San Francisco Crime Map: SFPD Data by Neighborhood

A San Francisco block can change fast after dark. One corner stays quiet. A few streets over, the police record gets louder.

This map helps the reader sort that difference. It shows where reported incidents stack up by neighborhood, and the early verdict is plain: central neighborhoods log the most police reports, while quieter residential areas log far fewer. But a dark cluster is a signal, not a sentence.

The file runs from 01/01/2018 to 08/17/2026, with the public record last checked on 08/18/2026. The main pattern holds steady across the page: Tenderloin, SoMa, Mission, and Downtown/Union Square sit near the top by total reports, while places like the Presidio, Sea Cliff, and much of the Sunset sit near the low end.

That headline needs one brake. More reports do not mean the same thing as personal danger. Busy districts pull in commuters, shoppers, nightlife traffic, parked cars, and transit riders. Those streets leave more police paperwork than quiet blocks lined mostly with homes.

The strongest read comes from type, not just total. Property crime drives much of the city’s record, with theft and vehicle theft doing a lot of the counting in central business and retail zones. Violent crime sits in tighter pockets, showing up more heavily near some nightlife corridors, dense transit areas, and a small set of repeat hot spots.

The map also has blind spots. It shows reported incidents, not every event. Some reports arrive a day later. Some incidents never get reported. Some records cannot be pinned to a neighborhood and fall into a county-style catchall instead of the map.

That leaves a clear use case. Use the neighborhood map to compare places side by side, then check the exact block and crime type before making a decision. A high total in the Financial District does not carry the same street meaning as a high total in a mostly residential area. The count tells where the noise is. The block tells what kind.

A few points matter most:

• Top total-report areas: Tenderloin, SoMa, Mission, Downtown/Union Square
• Lower-report areas: Presidio, Sea Cliff, Sunset
• Main city story: property crime outweighs violent crime in much of the map
• Main caution: totals reflect activity, traffic, and reporting volume, not just resident exposure
• Best way to read it: compare neighborhood totals, then narrow to the street, route, parking pattern, and offense type

The city file works best as an evidence board. It is for comparison, not prophecy.

Inside San Franciscos MOST Crime Filled Neighborhood (The Tenderloin)

sbb-itb-6d427fc

1. Aden's San Francisco City Report

This report is the base file behind the map. It uses approved SFPD incident records from Jan. 1, 2018, through Aug. 17, 2026, last checked Aug. 18, 2026 ^[1][2][3].

There is a short delay in the public file. New reports usually appear about one day later because supervisors must approve them first ^[1][2].

To keep the count clean, the report includes only Initial, Vehicle Initial, and Coplogic Initial records ^[2]. That cuts out repeat entries for the same event.

The map sorts incidents into 41 analysis neighborhoods using coordinate points from DataSF ^[2]. If a record has no coordinates, it falls into an unmapped "County" bucket. Those records still show up in city totals, but they do not land on the neighborhood map ^[2].

Incidents are also split into violent and non-violent groups using NIBRS ^[2]. In one recent cut of this file, the report counted 62,011 mapped incidents in this time window ^[7].

That number reflects reported geocoded events. It does not stand in for every incident that happened across the city.

Use this city report as the baseline. The next sections break the file into neighborhood, ZIP, map, and side-by-side comparison views.

2. Aden's ZIP San Francisco Crime Map and Report Page

The city baseline sets the frame. Then the map tightens the shot, dropping SFPD incidents into neighborhood-level markers. Open one, and the file gets specific: case number, CAD number, and crime category appear on the card ^[2][3].

The map sorts those incidents across the 41 DataSF analysis neighborhoods and tracks the numbers that matter at a glance: total incidents, violent incidents, trend lines, and top crime categories ^[2][3]. The signal is useful, but it needs restraint. Counts are not per-person risk. A busy retail strip or nightlife corridor will often carry more reports simply because more people move through it. That makes the map good for side-by-side comparison, not a stand-alone judgment of danger.

Another layer sits over the street picture. The page adds ZIP codes and police districts on top of the neighborhood map, including overlays such as 94102, 94110, and 94124, along with Bayview, Tenderloin, and Mission district filters, so the reader can compare local neighborhood lines with the administrative boundaries SFPD uses ^[8].

