Search "dog bite statistics by breed" and you will find dozens of pages ranking breeds from "most dangerous" to "safest." Almost all of them recycle the same fatality counts and present them as if they measured risk. They don't. This guide gives you the same numbers — sourced and dated — but also explains what they can and cannot prove, so you can use them without being misled. It is written for accident victims, dog owners, journalists, and anyone weighing the real-world numbers behind dog bite liability.
How many dog bites happen in the U.S. each year?
Dog bites are common because dogs are common. With nearly 90 million dogs in the country, even a very low per-dog bite rate produces millions of incidents. Children are bitten most often, and bites to the face and head are disproportionately common among young kids simply because of their height relative to a dog's mouth. Most bites are minor and never reported to a doctor, an insurer, or animal control — which is the first reason every "by breed" dataset is incomplete: the denominator (all bites) is largely invisible.
Two datasets capture the visible tip of the problem reasonably well: homeowners insurance liability claims (tracked by the insurance industry) and fatal attacks (recorded in death certificates via the CDC). Neither was designed to compare breeds, but together they tell us a great deal about scale, cost, and trend.
Dog bite insurance claims are surging — and getting more expensive
The clearest, most consistent dog bite dataset in America is not about breeds at all — it is the Insurance Information Institute's annual tally of homeowners liability claims, compiled with State Farm. It shows a steep, sustained rise in both how often dogs injure people badly enough to trigger a claim and how much those injuries cost.
The table below shows the full decade. Note that "dog bite claims" here include other dog-related injuries — for example, a person knocked down and fractured by a large dog — not only bites.
| Year | Value of claims ($M) | Number of claims | Average cost per claim |
|---|---|---|---|
| 2016 | $602.2 | 18,123 | $33,230 |
| 2017 | $686.3 | 18,522 | $37,051 |
| 2018 | $674.9 | 17,297 | $39,017 |
| 2019 | $796.8 | 17,802 | $44,760 |
| 2020 | $853.7 | 17,597 | $50,245 |
| 2021 | $881.9 | 17,989 | $49,025 |
| 2022 | $1,136.0 | 17,597 | $64,555 |
| 2023 | $1,116.0 | 19,062 | $58,545 |
| 2024 | $1,569.6 | 22,658 | $69,272 |
| 2025 | $1,862.1 | 28,450 | $65,450 |
| Change 2016–25 | +209.2% | +57.0% | +97.0% |
Why the rise? Insurers and analysts point to higher medical costs, larger settlements and jury awards, and more dogs in more homes following the pandemic-era adoption surge. The average cost per claim actually dipped 5.5% in 2025, but only because claim volume jumped so fast; the long-term cost trend is sharply upward.
Dog bite insurance claims by state
State totals track population, dog ownership, and — critically — each state's liability rules. States with strict-liability dog bite statutes and high medical and legal costs tend to produce both more claims and more expensive ones.
| Rank | State | Number of claims | Avg. cost per claim | Value of claims ($M) |
|---|---|---|---|---|
| 1 | California | 2,830 | $81,789 | $231.5 |
| 2 | Florida | 2,347 | $62,375 | $146.0 |
| 3 | Michigan | 1,432 | $68,018 | $97.0 |
| 4 | Ohio | 1,348 | $41,413 | $56.0 |
| 5 | Pennsylvania | 1,324 | $68,786 | $91.0 |
| 6 | Texas | 1,313 | $70,885 | $93.0 |
| 7 | New York | 1,308 | $92,154 | $121.0 |
| 8 | Illinois | 1,283 | $79,596 | $102.0 |
| 9 | Indiana | 824 | $53,396 | $44.0 |
| 10 | New Jersey | 818 | $77,447 | $63.0 |
| U.S. total | 28,450 | $65,450 | $1,862.1 |
Fatal dog attacks by breed: what the data actually shows
This is the number everyone wants, so here it is — with the caveats that make it honest. Fatal dog attacks are rare relative to the 4.5 million annual bites, but they are the one outcome serious enough to be recorded consistently. The CDC records dog-bite deaths through death certificates (CDC WONDER). Those records show a clear upward trend:
| Year | Recorded dog-bite deaths |
|---|---|
| 2019 | 48 |
| 2020 | 62 |
| 2021 | 81 |
| 2022 | 98 |
Crucially, CDC WONDER does not record breed. The CDC studied breeds once, in a landmark 2000 report (Sacks et al., published in the Journal of the AVMA) covering 1979–1998. That study found that at least 25 breeds were involved in 238 dog-bite-related fatalities, and that pit bull-type dogs and Rottweilers together accounted for about 51% (roughly 122 deaths) over the 20-year period. The CDC has not published breed-attributed fatality data since — by its own account, because breed identification in these reports proved too unreliable to support breed-specific conclusions or policy.
