About PawCurve
Research-based pet calculators with transparent methods.
Why this exists
Most pet calculators are a single line of arithmetic behind a nice button. The most common puppy weight formula on the web — weight ÷ age × 52 — overshoots by a median of 45%, and by 82% on toy breeds. It is not a rounding problem. It tells a Chihuahua owner to expect roughly twice the dog they will get.
That bothered us enough to do the work properly: download the raw veterinary dataset, recompute the growth curves from 8.5 million weight records, and then — the part almost nobody does — test the predictions against what those dogs actually grew into and publish the error bars. Our methodology page shows the whole derivation, including how the model is built, how it is tested, and how to read the uncertainty around each estimate.
Editorial standards
- Every formula names its source. We cite the author, year, journal, DOI, or primary source wherever available, so readers can trace the number back to evidence instead of vague claims.
- Uncertainty is shown clearly. When an estimate has a range or measured error, we publish it. That makes the answer more useful, not less.
- Scope is deliberate. PawCurve focuses on age, growth, lifespan, pregnancy timing, and feeding-start estimates: areas where published population data can be used responsibly.
- Clear safety boundaries. PawCurve does not provide medication dosing, symptom checking, or diagnosis. Those decisions belong with a veterinarian who can examine your pet.
- Health decisions deserve real context. Our tools support planning and comparison. When a pet seems unwell, is far outside an expected range, or needs medical care, the right next step is an examination by a veterinarian.
How to check every number yourself
Everything here is built to be checked, not just believed. The raw dataset is public and openly licensed, the full derivation is on our methodology page, and every accuracy figure is measured on tens of thousands of dogs held out of the fitting. You can trace any number back to its source — we built it that way on purpose.
Our trust model is evidence-first: named sources, visible assumptions, measured accuracy where possible, and a clear path from the calculator output back to the data behind it.
When expert reviewers contribute to a page or method, we will name the reviewer, their credentials, and the specific scope of their review, using the same source-first standard as the rest of this site.
Corrections
Accuracy is central to PawCurve. If a material number or interpretation changes, we document what changed and when, so readers can see the site's reasoning improve over time. If you spot a material issue, please contact us with the source you are comparing against.
Independence
We build openly, and with full credit, on data and standards from groups like WALTHAM, Mars Petcare, Banfield Pet Hospital, the University of Liverpool, and the American Kennel Club — but the site is ours alone. None of them are involved in it or have any say in what it concludes, and that independence is exactly what lets us flag where a widely repeated figure doesn't hold up.
Last reviewed 2026-07-17.