The headline number surprises people. Violent crime in the United States has fallen sharply from its early-1990s peak. The FBI's Uniform Crime Reporting program counted roughly 1.9 million violent crimes in 1992. By the most recent full-year estimates, that figure sits closer to 1.2 million — even as the country's population grew by more than 70 million people. Murder, the most reliably reported crime, dropped from a rate of 9.8 per 100,000 people in 1991 to roughly 5 to 6 per 100,000 in recent years.

But a national average tells you almost nothing about your neighborhood. Crime in America is intensely local. The difference between the safest large city and the most dangerous one is not a gap of 20 percent. It is often a gap of tenfold or more.

The Safest Big Cities

Among cities with populations above 100,000, a handful consistently rank at the bottom of the violent crime tables. Using FBI data and the rankings published annually by U.S. News & World Report, several names appear again and again.

Irvine, California is the standout. With a population around 300,000, Irvine routinely records violent crime rates below 60 per 100,000 residents — a fraction of the national rate. The city is affluent, well-policed, and demographically homogeneous, which criminologists note matters.

Naperville, Illinois and Gilbert, Arizona follow a similar pattern: large suburban cities with median household incomes well above the national figure and violent crime rates under 100 per 100,000.

Among the biggest cities, New York City remains the anomaly worth studying. With 8.3 million residents, it recorded fewer than 400 murders in a recent year — a murder rate lower than many cities one-tenth its size. San Diego and San Jose also post violent crime rates well below the national average despite their size.

The pattern is consistent: wealth, stable housing, dense policing, and low gun density correlate with safety. No single factor explains it all.

The Most Dangerous Cities

At the other end of the scale, a different set of names repeats year after year. St. Louis has posted the highest murder rate among large U.S. cities in multiple recent years, often exceeding 60 per 100,000 — roughly ten times the national figure. Memphis, Detroit, Baltimore, and New Orleans also rank near the top of violent crime tables.

What these cities share is not a single cause but a cluster of them. Concentrated poverty. Segregation that dates back generations. Declining population and a shrinking tax base. Easy access to illegal firearms. In St. Louis, the murder rate is heavily concentrated in a handful of north-side neighborhoods — a pattern repeated in Baltimore, Detroit, and Chicago.

That concentration matters. In Chicago, roughly half of all homicides occur in a small number of police districts on the South and West Sides. A resident of the Loop or Lincoln Park faces a violent crime risk far closer to Irvine's than to the citywide average.

Why the Averages Mislead

This is the single most important thing to understand about crime statistics: they describe places, not people. A citywide rate is an average across neighborhoods that may have almost nothing in common.

The Council on Criminal Justice and researchers at institutions like the University of Pennsylvania have documented that violent crime clusters at the block level. A study of gun violence in several major cities found that a small percentage of street segments accounted for the majority of shootings.

That means two families living three miles apart in the same city can face wildly different risks. It also means that "Is this city safe?" is the wrong question. The right question is: "Is this block, this neighborhood, this commute safe?"

What the Numbers Do and Don't Capture

There are real limits to the data. The FBI's numbers depend on what local departments report, and not every agency reports every year. The Bureau of Justice Statistics runs a separate survey, the National Crime Victimization Survey, which asks Americans directly whether they were victims of crime. It consistently finds that a large share of crime — especially assault and sexual assault — is never reported to police.

Definitions also vary. "Violent crime" in FBI terms covers four offenses: murder and nonnegligent manslaughter, rape, robbery, and aggravated assault. A city with aggressive reporting of aggravated assault can look more dangerous than a similar city that undercounts it.

And year-to-year swings can be noisy. A single bad year in a city of 600,000 can move the rate by double digits. Trends over five and ten years are more meaningful than any one year's ranking.

Practical Advice for Readers

If you are choosing where to live, work, or send your kids to school, here is how to use this information rather than be misled by it.

  • Zoom in past the city level. Use neighborhood-level tools like the FBI's crime data explorer, your local police department's public dashboard, and sites that map incidents by block. Compare specific addresses, not metro areas.
  • Look at five-year trends, not one year. A city improving steadily is a different bet than one spiking after years of decline.
  • Check what type of crime dominates. Property crime and violent crime move differently. A high burglary rate is a different problem than a high homicide rate.
  • Talk to people who live there. Data tells you what happened. Residents tell you how it feels and where the trouble actually is.
  • Weigh your own risk profile. Most violent crime involves people who know each other. Random stranger violence is rarer than the headlines suggest, and it is concentrated in specific places and times.