How closely related are these two?

What the number means

The coefficient of inbreeding is the chance that a kitten or puppy inherits two copies of the same gene from the same ancestor, one down each side of the pedigree. Zero means no shared ancestors anywhere in what you typed. Twenty five percent is what you get from a full brother and sister, or from a parent bred back to its own offspring.

It's a number about the pedigree you gave it and nothing else. Four generations is the usual view, and it'll always read lower than the truth, because every breed has shared ancestors further back than anybody types in.

How it is worked out

For every ancestor that turns up on both sides, the calculator counts the steps from the sire up to that ancestor and the steps from the dam up to the same ancestor, then adds a contribution that halves with every step. An ancestor that appears twice on one side contributes through both routes. That's Wright's method, published in 1922, and it's still the coefficient registries publish today.

One simplification, so you know where you stand. Wright's full formula also raises a contribution when the shared ancestor is itself inbred. There's nowhere here to tell the calculator that, so this tool leaves that part out. It's the second reason the figure reads low. The first is the four generations.

You can check it against the pairings whose answers are known. A full brother and sister give 25 percent. Half siblings give 12.5. First cousins give 6.25. The buttons above set those up so you can see the working.

What to do with it

There's no single safe figure, and anybody who gives you one is guessing. What the number is good for is comparing two possible pairings for the same animal, and watching the trend across your own program over years. A rising average across a closed line is the thing worth catching early.

Putting a tool like this on your own site

Search for a color or an inbreeding calculator and most of what comes up belongs to individual breeders rather than companies. A tool is the one page on a small breeder site that other breeders link to without being asked. If you want one on your own site, that is on the cat breeder website page and the dog breeder page, with prices on the packages page.

Questions

What's a good coefficient of inbreeding?

There's no agreed number, and it varies by breed and by how closed the studbook is. Comparing two pairings against each other is more useful than judging one against a figure somebody quoted.

Why is my result lower than the registry's?

Two reasons. This uses only the four generations you typed, while a registry that publishes a figure runs the same coefficient over every generation it holds. The Royal Kennel Club says its calculators use all the pedigree information it has and don't limit the number of generations used. The second reason is the simplification above: an inbred shared ancestor pushes the real figure higher than this shows.

Does it allow for the shared ancestor being inbred?

No. Wright's full formula raises a contribution when the ancestor on both sides is itself inbred, and there's no way to type that in here. So on a closed line, where shared ancestors are often inbred themselves, the true figure is higher than this one.

Does it work for dogs?

Yes. The method has nothing to do with species. Names in, shared ancestors found, contributions summed.

Do names have to match exactly?

Yes. Two spellings of the same cat are two different cats as far as the tool is concerned, so use the registered name on both sides.