How to Calculate Your Dog's Age (Accurately)

You already know the × 7 rule is wrong. The AVMA says so itself: "Contrary to popular belief, dogs do not age at a rate of 7 human years per dog year." The giveaway is that a one-year-old dog can have puppies, and a seven-year-old child cannot.

What replaced it is a more sophisticated-looking chart: 15 human years for the first year, +9 for the second, then a size-dependent number of years after that. It's on hundreds of sites. It's usually cited to the AVMA or the AAHA. We went to check, and this guide is what we found.

The chart nobody can source

Here's the table you'll recognise:

Three things went wrong when we tried to verify it.

1. The AAHA guidelines say the opposite

The AAHA Canine Life Stage Guidelines (JAAHA 2019;55:267–290) is the single most-cited source for this table. We downloaded the full PDF and searched it. The phrase "human years" appears zero times. There are no coefficients, no weight buckets, no 15/+9. What there is, is a refusal:

"For practical purposes, rather than attempt to calculate age equivalents to humans, the task force suggests that life stage be defined both by age and characteristics. Life stage divisions, although arbitrary, provide a framework…"

The document cited as the authority for converting dog years to human years is a document that specifically recommends not doing it.

2. The AVMA publishes something else entirely

The AVMA's page has no conversion chart. It publishes lifespan by size — small dogs 8–11 years, giant breeds 6–7 — and defines senior as the last 25% of expected life. Those are lifespan bands, not aging rates.

Look closely and the fingerprints show. AVMA's size bands are under-20 / 20–50 / 50–90 / over-90 lb. The folk chart uses ≤20 / 21–50 / 51–100 / over-100. Similar, but mutated. It looks very much like someone took AVMA's lifespan bands, shifted the cutoffs, and repurposed them as aging coefficients the AVMA never published.

3. The chart comes in variants

Plenty of sites use +4/+5/+6/+7 rather than +4/+5/+5.5/+6.5. Both cite the same authorities. Real standards don't come in flavours — you can look up the actual value. Folklore does, because there's nothing to check against.

A warning if you go looking yourself

Search for this and you'll get confirmation. AI search summaries state confidently that the AVMA publishes these coefficients, and attach a citation to the AVMA. Follow the citation and the page doesn't say it. The summary layer is laundering the folklore into an authoritative-looking claim, which is probably a good part of how the chart spread this far. The only thing that breaks the loop is opening the primary document.

What the evidence actually supports

Size matters — much less than the chart implies

The largest study of dog lifespan to date is McMillan et al. (2024) in Scientific Reports: 584,734 UK dogs, 155 breeds, peer-reviewed, with confidence intervals. Median lifespans:

Ten months separates small from large. Now consider what the folk chart claims: a giant dog ages at +6.5 human years annually while a small dog ages at +4 — roughly 60% faster, forever. The real data doesn't just fail to support that. It contradicts its magnitude.

The size effect is real. It's simply nowhere near what you've been told.

Muzzle shape matters more than size

In the same study:

The gap from medium-muzzled to flat-faced is 1.6 years — double the entire size effect. Medium-sized flat-faced breeds showed a 2.69-fold faster time to death than small long-muzzled ones (median 9.4 years versus 13.3).

Note the middle row, because the obvious reading is wrong and we nearly published it wrong ourselves: long muzzles are not best. Long-muzzled breeds sit behind medium-muzzled ones. The familiar "small long-nosed dogs live longest" claim comes from a specific combination — small and long-muzzled reach 13.3 years — not from long muzzles being better on their own. This is the difference between a main effect and an interaction, and it's an easy thing to get backwards.

So the strongest single predictor in the data is the one no dog age calculator asks about. Not because anyone judged it unimportant — because the formula everyone inherited only had a slot for size.

Life stage is the answer the professionals actually give

AAHA's four stages — puppy, young adult, mature adult, senior — are defined by developmental milestones and by proportion of lifespan, not by birthdays. Senior is the last 25% of expected life. That definition needs a lifespan figure to be usable, and AAHA deliberately doesn't supply one, because it's "breed and size dependent". Slot in McMillan's medians and you get a fully sourced model.

It's also just the more useful answer. "Your dog is 62 in human years" doesn't tell you to do anything. "Your dog has entered the last quarter of its expected life, so switch to twice-yearly check-ups" does.

What we'd suggest

1. Use life stage as the primary answer. Our dog age calculator gives you AAHA's stage plus the share of expected lifespan lived, using McMillan's data.
2. If you want a human-years number, use Wang et al. (2020) — 16 × ln(dog age) + 31 — and hold it loosely. It's built from 104 mostly-Labrador dogs and takes no account of size. Its authors say so.
3. Take the size chart off your mental shelf. Nobody can show you where it came from, and the real numbers are much flatter.
4. If your dog is flat-faced, pay attention to that rather than to its weight. It's the bigger factor by some distance.
5. Watch the dog, not the chart. AAHA's framing — stages defined "both by age and characteristics" — is doing real work. Two dogs of the same breed and age can be in very different shape.

A note on why this guide exists

We set out to build an ordinary dog age calculator. The plan was to implement the standard chart, cite the standard sources, and move on. Reading the sources broke the plan, and rebuilding it around what the documents actually say took considerably longer.

We're writing it up because the alternative — quietly shipping the same chart as everyone else — would have been easier and nobody would ever have known. That's exactly the reason to say it out loud.