Breeders are, or should be, the best judges of their dogs. I received a correspondence recently requesting that I address breeding theories other than the oft-discussed explanations of linebreeding, inbreeding, and outcrossing. This request was well written and asked that the investigation include breeding systems designed to reach a given goal, which all breeding activities should seek.
Important elements in the query included the acknowledgement that selection skills must be honed to perfection for any breeding plan to work. The writer fully understands that to breed quality dogs requires the breeder to have proficiency in pedigree analysis of the dogs being used, the identification of both desirable and undesirable traits associated with those dogs, and the ability to evaluate resulting offspring.
You can't get much more on target than that, as selection is the key to your breeding success, no matter what particular system or combinations thereof you might use in your breeding program. Consequently, in order to become a breeder of quality dogs, you must also become an objective, knowledgeable judge of your own stock. As parents are, or should be, the best teachers for their children, breeders are, or should be, the best judges of their dogs.
The breeding of like-type to like-type is an approach that has been used by some of our best breeders for decades. When you select an individual to retain in the gene pool, you are showing confidence in its correctness of type. Therefore, when you attempt to reproduce that type in the resulting offspring, you must select a mate of the same type for that individual. "Like begets like" is an old axiom that holds true. You are not trying to make changes. You are trying to preserve the qualities you already have by reducing genetic variation of type traits.
A strong breeding program produces animals of like type that conform to the window of correctness for breed type as described by the breed standard. There is a family resemblance among the animals that identifies them as products of a given breeding program. Others may say that these animals appear to be shaped by a cookie cutter. They are consistent in physique, athleticism, behavior, and performance. They have the same look (phenotype) because of selection. In a way, this method of breeding “like to like” simulates close breeding even if the parents are not closely bred because the breeder is always picking the same kind of dog.
Trial and error probably contributes to breeders learning to stick with the "like to like" theory as dogs lacking the desired characteristics are eliminated from the breeding population. And only the best of each generation is deemed worthy of getting into the gene pools of the future.