Through the week to 25 October, the NYPD’s preliminary count was 224 murders this year, a category that includes non-negligent manslaughter.
That is 18% fewer than the 272 at the same point of 2025, a fall that holds in both statistical checks we ran.
On each police force’s own published count, New York’s rate in 2025 was nevertheless the highest of five big cities, at 3.6 murders per 100,000 residents.
That was 3.4 times London’s, after a fall of 88% since 1990.
However, the gap with London is wider than before the pandemic. Taking 2017 to 2019 together, New York’s rate was 2.1 times London’s; in 2025 the multiple was 63% higher than then, a widening that holds in both checks.
It reached 4.2 times in 2022; the narrowing since could still be chance.
New York’s rate is the highest of the five, and above the Met’s figure for it
In January London’s Metropolitan Police, the Met, announced 97 homicides in 2025, a rate of 1.1 for every hundred thousand people.
Its release set that beside other cities’: “below many comparable global cities, including New York (2.8), Berlin (3.2), Milan (1.6) and Toronto (1.6).”
The Met’s figure for New York, 2.8, is lower than New York’s rate in any year since 1990 on the city’s own counts. The release gives no count, year or population for it, so the two can’t be reconciled.
Its Toronto figure, 1.6, matches neither of Toronto’s own rates for 2025.
The city police’s count gives 1.4 and Statistics Canada’s, for the wider metropolitan area, 0.9; we hold no Berlin or Milan counts to check.
On their own counts, London’s rate in 2025 was 1.1, and that of Paris and its inner ring, the three county-sized départements around the city, 1.3.
Toronto’s was 1.4 and Tokyo’s 0.3, counting the victims who died. New York’s was the highest in every year from 2016, when all five counts begin, except 2018.
New York's rate was the highest of the five in 2016 and in 2025
Administrative Counts from an agency’s own records: the calls, inspections, trips or transactions it logs. They carry no sampling margin, but they count events rather than people, and they can be cut only by what the agency records about each one.Homicides per 100,000 residents, each city's own count
The interactive chart needs JavaScript; its numbers are in the table below.
How to read this, and what it cannot show
Each row is a city and each dot a year; the figure at the right is 2025. Hover or tap a dot for its rate.
The numbers behind this chart
About these numbers. New York: murder and non-negligent manslaughter. London: homicide victims (Metropolitan Police area). Paris and inner ring: homicide victims, including assault resulting in death, by place (départements 75, 92, 93, 94). Toronto: the city's deceased victims by date. Tokyo: victims who died (murder, infanticide, robbery-murder); its recognised murder cases, which include attempts, run about three times higher.
| Group | 2016 | 2025 |
|---|---|---|
| New York | 3.8 | 3.6 |
| London | 1.3 | 1.1 |
| Paris and inner ring | 1.2 | 1.3 |
| Toronto | 2.7 | 1.4 |
| Tokyo | 0.2 | 0.3 |
Paris itself, without its inner ring, had a rate of 1.7, which would put it second, still under half New York’s.
Toronto’s 2025 was unusually low on its police’s preliminary count, 45 killings against 85 in 2024.
In 2024 New York’s rate was 1.7 times Toronto’s.
No two forces count a killing quite the same way; the table sets the definitions side by side.
London’s count, the Met’s, takes in manslaughter by gross negligence, but leaves out the City of London’s square mile and killings in the British Transport Police’s area, while its denominator is all of Greater London.
Tokyo’s count leaves out deaths recorded as assault, 5 in 2025.
On that broader count Tokyo’s rate is 0.29, and New York’s 12.2 times as high.
Since 2003 London’s rate has fallen faster than New York’s
In 1990 New York counted 2,245 murders on the State’s count, a rate of 30.6.
CompStat’s own table prints 2,262 for that year; either way the fall since is 88%.
Toronto’s metropolitan area, the one other series here that reaches back that far, went from 1.8 to 0.9.
Its rate halved in 2025 alone; in 2024 it was 1.9.
New York's murder rate fell by nearly nine-tenths from 1990 and is still the highest of the three
Administrative Counts from an agency’s own records: the calls, inspections, trips or transactions it logs. They carry no sampling margin, but they count events rather than people, and they can be cut only by what the agency records about each one.Homicides per 100,000 residents, each city's own count, as far back as its records go
The interactive chart needs JavaScript; its numbers are in the table below.
