Beyond the Headlines: How to Read NYC Crime Stats Like a Pro

This story was originally published by THE CITY. Sign up to get the latest New York City news delivered to you each morning

By Reuven Blau, THE CITY: Jan 16 5:00 a.m. EST

Murders. Shootings. Robberies.

The number of those major crimes in the city all fell sharply last year — with some plunging to their lowest levels in modern recorded history.

How much attention should New Yorkers pay to figures like these? And what numbers do criminologists and police officials believe matter most?

Those questions are taking on added meaning with President Donald Trump deploying the National Guard in cities such as Washington D.C., Chicago and Los Angeles, where the White House says crime is spiraling to dangerous levels — even though violent crimes are generally trending down in large U.S. cities.

“We see the headlines and hear the pundits talk about crime being out of control,” Police Commissioner Jessica Tisch told reporters last Tuesday as she announced the overall drop in so-called major crimes. “These numbers tell a very different story.”

THE CITY spoke with multiple experts who study crime data and asked what metrics they look at and how they interpret them.

Here’s what they said:

What are the main crime statistics, according to the police? 

When talking about crime in New York City, the conversation almost always begins with the so-called seven index crimes, or major crimes: murder, rape, robbery, felony assault, burglary, grand larceny (theft of property worth $1,000 or more) and grand larceny auto.

The NYPD posts those figures online each week as well as some historical data to put the information into context.

The FBI also collects index crime information from other cities.

Of those major crimes, which are the most useful and reliable to watch?

Homicide is widely regarded as the most important and reliable data point, as well as the most scrutinized by the public and policymakers. Some experts also rely on firearm violence statistics.

“I just look at shootings and murders,” said Peter Moskos, a professor at John Jay College of Criminal Justice.

The former Baltimore beat cop said he doesn’t trust other statistics because they are frequently underreported and left to interpretation.

In New York City, the number of homicides dropped by 20% last year, from 382 in 2024 to 305 in 2025, according to the NYPD. By contrast, there were over 2,200 murders in 1990 during the peak of the crack epidemic.

“People are more concerned about homicide,” said Alex Piquero, who served as the director of the U.S. Bureau of Justice Statistics during the Biden administration.

“That’s always what is the first story in the news, or they watch it on TV, and that’s what people look at,” he added.

The number of homicides is dropping across the country, federal data shows. “We’re moving in a really good direction in most places,” Piquero, who teaches at the University of Miami, told THE CITY.

Homicide rates are considered the most accurately recorded crime stats because the end result is undeniable. “There’s a body and the medical examiner’s office is involved,” said Alex Vitale, coordinator of the Policing and Social Justice Project at Brooklyn College.

“It’s not really an issue of reporting bias,” he continued. “Basically, when we see significant changes in homicide numbers over time, we have a lot of confidence in that.”

Similarly, auto theft numbers are highly reliable because people must report it to the police to remove the registration from their name and claim insurance, added Vitale, author of “The End of Policing,” who is also on the Mamdani transition team.

Which major crime statistics are less reliable?

Other index crimes, such as burglary and felony assault, are traditionally underreported, according to the National Crime Victimization Survey, which asks a group of random people “about crimes they may have experienced in the past six months.”

Still, that does not render the year-to-year trends meaningless. “When patterns hold over time, they do tell us something,” Vitale said.

Rape is the least reliable index crime reported, according to Vitale and other criminologists.

“The willingness of rape survivors to report to the police seems to be driven in part by some external factors,” he said. “And so that’s the lowest reliable number of the major crimes that we typically look at.”

In New York City, reported rapes rose 16% citywide in 2025, a spike the NYPD attributed largely to a state law change in 2024 that expanded the legal definition of rape. Nearly half of reported rape cases were domestic violence-related, police said.

The NYPD conducts spot check audits of its index crime figures but doesn’t post those reviews online or in any other public forum. Over the years, the audits have discovered top police officials downplaying some statistics. Top brass is under intense pressure to reduce crime in their areas and they are called out during routine “CompStat” meetings inside the department’s main office at 1 Police Plaza.

How accurately do the major crime stats reflect what’s actually happening?

Jaeok Kim, associate director for research at the Vera Institute’s Greater Justice New York, said the major crime categories don’t always line up with how people experience safety in their daily lives.

