The interesting thing about this comment, is that the horses shown in those pictures are not only from the same breed, but the vast majority of them are from the same family
Because there doesn't appear to be a breed standard. You've chosen multiple horses that look NOTHING alike and are rebuilding using them. Where is the standards that you are using? If I hadn't have seen this thread and I saw all those horses posted, I never would have thought there was a certain 'breed' attached to them. Never.
(and the old lines that we are trying to correct with the influence of the modern lines). The majority of the horses being shown were not chosen.
They were born into the breed, most before our current staff became involved with the breed. Granted, they are shown in some pretty horrible pictures! Most of us know about how photos can distort the horse, and how easy it is to get a casual shot of a horse leaning toward the owner/handler. With that said, many of the comments on leg structure are valid. We are bringing in new blood to try and improve that. The Foundation horses are judged based upon their joint angles, and ratios of body parts to their own body.
As an example, the ideal shoulder slope should be a 45 degree line from the point of shoulder to the withers. The shoulder angle from the scapula to the humerus should be 90 degrees or larger. Some horses approved have a 47 degree shoulder with a 79 degree angle between scapula and humerous, while others have a 51 degree shoulder with a 92 degree angle. Very few draft horses have a true 45 degree shoulder slope, sadly, which is why the horses must be within 70% of the ideal.
Another example is length of back. The ideal is between 33 and 35% of the horse's body length. So what do you do if you have a horse with a 45/90 shoulder and a 42% back? Well, you average out the flaws via a scoring system. The horses are scored on a scale of 1 to 10 for each point. Points are based on the deviation from the ideal formula. Each degree off from the ideal removes a point, but points can be gained back for specific conformational improvements, such as the open shoulder angle mentioned above). To receive a passing score in one area, a horse can have a 48 degree shoulder, or a 36% back, but not both. (I am overly simplifying the conformational analysis process to prevent from writing too much of a book).
Fatal flaws are cause for denial, but we see few of those in the horses applying. Each owner received a score sheet with their horse's analysis that very clearly breaks down the horse's flaws.
These scores are the owner's to share or not, as they are pretty harsh. We mention how to improve on those areas, and offer free advice for analyzing breeding pairs. We don't just turn them loose and say "ok, go make more!". In fact, I have spent hours explaining genetics (including genetic diseases, and DNA testing for them) conformation (including how conformation effects soundness and movement) and such to the horse owners. I will not (legally can not) tell them how to breed or care for their hoses, but we offer every tool possible for these horse owners to help the breed improve not stagnate. The misunderstanding here though:
Seems to be that the horses shown (horribly) in this thread are the horses we have been searching for. Instead, there are only 2 Foundation horses shown, horse number 2 and horse number 3. Horse number 3 barely passed the Foundation requirements. Her long neck made up for her sloped shoulder, and her straight legs made up for her long back (again, I'm overly simplifying the analysis process). Since horse number 3 had to be approved to be registered, she has fewer flaws than some of the horses born into the breed, and when bred back to the old lines, her conformation improved the set of the legs under the horse, increased the length of neck, but the sire shortened the back, and lowered the angle of the shoulder.
Here is the resulting foal.
This filly is the goal. A cross between the old lines, and the new, with the conformational flaws (mostly) removed. That filly is registered as a G4, which means she is not yet the ideal, and only carries 87% draft ancestry. She is shown at 5 months of age.