3. Aden's ZIP San Francisco ZIP report

The ZIP report adds a postal-area lens to the neighborhood map. In San Francisco, that means the same SFPD incident file appears one level up: by ZIP, side by side, block noise compressed into a cleaner city pattern ^[1][2][3].

The main ZIP metrics are incident volume, incidents per 1,000 residents, leading crime categories, and the Safety Pulse Score (0–100). Read together, they show where police recorded more trouble, how that total sits against local population, what type of trouble shows up most, and how the score reads the file at a glance. Use this view to compare ZIPs directly.

There is one line to keep in sight. The ZIP report uses geocoded incident data, not strict USPS boundaries. Incidents without coordinates do not appear in map views, and SFPD says the records can change over time ^[1][2][3]. That makes the ZIP report the cleanest baseline for comparison, but not a final verdict on every street.

4. Aden's ZIP San Francisco Comparison Report

The city file narrows here. After the citywide map and the ZIP view, the comparison report shows how each part of San Francisco lines up against the rest.

Aden's ZIP comparison report ranks 41 analysis neighborhoods by total incidents, per-capita rate, and crime mix ^[2][4]. The point is comparison. Not a stand-alone verdict on danger.

Per-capita rate is the key reality check. Dense business districts like the Financial District can log heavy police volume while telling a different story for the people who actually live there ^[4]. Big raw numbers can come from foot traffic, offices, retail, and nightlife. That does not always mean the same resident-level exposure.

The report uses the same geocoded SFPD incident file, leaves out supplemental updates, and places incidents that could not be mapped into the "County" bucket so neighborhood-to-neighborhood comparisons stay clean ^[1][2].

Population figures come from the 2023 American Community Survey. That matters because small gaps between neighborhoods may sit inside the survey's margin of error. For a cleaner read, use the local-data vs. citywide-data toggle. It shows whether a neighborhood lands above or below the city average, and which crime categories are pulling it there ^[4].

That comparison frame sits underneath the neighborhood rankings that follow.

5. Highest-Incident Neighborhoods

The comparison view puts Tenderloin, SoMa, the Mission, and Downtown/Union Square at the top for incident totals across San Francisco’s 41 analysis neighborhoods ^[2][7].

That ranking needs a second look before it turns into a verdict.

Downtown/Civic Center has 46,662 residents. The Financial District has 8,329 ^[7]. A high total does not automatically mean a resident faces the same level of risk in both places. In the city’s busiest zones, police counts often climb because the streets carry more than residents alone: offices, bars, late-night foot traffic, shoppers, visitors, and steady movement near transit lines ^[2][7].

A large share of the total in these central districts comes from vehicle theft ^[7]. Violent incidents also gather in a small set of central neighborhoods ^[7].

The file has one guardrail built in: supplemental updates are left out, which helps avoid double-counting. These numbers reflect initial reports only ^[2].

SoMa and the Tenderloin do post high counts. That signal is real, especially on commercial corridors and blocks tied closely to transit. But neighborhood totals can cast too wide a shadow. The sharper read comes from the exact block and the exact crime type. The next sections break those totals apart into property and violent crime.

6. Middle-Incident Neighborhoods

Below the city's hottest zones, the middle tier settles into a steadier pattern. The spikes soften here. Trouble spreads across more of the neighborhood instead of piling onto just a few blocks.

Nob Hill leads this band. Financial District stays in the same range despite its small residential base. Downtown/Civic Center holds the lower edge of the tier ^[7].

In this part of the map, vehicle theft does much of the work. The pattern tracks places with more stores, office activity, parked cars, and passing foot traffic more than it tracks how many residents sleep there at night. The lower tier tilts the other way.

7. Lower-Incident Neighborhoods

The quiet edge of the city tells a different story.

The Presidio, Sea Cliff, and the Sunset sit at the low end of the SFPD file, which covers January 1, 2018, through August 17, 2026 ^[1][4]. These are mostly residential places. They carry far less of the foot traffic, nightlife, retail churn, and visitor flow that push police reports higher elsewhere ^[2][3].

The split becomes clearer when the map isolates crime type. In these neighborhoods, property crime leads the file. Violent incidents appear much less often ^[2][4].