Felony assault, for instance, is often used as a shorthand for whether a city feels safe, feeding the idea that those cases reflect random acts of violence by strangers, she said. But when you look more closely at the data, that picture starts to break down.

About half of felony assaults involve someone the victim already knows, Kim said, which makes it a far less straightforward measure of public safety than it’s often treated as.

The focus on crime stats also fails to address underlying causes that led up to the offense being committed.

Kim said that’s why she and other researchers look well beyond crime data when trying to understand safety.

In recent work, she said, researchers have paired crime statistics with measures of housing and food security, environmental conditions and economic stability. Those factors include things like education levels, poverty rates, noise and air pollution, and how many residents are struggling to afford basic needs.Research has shown that higher poverty neighborhoods frequently also have higher rates of violent crime.

Just as important, Kim said, is asking people directly what makes them feel unsafe in their own communities. When residents are part of defining the problem — alongside city agencies — the picture of safety becomes far broader than crime alone.

That approach, she added, has another advantage: it points to solutions. While violent crime can be difficult to prevent in a targeted way, issues like food insecurity, housing instability, pollution and job access are often more concrete and more actionable. Research has shown that improving those has been associated with crime reductions.

Vitale, who is also a sociologist at Brooklyn College, said “disorder” is among the most poorly measured aspects of public safety.

Encounters that often shape how safe people feel, such as seeing a homeless person who appears disoriented or in crisis in a subway station, do not show up in crime statistics, he said. Many quality-of-life issues are not crimes at all, and even when they are, they frequently go unreported.

Crime statistics seem to go up and down a lot throughout the year. What does that mean?

Almost all the experts agree that sudden spikes or drops in crime from week to week or month to month should not be given too much credence.

“I normally would want to look at at least a year’s worth of data before drawing any significant conclusions,” Vitale said. “And you know, ideally longer than that.”

He cautioned that sensational coverage of some murders, like a mentally ill person shoving someone on the subway tracks, should not be taken out of context when looking for trends. “Yes, homicides sometimes come in spikes, but that doesn’t necessarily tell us what the overall trend is,” he added.

Piquero, a criminologist, said he starts with the year-to-year trend when reviewing crime data and asks himself, “Is it going down, up, or steady?”

Not everyone agrees with that approach. One expert says it’s important to look out for any major trends happening during the year. “Nothing magic about Dec. 31 as a cut off,” said Moskos, author of “Back from the Brink: Inside the NYPD and New York City’s Extraordinary 1990s Crime Drop.”

Overall, the experts also noted that crime often goes up in the summer when the weather is warmer and people spend more time outside — and teens are home from school.

What figures don’t matter? 

Vitale and some other criminologists contend that the number of people locked up doesn’t make a difference to public safety.

The population of  Rikers has gone from a decades low of under 4,000 during the peak of the pandemic to around 7,000 right now.

Almost half of the people detained there suffer from some sort of mental health issue, according to the city’s Department of Correction.

“When we look at the population at Rikers, what we see is that a huge proportion of it is basically people in acute crisis,” Vitale said. “They’re unhoused, they’re unemployed, they have mental health and substance use issues, and yes, some of them are a source of harm and burdens and quality of life issues for the community.”

But sending them to Rikers, where it costs an estimated $500,000 per person a year, is “the most expensive thing we could possibly do,” he said.  “If we look long term, we do not see a simple correspondence between the population of Rikers and crime rates.”

He pointed out that the number of people in city jails dropped during the entire Bloomberg administration from 2002 to 2013. Yet murder and other crime plummeted to record lows.

Moskos, the former Baltimore police officer, has a different take. The number of people in jail does matter to public safety, he said.

“You hate to judge effectiveness by the number of people in jail, and I would hate for there to be a goal or quota, but of course if more people who would be hurting people on the street are in Rikers it means less people being hurt in the street,” he told THE CITY.

So which numbers should people really be paying attention to?

Criminologists say indicators such as overdose deaths, homelessness, and health insurance coverage rates often reveal more about public safety than crime statistics alone.

“That’s something that people often miss,” said Kim from the Vera Institute.

The number of people being admitted to hospital emergency rooms after suffering some type of assault is also a potential indicator of public safety. But those figures aren’t updated regularly so they are hard to keep a close eye on and flag for trends.