| Breed group | CDC/AVMA (Sacks et al.), 1979–1998 | DogsBite.org, 2005–2017 |
|---|---|---|
| Pit bull-type | Part of ~51% with Rottweilers (122 deaths combined) | ~66% (284 of 433) |
| Rottweiler | Part of the combined 51% | ~10% |
| Other (German Shepherd, mixed breeds, Husky, Mastiff, etc.) | Remaining ~49% across ≥23 breeds | Remaining ~24% across many breeds/mixes |
| Distinct breeds identified | At least 25 | Dozens (incl. many mixes) |
Why "bite rate by breed" is usually the wrong question
This is the single most important thing to understand about every breed ranking online, and it is why the AVMA, the ASPCA, and the CDC all decline to endorse breed-specific danger lists. A rate is incidents divided by population. We have a rough numerator (recorded incidents) but no trustworthy denominator (how many pit bulls, Labs, or Rottweilers are alive in the U.S.). Without it, "pit bulls cause X% of deaths" tells you about the population mix and reporting patterns — not the odds that a given dog of that type will hurt someone.
A useful analogy: the most-driven car models appear in the most accidents. That does not make them the most dangerous cars; it reflects how many are on the road. Breed bite counts work the same way.
Breed identification in bite reports is unreliable
The second structural problem is that the breed label attached to a biting dog is often a guess. "Pit bull" is not a single breed; it is an umbrella term covering the American Pit Bull Terrier, American Staffordshire Terrier, Staffordshire Bull Terrier, and a wide range of mixed-breed dogs with a certain look. Studies of shelters and veterinary clinics have repeatedly found that visual breed identification — including by professionals — frequently disagrees with DNA testing, and that dogs are labeled "pit bull" far more often than DNA confirms. When the identification feeding a dataset is inconsistent, breed-level comparisons inherit that error. Insurers know this, too: the Insurance Information Institute notes that many insurers do not even record breed, and some do not ask, in part because of the identification problem.
Nonfatal bites by breed: hospital and trauma data
Fatal attacks are a tiny, atypical slice of all bites. For the far larger universe of nonfatal bites, the evidence comes from individual hospital and trauma-center studies — useful, but small, local, and not nationally representative. They also illustrate the exposure effect, because common family breeds show up prominently.
| Study / setting | Top breeds reported | Source |
|---|---|---|
| Pediatric facial bites (children's hospital) | Mixed breed 23.0%, Labrador retriever 13.7%, Rottweiler 4.9%, German shepherd 4.4% | Chen et al., 2013 (NIH/PMC) |
| Pooled Level-I trauma center studies | Pit bull-type 22.5%, mixed breed 21.2%, German shepherd next highest | DogsBite.org summary of trauma studies |
Notice that a breed widely considered a "family dog" — the Labrador retriever, the most popular breed in America for decades — ranks near the top of pediatric bite studies. That is not evidence Labs are dangerous; it is evidence that there are enormous numbers of them around children. The same exposure logic cuts in every direction.
Popularity vs. reports: the confounder in one picture
Behavior research consistently finds that individual factors predict bite risk better than breed: whether a dog is spayed or neutered, whether it is chained or isolated, prior aggression, the owner's handling, and the situation (resource guarding, pain, fear, an unsupervised child). Breed is a weak predictor once these are accounted for, which is why major veterinary and animal-welfare organizations oppose breed-specific bans in favor of owner- and behavior-based rules.
Who gets bitten most — and where
Breed obsession distracts from the more actionable risk factors: who is bitten, and in what situation. Young children are bitten most often and most severely, frequently in the face, and usually by a familiar dog in or near the home rather than a stray. Delivery and postal workers are a second high-exposure group — USPS data is one of the few consistent national series on bites to adults, and it shows attacks rising for years before dipping in 2025. Most serious bites happen during ordinary interactions (reaching for food or toys, startling a sleeping dog, an unsupervised child and dog together), which is exactly why prevention experts focus on supervision and behavior rather than breed labels.
What the numbers mean if a dog bites you
There are broadly three legal frameworks. Under a strict-liability statute, the owner is liable for an unprovoked bite regardless of whether the dog had shown aggression before. Under the one-bite rule, an owner is liable if they knew or should have known the dog was dangerous — so the victim must show prior knowledge. Under negligence, liability turns on whether the owner was unreasonably careless (for example, violating a leash law). Compensation for a bite typically comes from the owner's homeowners or renters insurance, and the claims data above shows why these cases matter financially: the average paid claim now exceeds $65,000 nationally, and far more in states like New York and California.