How to read this, and what it cannot show
One line per city, from 1990 or from its first year after it. Hover or tap a point for the year’s rate.
The numbers behind this chart
About these numbers. New York: murder and non-negligent manslaughter (the State's count to 2000, the NYPD's from 2001). Toronto: the metropolitan area's victims by the year reported (Statistics Canada's rate). London: the Metropolitan Police's homicide victims by the month recorded; no consistent series before 2003 is published. Paris's and Tokyo's comparable counts start in 2016.
| Group | 1990 | 1991 | 1992 | 1993 | 1994 | 1995 | 1996 | 1997 | 1998 | 1999 | 2000 | 2001 | 2002 | 2003 | 2004 | 2005 | 2006 | 2007 | 2008 | 2009 | 2010 | 2011 | 2012 | 2013 | 2014 | 2015 | 2016 | 2017 | 2018 | 2019 | 2020 | 2021 | 2022 | 2023 | 2024 | 2025 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| New York City | 30.6 | 29.2 | 26.9 | 25.9 | 20.6 | 15.4 | 12.8 | 9.9 | 8.1 | 8.4 | 8.3 | 8.1 | 7.3 | 7.4 | 7.1 | 6.7 | 7.5 | 6.2 | 6.5 | 5.8 | 6.5 | 6.2 | 5.0 | 3.9 | 3.9 | 4.0 | 3.8 | 3.3 | 3.3 | 3.6 | 5.3 | 5.8 | 5.2 | 4.6 | 4.4 | 3.6 |
| Toronto metropolitan area | 1.8 | 2.5 | 2.1 | 1.7 | 1.9 | 1.7 | 1.8 | 1.8 | 1.6 | 1.3 | 1.7 | 1.6 | 1.8 | 1.9 | 1.8 | 2.0 | 1.8 | 2.1 | 1.9 | 1.6 | 1.4 | 1.5 | 1.4 | 1.4 | 1.4 | 1.3 | 1.6 | 1.5 | 2.3 | 2.0 | 1.6 | 1.9 | 2.0 | 1.8 | 1.9 | 0.9 |
| London (Metropolitan Police area, from 2003) | – | – | – | – | – | – | – | – | – | – | – | – | – | 2.9 | 2.5 | 2.4 | 2.3 | 2.1 | 2.0 | 1.6 | 1.6 | 1.4 | 1.3 | 1.3 | 1.1 | 1.4 | 1.3 | 1.6 | 1.6 | 1.7 | 1.5 | 1.5 | 1.2 | 1.2 | 1.2 | 1.1 |
London’s consistent count, the Met’s, starts in 2003, at 2.9, when New York’s rate was 7.4.
Since then London’s rate has fallen by 64%, New York’s by 51%.
So New York’s multiple of London’s has moved, from 2.5 in 2003 to 4.3 in 2011.
It was smallest from 2017 to 2019, between 2.0 and 2.1.
Then New York’s rate rose to 5.8 in 2021, while London’s fell, unevenly, from 1.7 in 2019 to its lowest in 2025.
New York's rate was about twice London's in 2017 to 2019, and is more than three times it now
Administrative Counts from an agency’s own records: the calls, inspections, trips or transactions it logs. They carry no sampling margin, but they count events rather than people, and they can be cut only by what the agency records about each one.New York's homicides per resident as a multiple of London's
The interactive chart needs JavaScript; its numbers are in the table below.
How to read this, and what it cannot show
One line; higher means a wider gap. Hover or tap a point for the year’s multiple.
The numbers behind this chart
About these numbers. Each city's own count on its own population: New York's murders and London's homicide victims (which also count manslaughter by gross negligence). London's series starts in 2003.
| Group | 2003 | 2004 | 2005 | 2006 | 2007 | 2008 | 2009 | 2010 | 2011 | 2012 | 2013 | 2014 | 2015 | 2016 | 2017 | 2018 | 2019 | 2020 | 2021 | 2022 | 2023 | 2024 | 2025 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| New York's rate divided by London's | 2.5 | 2.8 | 2.8 | 3.3 | 2.9 | 3.3 | 3.5 | 4.2 | 4.3 | 3.9 | 3.0 | 3.5 | 2.9 | 2.9 | 2.0 | 2.1 | 2.1 | 3.5 | 3.8 | 4.2 | 3.9 | 3.7 | 3.4 |
The Met credits its own work: “London’s record‑low homicide rate is the result of relentless work,” its Commissioner said in January. That is the force’s reading of its own city, which these counts can’t test.
This year New York’s murders are down by almost a fifth, and London’s and Toronto’s changes could still be chance
In the department’s own words, the first nine months’ 203 beat “the previous record of 215 set in 2017”: a record count for those months, not a record rate.
London’s recorded homicide offences, a separate file from its victims count, came to 73 in January to August, against 62 in the same months of 2025.
Toronto’s police counted 23 killings against 28 over the same months, on a preliminary count. Both gaps could still be chance.
Japan’s national police agency counted 96 murder cases in Tokyo in January to September, against 81, but that count includes attempts, so it isn’t set beside the others. The departmental series used here for Paris has no 2026 figure.
London passed New York for two single months, never for a whole year
“London’s murder rate has passed New York’s.” For two single months, never for a whole year. In two of the 144 months from 2014 to 2025, London’s homicides per resident exceeded New York’s murders per resident.
February 2018 was one, 18 against 15.
December 2019 was the other, 18 against 17.
Those set London’s month recorded against New York’s month of the killing. For 2018 as a whole, London’s rate was 1.6 and New York’s 3.3.
“Toronto is safer than New York.” On homicide in 2025, Toronto’s recorded rate was far lower. In 2018 the city of Toronto’s rate, 3.37, was almost level with New York’s 3.34, a shade higher.
“Tokyo has almost no murders.” The lowest rate here, though “almost none” has no threshold to test. Of 107 murder cases its police recognised in 2025, attempts included, 33 victims died.
With infanticide and robbery-murder the deaths come to 37, the count we compare.
“New York has never been safer from murder.” Not established for rates: the lowest whole-year rate in this series is still 2017’s 3.3, and 2026 isn’t complete.
Complete 2026 counts will give this year’s multiple
The 2025 counts give the order of these five published measures. Whether New York moved closer to London this year needs both forces’ complete 2026 counts and the year’s populations.
Those will give the observed multiple. Its 90% range may still leave the direction open, as it does for the change since 2022, and a record count from either force would not settle it.
Questions this raises
Are the five counts comparable?
Close, not identical, and the differences cut both ways: London’s count takes in kinds of manslaughter New York’s leaves out, while its area leaves out the City’s square mile and the railways. The table gives each definition.
They also count at different moments: the year reported in New York and in Statistics Canada’s figures, the month recorded in London, the date of the killing in Toronto’s police file. Every multiple is computed from unrounded rates, so it can differ slightly from the rounded rates it divides.
Why Paris and its inner ring, and the City of Toronto?
Paris with its three inner départements, Hauts-de-Seine, Seine-Saint-Denis and Val-de-Marne, is closer to New York’s population than Paris alone. Paris alone had 35 homicides in 2025.
For Toronto, the city police’s file gives the current year and runs from 2004, so it carries the comparison; Statistics Canada’s count for the much larger metropolitan area carries the long view, because it reaches back to 1981. The two never share a line.
Why does London’s long view start in 2003?
The Met’s homicide data, the consistent series used here, starts in January 2003. Earlier figures we found are on other bases, so we leave them out rather than join them to it.
How sure are the changes?
A count of killings swings by chance, so each change is tested twice: as a ratio of two counts with a 90% Poisson range, and again with that range widened by the extra swing in the city’s own monthly counts (New York’s from 2022, London’s and Toronto’s from 2015).
Widened, New York’s change this year runs from -30% to -3%, the only one of the three that clears zero.
London’s runs from -11% to 56%, Toronto’s from -48% to 30%.
The gap with London is tested as a ratio of the two cities’ rate ratios, widened by the larger swing in the two cities’ monthly counts from 2014 to 2025. Against 2017 to 2019 the 2025 multiple is 63% higher (29% to 107%).
Against 2022 its change runs from -40% to 7%. The widening is a sensitivity check: a swing that persists for months can be larger than it allows, and later revisions are not in it.
Under the hood
Every count is the agency’s own, read from the file it publishes: the NYPD’s yearly table and its CompStat report for 2001 on, the State’s index crimes before; the Met’s homicide victims by month recorded, and for this year its recorded homicide offences, a separate file whose 2025 total came within one of the victims count; France’s interior ministry’s victims by département; the Toronto police’s victims by date, its open-data file to 2023 and its feature service after; the Tokyo police yearbook’s victims who died under murder, infanticide and robbery-murder. Japan’s national police agency’s 2025 murder group, 110, is the yearbook’s 107 cases plus 3 of infanticide. The month-by-month comparison with London uses the NYPD’s complaint records, dated by the month of the killing. The NYPD’s yearly table doesn’t say whether killings in the city’s jails, whose complaints CompStat has counted apart since 2013, are in its murder count.
