Most LAPD calls for service aren't calls from the public

Most of the entries in the Los Angeles Police Department's Calls for Service data aren't calls from the public. In a dataset containing 3,908,113 records from January 1, 2024, through August 8, 2026, the single largest call type is "CODE 6," which an officer uses to log that they are at a scene. It appears 1,800,368 times.

The second-largest type is "902 TRAFFIC STOP," with 273,526 records. Together, those two officer-initiated categories account for 2,073,894 calls, or 53.1% of everything in the dataset. A majority of the file therefore records activity officers logged themselves, rather than requests the department received from residents.

That distinction changes how I read two maps built from the records. One shows what residents call the police about, after officer-initiated activity is removed. The other shows where calls are most likely to begin with an officer rather than a member of the public.

The part of the city the data can place

Only 3,025,845 of the 3,908,113 calls carry one of LAPD's 1,135 reporting districts, so 77.4% can be mapped. The other 22.6% have no reporting district and are invisible here. Missing locations also affect one kind of record more than another, because Code 6 entries disproportionately lack a reporting district.

Among the calls that can be placed, 1,196,477 are officer-initiated, a share of 39.5%. That is far below the officer share suggested by Code 6 and traffic stops across the full dataset, which means the mapped records under-represent officer activity. Reporting districts add another limitation: they are police geography, not neighborhoods, even when their boundaries cut across places Angelenos would recognize by name.

Removing the officer-initiated records leaves 1,829,368 mapped calls from the public. To make those calls legible, I grouped 999 distinct radio codes into nine families using their code prefixes. This is my taxonomy, not an official LAPD classification, and roughly 10% of public calls wind up in a residual "public order and other" bucket.

What Los Angeles calls the police about

Assault and weapons and noise and disturbance together make up almost two fifths of what the public calls about.

A map of each district's most common category wouldn't reveal much, because the raw winner is nearly the same almost everywhere. Instead, the first map colors each reporting district by the category that stands out most relative to its citywide share. A district marked for burglar alarms isn't necessarily a place where alarms produce more calls than anything else. It's the place where alarms sit furthest above the rate the rest of the city produces.

Of the 1,135 reporting districts, 1,094 had at least 200 public calls and qualified for the map. Burglar alarms stand out in the most districts, and alarms, disputes and trespass between them account for more than half of the map.

Those relative winners form coherent patches rather than noise scattered across the city. Burglar alarms dominate a large contiguous area through the west San Fernando Valley and the Santa Monica Mountains. Domestic and neighbor disputes, by contrast, run in a band through South and Central Los Angeles.

The raw codes underneath the families give the map some texture. "906 CODE 30 RINGER," a burglar alarm ringing, appears 110,147 times, while "415 MAN," a man disturbing the peace, appears 95,135 times. The data also include 52,369 calls for "921 TRESPASS SUSP," 51,133 for "904 AMB," 38,914 for "507 RADIO," 38,251 for "900 UNKNOWN TROUBLE," 34,449 for "620 DOM VIOL," and 32,157 for "507 PARTY."

None of this makes the first map a crime map. Calls measure willingness to contact the police, not what happens in a place, and two areas with identical conditions can produce very different patterns depending on who feels able or inclined to dial. The analysis holds as a picture of what people call about, but it can't establish the underlying incidence of crime or disorder.

Where officer-initiated records are concentrated

The second map puts officer activity back into view by coloring each district according to the share of mapped calls that officers initiated. Across districts, the median is 31.6%. The range runs from 4.5% to 97.2%, and the highest shares cluster in South Los Angeles, downtown and the harbor area.

That spread describes who creates the record as much as it describes where LAPD work occurs. An officer creates a record by logging a traffic stop or going Code 6 at a scene; a resident creates one by asking the department to respond. Both are stored as Calls for Service, and the field does not distinguish them, which is why the share above has to be derived from the call types.

Code 6 entries often lack a reporting district, so the mapped records omit much of the officer-generated activity. That makes the second map a conservative account of it: Code 6 is the largest call type in the full file and the one most likely to arrive without a location, so much of what makes officers the dominant source citywide drops out when the records are placed on a map.