That does not mean no risk. The records show reported incidents, and some entries can be hidden under privacy rules ^[9].

So the lower totals point to the shape of the place: more homes, less commercial activity, fewer reasons for heavy report volume. They do not mean the file catches every police interaction. The next view pulls property crime away from violent crime, and the contrast tightens.

8. Property Crime Patterns by Neighborhood

Property crime gathers in a few parts of the city. It does not spread evenly block to block. Once violent crime is pulled out, the map points to the same places again and again: commercial corridors, retail strips, transit-heavy zones, and areas built for constant foot traffic.

The Financial District ranks near the top for property crime, with other high counts in dense retail, tourism, and transit areas. On the map, those sections tend to hold the city's brightest signals. That matters because the violent-crime pattern that follows is tighter, narrower, and more tied to exact blocks than broad business zones.

Raw count does not equal resident risk. A business district can collect more reports for a simple reason: more people pass through, more cars stop, more bags get left behind, and more chances for theft appear under brighter lights and heavier traffic.

Method note. Neighborhood tools now exclude duplicate incident reports and public-filed reports, so older snapshots may not match newer exports ^[4].

9. Violent Crime Patterns by Neighborhood

Property crime reaches across the city. Violent crime gathers in tighter pockets.

The pattern is sharper near commercial corridors, nightlife zones, tourism blocks, and transit lines. A few neighborhoods carry much of the recorded activity, while many others stay quieter on the map.

This section tracks where violent incidents concentrate, not how the map itself is built.

Per-capita rates can push small-population places like McLaren Park, Lincoln Park, and the Financial District higher than expected, even when the total number of incidents is lower. At the same time, the Tenderloin and Mission stay high when the count is measured in plain totals. That split matters. For violent crime, offense type often tells the cleaner story than raw totals alone.

The map shows reported incidents, not every event. The next section reads the city by crime type.

Reading the Map by Crime Type

Crime type explains the map faster than the neighborhood ranking does.

Property crime gathers where the city stays busy: shopping streets, tourist blocks, transit hubs, and places where cars sit and cycle through all day. Violent crime tends to tighten around nightlife zones, crowded corridors, and the same recurring hot spots. The table shows where each category appears most often, what those counts usually mean, and where the picture can bend out of shape.

Category	Neighborhoods Where It Appears Most	What the Count Indicates	Key Cautions
Larceny / Theft	Financial District, Union Square, Mission [3][4]	High volume of opportunistic theft in commercial and tourist areas [3].	High counts often reflect visitor volume rather than resident risk [3].
Burglary	Areas with high commercial or residential density [3]	Targeted property entry; may be grouped with related vandalism charges [3].	A single case number may include multiple offenses, such as burglary plus vandalism [3].
Motor Vehicle Theft	Transit hubs, large parks, Mission [3][4]	High vehicle turnover and availability of targets.	Some reports are filed online and may lack exact GPS coordinates [6].
Robbery	Commercial corridors and recurring hot spots [3]	Direct confrontation; classified as violent crime [3].	Use the violence filter to separate these from property crimes [3].
Assault	Tenderloin, Mission, South of Market [5]	Street activity and nightlife intersections; incidents often peak late at night [3].	Mission counts are spread over a larger area than the more concentrated Tenderloin [5].
Arson / Vandalism	Citywide; often logged as a secondary charge [3]	Intentional property damage; lower frequency overall [3].	Small counts can look like a pattern in quiet areas [3].

Color matters. Shape matters too.

Blue markers usually point to nonviolent property crime. Orange or red markers usually mark violent incidents, and red can also signal an arrest. Read those counts as exposure to activity, not only as a measure of resident risk ^[3].

The map still shows reported incidents, not the full amount of neighborhood exposure.

Strengths and Limits of This Neighborhood View

The file behind this map runs from January 1, 2018, through August 17, 2026. DataSF, the city's open records portal, shows the source dataset was last updated on August 17, 2026 ^[1]. The timeline is long. The blind spots still matter.

The main strength is simple: this starts with official city records. Every point traces back to DataSF ^[1]. The map and neighborhood views add street-level context that the raw feed does not show on its own ^[3]. A 30-day pattern makes more sense when set against the city baseline, and straight totals can look louder than they are in places built around offices, transit, nightlife, or visitor traffic ^[3].

