Does Ring prevent break-ins?

For years, the most quoted statistic in the video-doorbell world was that Ring cameras cut burglaries 55% in a Los Angeles neighborhood pilot. Ring installed about 40 doorbells in the Wilshire Park neighborhood starting in 2015 and announced the result in March 2016, at a joint press event with the LAPD. The number traveled from there into news coverage and a thousand blog posts. But Ring never released the data or methodology behind it; when MIT Technology Review asked in 2018, the company would not supply any.

The LAPD's own follow-up analysis, covering ten months from July 2015, found a 21% year-over-year decline — 42% after adjusting for the surrounding area — and in 2017 Wilshire Park had more burglaries than in any of the previous seven years. The only test MIT Technology Review found that was run independently of Ring, in West Valley City, Utah, found the neighborhood without the doorbells fared better. If the effect were large, someone should have been able to point at it.

So retire the 55% number entirely. It is not a weak statistic; it is an unverifiable one, and any article still citing it as fact has told you something useful about its sourcing standards.

The independent picture is thin: the few attempts to link video doorbells to measurable reductions in break-ins have come up empty or inconclusive. Honest hedges apply — crime studies are genuinely hard, adoption is uneven across neighborhoods, and absence of evidence is not proof of no effect. But after years of mass deployment, the burden of proof sits squarely on the claim, and the claim keeps failing to meet it.

The verified offender research assigns cameras a modest, mixed role rather than a protective one. In the 2013 UNC Charlotte survey of 422 incarcerated burglars (industry-funded, and disclosed as such), outdoor cameras appear inside the broader "signs of security" bundle that burglars weigh — real, but well below occupancy and escape routes. And in KGW-TV's 2017 interviews with 86 Oregon inmates, cameras cut both ways: a deterrent to some, a hint of valuables inside to others.

In other words: even the most camera-favorable data treats a camera as one cue among many, not a wall. The doorbell's specific problem is what it does after the cue fails.

Strip the branding and a video doorbell does three things: watches one narrow approach to the house, records what crosses it, and sends a push notification. Everything beyond that routes through a distant human — you — noticing the notification, judging it correctly, and doing something useful about it, remotely, in time to matter.

Each link in that chain is weak at exactly the wrong moment. FBI data for 2024 shows daytime residential burglaries (216,601) outnumbering nighttime ones (174,053): burglars work your work hours, which means the notification lands where you are — a meeting, a commute, a phone on silent — not where the burglary is. The doorbell's output is information. The break-in is a physical event. Nothing in the product converts one into the other.

Run the standard residential burglary past a doorbell and watch what it contributes at each step:

Nothing malfunctioned in that timeline. That is the point. The product did its job at every step — and its job was documentation and messaging, not intervention. The chime rang, the clip saved, the notification sent. The burglary proceeded.

Fairness requires the other column, because it is real. Package theft is the doorbell's home turf: SafeWise's 2025 survey of 7,500 US adults found 31% had a package stolen in the past twelve months, with a median loss of $143, and the USPS Office of Inspector General's May 2025 whitepaper stands as the strongest federal source on the problem. A doorbell documents porch theft, guides deliveries, and occasionally identifies a repeat offender for an entire street.

The occupancy knock is the other genuine win. Since burglars overwhelmingly ring first, a doorbell you answer quickly — from anywhere — lets the house pass the test it is actually being given. Add everyday visitor screening and evidence quality that police reports and insurance claims can actually use, and the verdict writes itself: a video doorbell is worth having for what it does. Just be precise about what that is.

The clearest evidence that notifications alone fall short comes from the notification vendors themselves, all of whom now sell response as an upgrade. As of mid-2026: Ring offers Virtual Security Guard, a professional-agent add-on around $99 a month; SimpliSafe's Active Guard tier, around $80 a month for 24/7 coverage, adds human agents who can intervene through up to three outdoor cameras; Deep Sentinel builds its entire product on live guards in a surveillance center, from roughly $100 a month; Alarm.com introduced an automated AI deterrence service, AID, at CES 2025; and RAD launched its RADCam talking camera in 2024.

These are competent monitoring companies making a rational admission: watching and notifying was never the whole job, so they are renting you the missing piece — a human, or lately an AI, that can act while the event is happening. The market has priced the gap this article describes at roughly eighty to a hundred dollars a month, indefinitely.

If you typed "does Ring prevent break-ins" into a search bar, you were probably shopping the home security aisle for something that goes beyond notifying you. Everything in that aisle — Ring, Nest, ADT, SimpliSafe, and the rest — watches, records, and notifies, and does it competently; the premium tiers add a person who responds from somewhere else. PerimeterEye belongs to a different category: an Automatic Property Defense System, which acts on the property itself, in seconds, automatically, while you watch live. PerimeterEye is not an alarm-monitoring service — alerts and control go directly to you.

Replay the timeline. The knock still happens — but the system has already verified the person, and it knows the household, so a delivery or a family arrival passes quietly. A stranger who steps off the approved path gets a different property: a motorized searchlight locks onto them and physically follows them, a live targeted voice describes them specifically — the person, the clothing, the location — and if they stay, a 130 dB siren sounds while priority alerts with live video reach your phone. You can watch, talk live, or take manual control at any point, and there is nothing to arm or disarm.

No security system can guarantee prevention of every incident. The claim here is narrower and checkable: during those first two minutes — the only minutes that matter — something happens at the scene, in light and voice and sound, instead of only in your pocket.