If a dog has injured you or a family member, the practical questions are which liability rule your state uses, whether insurance coverage applies, how your medical costs and lost income are documented, and what your claim may be worth. A general-purpose AI Lawyer tool can help you understand these concepts, organize records, and draft questions — but it is not a substitute for a licensed attorney. Dog bite injuries, especially to children or involving the face, can have lasting medical and legal consequences; consult a qualified personal injury lawyer in your state before making decisions about a claim.
Do breed bans actually reduce dog bites?
Breed-specific bans — outlawing or restricting "pit bulls" and similar dogs — are politically popular after high-profile attacks, but the evidence does not support them. The AVMA notes that in controlled studies breed is a poor predictor of biting, and the unreliability of breed identification (the same problem that undermines the statistics above) makes bans hard to enforce fairly. Insurance regulation is moving in parallel: a few states now limit insurers' ability to deny homeowners coverage based purely on breed. Most public-health guidance favors breed-neutral "dangerous dog" laws that target individual animals and owners with a record of aggression. This is also where the law gets genuinely complicated, and where understanding your local rules matters; tools like Legal AI can help you research your jurisdiction's ordinances and liability standard before you act.
Dog bite settlements and related injury claims
Dog bite claims sit alongside other personal-injury matters in how they are valued: medical bills, future care, lost wages, scarring and disfigurement, and pain and suffering. For realistic benchmarks by injury severity, see our companion guide to dog bite settlement amounts, or use the free dog bite settlement calculator to estimate a state-aware range. If you are researching what an injury claim might be worth more broadly, our data-driven guides to the average car accident settlement and, in the tragic cases of fatal attacks, the wrongful death settlement process explain how damages are calculated and what factors move the number. For families dealing with a death, our explainer on whether wrongful death settlements are taxable covers the financial aftermath.
Data sources and limitations
Specifically: insurance claim data (Triple-I/State Farm) captures only injuries serious enough to generate a homeowners or renters claim, and it bundles bites with other dog-related injuries. CDC WONDER fatality data is reliable for counts but records no breed. The 2000 CDC/AVMA study is authoritative but old (1979–1998) and its authors warned against using it for breed-specific policy. DogsBite.org compiles modern fatality counts from news reports and is an advocacy organization, so its breed attributions should be read as media-reported, not DNA-confirmed. Hospital bite studies are small and local. We have labeled every figure with its source and date so you can weigh it accordingly.
Frequently asked questions
Which dog breed has the highest bite rate?
There is no reliable answer, because a true "rate" requires knowing how many dogs of each breed exist, and no trustworthy national breed-population count exists. By raw count, pit bull-type dogs appear most often in recorded fatal attacks (about 66% in the DogsBite.org 2005–2017 dataset), but that reflects population size, reporting, and inconsistent breed labeling as much as any per-dog tendency. The CDC stopped attributing fatal attacks to breeds in 2000 for exactly this reason.
What breed of dog is least likely to bite?
No dataset reliably ranks breeds from safest to most dangerous, so any "least likely to bite" list is speculative. Veterinary and animal-welfare organizations emphasize that individual factors — training, socialization, neutering, supervision, and the situation — predict bite risk far better than breed. Even breeds with gentle reputations bite; Labrador retrievers, among the most popular family dogs, appear near the top of some pediatric bite studies simply because there are so many of them.
Does the CDC publish dog bite statistics by breed?
Not anymore. The CDC published a single breed-focused study in 2000 (covering 1979–1998) and has not released breed-attributed fatality data since, citing the unreliability of breed identification. Current CDC WONDER data records dog-bite deaths but not breed. Any page claiming "current CDC breed statistics" is misattributing other sources.
How much does the average dog bite claim cost?
The national average homeowners insurance payout for a dog-related injury claim was $65,450 in 2025, according to Triple-I/State Farm. Averages vary widely by state — from about $41,000 in Ohio to over $92,000 in New York — driven by injury severity, medical costs, and local liability rules.
Are dog owners legally responsible if their dog bites someone?
Usually, yes, but the standard depends on the state. Twenty-nine states have strict-liability statutes making owners responsible even for a first bite; others use the one-bite rule or negligence; four states have no specific statute. Compensation typically comes from the owner's homeowners or renters insurance. Because the rules and exceptions (such as provocation or trespassing) vary, consult a personal injury attorney licensed in your state.
Why do pit bulls appear so often in fatal-attack statistics?
Several reasons combine: "pit bull" is a broad label covering several breeds and many mixes, so it captures more dogs than a single breed would; these dogs are numerous in the U.S.; and bites by powerful, large-headed dogs are more likely to be severe and therefore recorded. Visual breed misidentification also inflates the "pit bull" category. Together these mean the high counts reflect labeling, population, and severity — not a measured per-dog risk rate.
This article is for general informational purposes and is not legal advice. Statistics are sourced and dated above; figures change as agencies update their data. If a dog has injured you or your family, speak with a licensed personal injury attorney in your state.