Some limits sit in the background of every map check:

• Reporting lag: Most incidents land in the feed one to two days after they happen ^[2].
• Underreporting: Domestic incidents and minor property thefts often never become formal reports ^[2].
• Reclassification: SFPD can change how an event is labeled after the first entry appears ^[2].
• Boundary mismatch: SFPD's 41 analysis neighborhoods do not always line up with ZIP code edges, so a count that seems tied to one neighborhood can spill across two areas ^[3][2].

That same warning runs through the neighborhood rankings, property crime pages, and violent crime pages: raw counts are not per-capita rates. High totals often track places with dense retail, bars, transit stops, and tourist flow, not a higher day-to-day risk for people who live nearby ^[3][2].

The table below separates what each view does well from where it can send the wrong signal.

Source or View	Pros	Cons	Best Use
SFPD / DataSF	Official sourcing; includes self-service online reports; refreshes after approval [1]	Reporting lag of 1–2 days; accuracy isn't guaranteed; subject to reclassification [1][2]	Verifying specific case numbers or official citywide totals
Aden's ZIP Map	Interactive markers; adjustable 30–90 day ranges; color-coded by severity [3]	Requires exact coordinates; incidents in the "County" bucket may not appear as map markers [3]	Visualizing crime clusters near a specific address
Neighborhood / ZIP Reports	Groups offenses by case number; shows local patterns and peak activity times [3]	Raw counts can mislead in high-traffic commercial zones [3]	Understanding local patterns and activity concentration
Aden's Comparison Pages	Ranks 41 neighborhoods; provides Safety Pulse context vs. city average [3]	Focuses on raw volume, not per-capita rates [3][2]	Comparing safety levels across different residential areas

A dark cluster is a signal to inspect the block more closely, not proof of resident risk. The neighborhood rankings make more sense when read with these limits in view.

Conclusion

The ranking and the crime mix land on the same rule: a total by itself does not tell the whole story. A neighborhood stacked with larceny reports says one thing. A place with fewer incidents but a heavier share of assaults or robberies says another.

Across the Jan. 1, 2018–Aug. 17, 2026 file, high totals do not automatically mean higher risk for residents. Areas like the Financial District carry heavy volume largely because of foot traffic and property crime, not because every block exposes residents to the same level of danger. These are reported incidents, not final counts, and some records do not map cleanly ^[1][4].

Start with the Aden's ZIP San Francisco city report. Then move to the Aden's ZIP San Francisco comparison report, the Aden's ZIP San Francisco ZIP report, and the Aden's ZIP San Francisco crime map and report page to trace what is driving the count.

A dark cluster is a signal. Not a verdict on the neighborhood.

FAQs

How should I compare two San Francisco neighborhoods fairly?

Compare neighborhoods with per-capita crime rates, not just raw incident totals. A busy strip lined with shops, bars, offices, or tourist traffic can stack up reports fast. That does not always mean the surrounding neighborhood is more dangerous than a quieter area with fewer people passing through.

Use official neighborhood snapshots and CompStat trends as the city’s paper trail. Keep violent and property crime separate. They do not move the same way across the map, and the difference matters when the decision is where to live, park, walk, or invest.

Why do downtown areas show more crime reports?

Downtown usually looks darker on the map. That does not mean every block is more dangerous than it seems.

Police incident data is a street log: calls for service and filed reports, not a full count of every crime that happened.

Busy core districts pull in more reports for simple reasons. More people pass through. More stores open and close. More cars sit in garages and lots. More property cases get reported, especially larceny and burglary.

So the map records two things at once: where trouble happened, and where reporting happens more often. In the city center, those lines often overlap.

What crimes matter most on this map?

Focus stays on violent and property crimes. Those are the categories SFPD CompStat follows most closely when incidents are mapped and neighborhoods are compared.

The records on the map run from 2018 to the present and are pinned to the place where each incident happened. That matters. The map is built to show which neighborhoods carry heavier or lighter counts in those two crime groups, not to make the whole city feel like one single number.

Think block by block, not headline by headline. One part of San Francisco can stay quiet while another logs more police reports a few streets away.