built by assigning each census tract's whole population to one district by its centroid, so population near a district boundary can be misattributed and the reconciliation check cannot detect it.
- Vintage mismatch. Population is ACS 2019-2023 (5-year); the stops are 2023-2025. The denominator lags the numerator.
- District-grained. MPD stop data has no coordinates and no ward, only a police district. All geography here is at the seven-district level.
Other districts (numeric order, not a ranking):
District 1 District 2 District 4 District 5 District 6 District 7 | Citywide overview