Yes, it’s late May and there are still more than seven full months remaining in 2026, but there has been enough data collected on this year to say that this year will likely set a new record for lowest murder rate ever recorded by the FBI when their Crime in the Nation report comes out sometime next summer or fall.
My confidence in this assertion comes from two main factors — first, the US murder rate in 2025 was likely at or below the lowest rate ever recorded by the FBI. I spent a good deal of ink on this assessment last year and don’t really want to rehash it all right now. The 2025 murder rate will be 4.1 per 100k when the FBI first reports it and should still be a record when the FBI revises it next year (the FBI reported a record low rate of 4.4 per 100,000 in 2014).
And, second, the available evidence strongly points to another sizable drop in murder nationally in 2026, even if it’s too soon to predict the exact strength of that drop.
Putting those two together — if murder was at or below the lowest rate ever recorded last year, and it’s falling even more this year then it isn’t a huge logical leap to assert that this year will set a new record low rate.
We have lots of data showing a big drop in murder is happening in 2026. The Gun ViolenceArchive is showing a 10.3 percent drop in 2026 relative to 2025 as of mid-May, and the Real-Time Crime Index and Major Cities Chiefs Association reported 19.9 and 17.7 percent drops through March respectively.
The GVA is particularly useful for our analytic purposes because it shows a large drop in gun violence in 2026…
But the degree of decline appears to be slowing, at least it has been over the last few months.
Another way to evaluate this year's trend is by collectijg data in 30 big cities with available data through the end of April. Doing so confirms a large drop is likely occurring with murder down 19.1 percent in that sample. This 30 city sample also points to the murder decline gradually slowing down in a way that it didn’t in 2024 and 2025. Murder in this sample was down more than 30 percent through January, 22.3 percent through February, 20.8 percent through March, and 19.1 percent through April.
Even at this point of the year, just four months in, a sample of this size is usually very good at predicting the scope of what the year’s murder change will be. I say usually, because it wasn’t in the extraordinary circumstances of 2020 (murder did not start truly surging until June/July) and 2021 (the huge YTD reflected murder surging in June/July 2020 and staying high). Every other year since 2018, however, the change in murder through April in this sample of cities has been within 2.5 percentage points of the final year end percent change reported by the FBI.
It’s reasonable to think that the decline in 2026 will be less big than it was in 2025 and perhaps 2024 as well, so this year’s decline probably won’t be 20ish percent. But 2024 and 2025 had the largest drops on record, and the drop by the end of the year should still be pretty sizable.
With a hypothetical 10 percent drop in murder in 2026 (and it’s far too early to say just how big this year’s decline may actually be), you’d get roughly the fewest murders nationally since 1967 and a murder rate below 4 per 100,000 which would easily be the lowest murder rate ever recorded. I’ve worked out what that all might look like in the below graph and included an expected upwards revision of the 2024 and 2025 murder counts by 750. Please remember that the 2024-2026 counts and rates are still inexact at this point.
As an aside, I will note that I use the term ‘ever recorded’ to convey the fact that we do not know when the FBI reported fewer murders. The FBI started estimating a national murder rate in the 1950s and has used a reasonably consistent methodology since 1958. Since then, they haven’t reported a lower murder rate, so saying it’s the lowest murder rate in 65 years (or however many years) reminds me of one of my favorite Simpsons jokes.
It’s also worth noting that 2025 had (and 2026 will likely have) the lowest murder rate ever recorded by the FBI, but it did not have the lowest homicide rate ever recorded by the CDC (or its predecessors). The CDC is reporting a 2025 homicide rate of 5.1 per 100k which is ever so slightly above 2014’s rate (5 per 100k) and a bit more above the recorded homicide rate in the 1950s — bottomed out at 4.5 per 100k a few times.
Still, that’s largely a distinction without a difference as far as I’m concerned. The bottom line is that murder and homicide are incredibly low by historical standards and both measures are either at or headed for record lows.
We are so far into uncharted murder reduction waters that I wouldn’t hazard even a guess as to what specifically the future holds. The enormity of 2020’s murder spike was certainly not known this far into the that year, though the sample data pointed to a sizable increase likely occurring.
Something could still happen which reverses this trend this year, though the odds of a second, sudden, once-in-a-lifetime murder surge occurring in is pretty unlikely. The rate of decline could become more blunted as 2026 goes on, but I'dbe very surprised if it completely reverses.
The US likely set a record for the lowest murder rate ever recorded in 2025, but even if it didn’t then a record low feels like a foregone conclusion sitting here in the first half of 2026.
New on the Jeff-alytics Podcast
What actually drives change on an issue like gun violence? Is it policy? Culture? Policing? Or something harder to measure?
To help answer that question I’m turning to Senator Chris Murphy of Connecticut. Senator Murphy has been one of Washington’s leading voices on gun violence reduction, a passion that was shaped by his experience in the aftermath of Sandy Hook, where he spent time with families and saw firsthand the impact of that tragedy.
In this episode, we talk about the recent decline in gun violence nationally, what role federal legislation and local interventions may have played, and how the broader movement around this issue has evolved over time.
You can also catch it on the Jeff-alytics YouTube page where I’ll be posting episodes and video clips, so be sure to like and subscribe there if you’re so obliged!
And while you’re here, be sure to check out these other recent great episodes:
Manhattan Institute Senior Editor Charles Fein Lehman
Civil rights attorney Jill Collin Jefferson
Law professor Rachel Harmon
What I find especially interesting is murders are down across the globe. So whatever is causing the trend is best explained by a global phenomenon. (E.g. teens inside on phones instead of going outside and taking on risk?)
I would offer that the spike in murders in the 30 largest cities and an apparent (deduced) simulatanious decline in rural areas during the covid spike might have been due to the rapid installation of 5G mostly concentrated in cities and nearby suburbs.
As noted in my longer comment those repeated frequencies cause irritation which can lead to murder and to increased traffic fatalities.