Every rate divides by the city’s residents in the same year, from its own statistical office: the Census Bureau’s 1 July estimates for New York, which join several vintages; ONS’s mid-year estimates for London; INSEE’s 1 January estimates averaged to mid-year for Paris; Statistics Canada’s 1 July estimates for Toronto; and Tokyo’s 1 October estimates. New York’s count alone fell 86%, from 2,245 to 309.
Tokyo revised its population for recent years to its 2025 census on the day these files were read, after they were fetched. If 2025’s population is lowered by that revision’s share of 1 January 2026, Tokyo’s rate stays at 0.3 and New York’s multiple of it would be 13.8. The figures for London, New York and Toronto’s metropolitan area reproduce the published ones in the box below; the data and the code are linked at the end.
Continue exploring New York
New York's shootings fell to the fewest on file; its felony assaults are back at late-1990s levels
Shootings in 2025 were the fewest in the NYPD's open files, which begin in 2006; recorded felony assaults passed 1998's count, on a definition that has widened.
Per subway ride, major felonies fell for thirty years. Felony assault broke from the fall.
In January to May 2026, felony assault per subway entry was about twice its 2019 rate, counting unpaid entries, while the six major felonies together sit close to 2019.
- Explore this dataset: the data and the code behind every number above.
- All articles · Search · RSS
SparkyData Weekly
Every week, the most surprising things New York's public data reveals, and the evidence behind them.
Free, weekly, and one click to leave. Your address is held by Buttondown, which sends it; see the privacy policy.
Sources
- NYPD — CompStat citywide report, week of 19 to 25 October 2026, and the seven major felony offenses, 2000-2025 (The report for the week ending 25 October 2026, retrieved 29 October 2026; the historical table through 2025) — New York's murders by year from 2001 and in 2026 to date
- NYPD — Complaint data (NYC Open Data), summed by month (January 2014 to December 2025) — New York's murders by month of occurrence, for the month-by-month comparison with London
- NYPD — press release of 6 October 2026 on the first nine months of 2026 (Published 6 October 2026) — The fewest murders in the department's records for January to September, quoted
- New York State Division of Criminal Justice Services — Index crimes reported to police by county (1990-2000 as published) — New York City's murders by year before 2001
- Metropolitan Police Service — Homicide dashboard data and borough-level crime (London Datastore) (Homicide victims January 2003 to December 2025; recorded offences October 2024 to September 2026; retrieved 29 October 2026) — London's homicide victims by year and month and its 2026 offences
- Metropolitan Police Service — London's homicide rate falls to record low (release of 12 January 2026) (Published 12 January 2026; retrieved 29 October 2026) — The 97 homicides, the rate of 1.1 and the comparison with other cities, quoted
- SSMSI (Ministère de l'Intérieur) — Bases statistiques départementales de la délinquance enregistrée, and their documentation (Edition of July 2026, years 2016-2025) — Homicide victims in Paris and the three inner départements by year
- Toronto Police Service — Homicides (open data) and its documentation (January 2004 to September 2026, retrieved 29 October 2026) — Deceased victims in the City of Toronto by the date of the killing
- Statistics Canada — Table 35-10-0071-01, Number and rate of homicide victims, by census metropolitan area (1981-2025) — The Toronto metropolitan area's homicides and rate, the long view
- Tokyo Metropolitan Police Department — Statistical yearbook (警視庁の統計), Table 30; National Police Agency — Crime statistics (犯罪統計資料) (Yearbooks for 2016-2025; the agency's provisional counts for January to September 2026) — Tokyo's victims who died, its deaths from assault, its recognised murder cases, and the 2026 cases to September
- Population estimates — U.S. Census Bureau, ONS (Nomis), INSEE, Statistics Canada, Tokyo Metropolitan Government (Retrieved 29 October 2026; Tokyo's as published before its revision to the 2025 census that day) — Every rate's denominator, each city's residents in the same year
How to cite
Reference
BibTeX
Quote the version: a correction bumps it, and the corrections log says what changed. Charts can be embedded from the “Embed this chart” link under each one.