What a coefficient of inbreeding actually tells you

Coefficient of inbreeding — COI, sometimes written CoI — is one of those numbers that everybody quotes and few people can define. It is worth ten minutes, because it is one of the very few figures in dog breeding that describes a pairing rather than a dog, and because a small caveat printed next to it changes what it means entirely.

What the number actually measures

A coefficient of inbreeding is a probability, expressed as a percentage. It estimates the chance that, at any given position in the genome, the two copies a dog inherited — one from each parent — are identical by descent: that is, both are copies of the same gene in the same shared ancestor.

Put less technically: it is a measure of how closely related a dog’s parents were, summarised over the whole recorded pedigree rather than over one relationship. A COI of 0% means the calculation found no common ancestors in the pedigree it could see. A COI of 25% means the parents were about as closely related as full siblings.

What the standard relationships produce

These are the figures the arithmetic gives for a single relationship, assuming the rest of the pedigree is unrelated — which in a closed breed population it almost never is, so treat them as landmarks rather than as answers:

In a real pedigree these relationships stack. Two dogs who are “only” first cousins on paper can produce a much higher figure once the rest of the pedigree is counted, because the same influential ancestors appear repeatedly further back.

Where you will see one, and what sits next to it

The Royal Kennel Club publishes a COI on its public dog records. Three things appear alongside it, and all three matter:

• The dog’s own figure, as a percentage.
• How many generations the calculation had, and how many of those were complete — for example, twenty generations available of which seven are complete.
• The breed average, so the figure has something to be read against.

The caveat that changes everything: incomplete generations

This is the part that gets skipped, and it is the most important sentence on this page. A COI can only find the common ancestors that are actually recorded.

If a pedigree runs twenty generations deep on paper but only seven of those are complete, then thirteen generations of possible shared ancestry are simply invisible to the sum. Ancestry that is missing cannot add to the figure — which means an incomplete pedigree can only ever make a dog look less inbred than it is, never more.

The number that matters is the pairing’s, not the dog’s

A stud dog’s own COI tells you how related his parents were. It does not tell you what your litter would be, because that depends entirely on how he and your bitch are related to each other.

Two dogs each with a low COI can produce a high-COI litter if they share the same recent ancestors. Two dogs with unremarkable figures from different lines can produce a lower one than either. So the useful question to a stud owner is not “what is his COI?” but “what would this pairing’s COI be, and how does that compare with the breed average?” The Royal Kennel Club offers a mate-selection service that calculates the figure for a hypothetical pairing.

Why breeders pay attention to it at all

Two reasons, and they are different.

Recessive conditions. Many inherited diseases need two faulty copies of a gene to cause illness. The more likely it is that a puppy inherits two copies of the same thing from a shared ancestor, the more likely it is that both copies are the faulty one. That is the direct, mechanical link between relatedness and risk.

Inbreeding depression. Across domestic animal species, rising inbreeding is generally associated with declines in fertility, litter size, immune competence and longevity. This is a broad finding in animal breeding rather than a per-breed prediction — and it is the reason breed health strategies talk about the direction of travel in a whole population, not just about individual matings.

What COI does not tell you

It is a probability derived from a family tree, not a look at the dog. It cannot tell you which variants a dog actually carries. A dog with a very low COI can still be a carrier of a serious recessive condition, and a dog with a higher COI can be clear of everything its breed is tested for.

So COI does not replace DNA testing or the screening schemes; it sits alongside them. See health testing explained for what the tests themselves answer.

What Breed Right shows

Where a stud owner has given us a coefficient of inbreeding, it appears on the listing labelled “COI (declared)” — because that is exactly what it is. We do not calculate pedigrees, and we do not hold or copy Royal Kennel Club data, so a COI on this site is the breeder’s own figure. If a listing shows no COI, it means nobody gave us one; it is not a finding about the dog.

Questions worth asking

• What would this pairing’s COI be?
• How does that compare with the breed average?
• Over how many generations was it calculated, and how many are complete?
• Which ancestors appear on both sides of the pedigree?
• What does our breed club’s health guidance say about it?

That last one matters more than any general rule. Breeds differ enormously in population size and history, and the people who know what a figure means in your breed are the ones working on that breed’s health.