Phenotype recording

Horse breeding is characterized by an extremely wide spectrum of traits to be considered when making breeding decisions (this issue is addressed in more detail in the section trait groups). The descriptions of the breeding goals provide some hints for horse owners, riders and breeders where to focus. However, when it comes to specific aspects, it is often not easy to find clear directives on the basis of the available trait definitions and the accessible trait information (phenotypes or breeding values).
Experiences with the new traits (for example from the trait groups health and behavior) are still limited, and we are also not rarely challenged by the long-established conformation and performance traits.

The experienced judges and horse people surely know what to look at when asked to evaluate for example the trot of a horse. However, individuals are very likely to differ with regard to how they see and weight certain aspects for their evaluation. The outcome of the evaluation does not reflect these details. From the traditional system of assigning grades on a scale from bad to excellent, called valuating scoring, you get some scores for few globally defined traits. Because the evaluation is relative to the breeding goal, you only know how close to or far from the breeding optimum the horse's phenotype appeared to be. The maximum score of 10 tells you, that the respective trait expression was seen as ideally reflecting the breeding goal; any lower score does not give any hint to what was not in line with breeding goal. Subjectivity, low repeatabilities and lacking transparency give reason to major concern and have motivated searching for alternatives.

Linear descriptions of traits relative to biological extremes have the clear advantage of independence from breeding goals. There are many examples for successful use of linear traits in animal breeding, and more recently the importance of linear systems for horses have increased, too. They are referred to as linear scoring or - in order to avoid confusion and misinterpretation with the valuating scoring - linear profiling. Instead of expressing how bad or good some phenotype is relative to a theoretical optimum (valuating score), the linear value reflects the proximity of the phenotype to the most extreme expressions. For example, the length of the neck of a horse can be described between extremely short and extremely long - without any consideration of which neck length would probably fit the breeding goal best. The second beneficial change in linear systems is the refinement of trait definitions, being linked to the departure from the reference to more globally defined breeding goals. This more objective and more detailed information of linear profiles provides transparency of the assessments of horses that cannot be achieved in the traditional system of valuating scoring.
A judge may tend to give a lower (= worse) valuating score for trot because of over-excited knee action or because of lacking thrust or hind limb activity. In the linear system, distinct linear values are assigned to the different aspects of trot which are relevant for the breeder and rider. Accordingly, it is clearly documented which horse showed extreme knee action or inactive hind limbs.

More detailed information on linear profiling practices and their development can be found in the respective sections: