Data Methodology

Primary Data Source

All crime data on PlainCrime is sourced from the FBI's Uniform Crime Reporting (UCR) program, the nation's most comprehensive standardized crime data collection. Specifically:

• Table 8, Offenses Known to Law Enforcement by State by City
• Table 10, Offenses Known to Law Enforcement by State by Metropolitan and Nonmetropolitan Counties
• FBI Crime Data Explorer (CDE) API, State-level crime trends

Data Vintage

PlainCrime uses the most recently available FBI UCR annual release. UCR data is published approximately 1 year after the reporting period. The database reflects city-level crime data for 8,986 jurisdictions across the United States. UCR data represents the most comprehensive standardized crime statistics available at the city level, though participation rates vary by state and year. Some states have nearly complete agency participation, while others have significant gaps.

Processing Pipeline

1. FBI UCR Table 8 and Table 10 CSV files are downloaded from the FBI's public data portal.
2. City-level offense totals for violent crime (murder, rape, robbery, aggravated assault) and property crime (burglary, larceny-theft, motor vehicle theft, arson) are extracted.
3. Per-capita crime rates are computed using FBI-reported population figures: Rate = (Offenses ÷ Population) × 100,000.
4. Safety grades are calculated by comparing each jurisdiction's crime rate against national UCR averages for cities of similar size. The size-adjusted comparison is critical because crime rates naturally correlate with population density and urbanization, comparing a small rural town directly against a major metropolitan area would produce misleading results. Our grading methodology groups jurisdictions into population tiers and evaluates each city against its peers.
5. State-level trend data is pulled from the FBI Crime Data Explorer API and joined to state records.
6. All data is loaded into a structured SQLite database serving city, county, and state pages.

Crime Categories

The FBI Uniform Crime Reporting program categorizes offenses into two broad groups. Violent crimes involve the use or threat of force against a person. Property crimes involve the taking or destruction of property without force or threat against a person.

How the Source Agency Collects Data

The FBI Uniform Crime Reporting program collects crime data from approximately 18,000 city, county, university, state, tribal, and federal law enforcement agencies across the United States. Participation is voluntary, agencies submit monthly returns reporting the number of offenses known to law enforcement in their jurisdiction. The FBI standardizes crime definitions across agencies using the UCR Summary Reporting System classification rules so that crimes reported in different states follow the same counting methodology.

State-level UCR programs serve as intermediaries, collecting data from local agencies within each state and forwarding it to the FBI. The FBI reviews submissions for accuracy, contacts agencies about anomalies, and publishes the compiled data in the annual Crime in the United States report.

Data Accuracy Commitment

PlainCrime presents FBI Uniform Crime Reporting data without modification. Crime counts, per-capita rates, and safety grades are computed directly from the published UCR tables, which the FBI also distributes through the FBI Crime Data Explorer. For broader victimization context, we reference the Bureau of Justice Statistics. We do not estimate crime for non-reporting agencies or adjust figures for underreporting. If you find any data that appears incorrect, please contact us at hello@plaincrime.com and we will verify against the FBI source data.

Limitations

• UCR participation is voluntary, not all law enforcement agencies submit data every year. Missing data does not mean zero crime.
• Reporting practices, crime definitions, and local factors vary across jurisdictions, making direct city-to-city comparisons imperfect.
• Population figures used for rate calculations come from FBI-reported populations, which may differ from Census estimates.
• Crime statistics should not be used as the sole basis for decisions about personal safety or relocation.
• PlainCrime is not affiliated with the FBI or any government agency.

Per-Source Breakdown

PlainCrime joins data from three distinct FBI publications. Each source has a different update cadence, geographic granularity, and known data-quality profile. We document each one explicitly so analysts can match figures back to the FBI source of record.

Refresh Cadence

The FBI UCR program publishes its annual Crime in the United States release in September or early October each year, covering the prior calendar year. PlainCrime ingests the new release within two weeks of publication and rebuilds the database to reflect the latest figures. Between annual releases, the FBI Crime Data Explorer API provides continuous updates as agencies submit late or revised filings; PlainCrime pulls these updates monthly to keep state-level trend tables current.

Quarterly NIBRS supplements add incident-based detail throughout the year. PlainCrime does not currently surface NIBRS incident-level fields on every page, but the underlying tables include the supplemental columns for analysts who query the database directly.

Each detail page on PlainCrime carries a vintage timestamp drawn from the most recent ETL run. When you see "FBI UCR 2024" on a page, the underlying numbers were ingested from the FBI release published on September 23, 2025.

NIBRS Transition Notes

The FBI completed its transition from the legacy Summary Reporting System (SRS) to the National Incident-Based Reporting System (NIBRS) in 2021. NIBRS captures richer detail per incident, multiple offenses, victim and offender characteristics, weapon usage, location type, but the transition produced a two-year coverage gap as agencies migrated their reporting infrastructure. Some states (notably Florida, California, Pennsylvania, and New York) saw under-70% agency participation during the 2021–2022 transition window. The 2023 and 2024 releases improved coverage through statistical estimation, but inconsistencies remain when comparing pre-2021 SRS data to post-2021 NIBRS data for the same jurisdiction.

Per-100K rates calculated from NIBRS-era and SRS-era data are not strictly comparable. Trend lines that span the 2021 cutover should be read as directional rather than precise. PlainCrime renders the discontinuity transparently on per-state trend pages by drawing a vertical reference line at 2021 wherever the underlying data crosses the boundary.

Contact for Corrections

If you find any data on PlainCrime that appears incorrect, please email hello@plaincrime.com with the page URL and the value you believe is wrong. We will verify against the FBI source data and update within one business day if a correction is required.

For agencies and law enforcement personnel: if your jurisdiction's data is missing, partial, or appears inconsistent with your own records, the issue typically lies upstream at the FBI submission layer rather than in PlainCrime's ingestion pipeline. We will help you trace the source row and connect you with the FBI UCR program manager if the upstream data needs revision.

Editorial review and methodology refinement happens on a rolling basis. Methodology updates are logged at the top of this page with a <time datetime> stamp so analysts can verify which version of the methodology applies to the data they are viewing.