What age should a horse be started under saddle and what do Anderson, Parelli, and Schiller recommend?

The question of age for starting a horse under saddle is one that Clinton Anderson, Pat Parelli, and Warwick Schiller each address, and while their practical answers are similar, their reasoning reflects their different philosophical frameworks. Clinton Anderson starts horses at two years old in his program, which is consistent with the performance horse industry standard in western disciplines. His reasoning is practical and biomechanical: a two-year-old horse has enough physical development to carry a rider's weight for short sessions, and the work done at two is light enough — walk, trot, basic direction control — that it does not stress developing joints. He distinguishes starting, which involves introducing the horse to carrying a rider, from finishing, which involves the athletic demands of competition. Starting at two; finishing at four and beyond. Pat Parelli's position acknowledges that starting age depends on the horse's mental and physical development more than its chronological age. He has started horses at two and has also waited until four with horses that were not mentally ready. His Level system provides an objective measure: a horse that is solid at Level 2 on the ground is ready to be started regardless of age. A horse that is not at Level 2 at three years old should not be started regardless of age. Warwick Schiller's view has evolved toward waiting longer than industry standard, and he has spoken specifically about the emerging research on skeletal development in horses — including evidence that the spine and back are among the last structures to fully develop, closer to five or six years old in many breeds. He does not advocate waiting until six to start horses, but he does advocate lighter, shorter sessions when starting young horses and significantly longer gaps between sessions at two and three than most training programs allow. His position is that the industry standard of starting at two is commercially driven rather than biomechanically optimal.