I mean yeah, sure. One of the recent projects I was on included a survey of ortho programs to find out how many people were either home matches or matched after doing an audition, versus the number matching to an unaffiliated program. Turns out it's a majority that land at home or an away. So it's already becoming a norm and I see no issue with that. I can't imagine a better way to assess whether you want to train/work with someone for many years, than to test them out for an entire month.
Right, and no one will know how that dynamic changes once step goes P/F. If 10 people enter medical school interested in ortho, 3 of them get 240s and the rest get 210s, then only 3-4 people are going to apply ortho (and do sub-Is). After step goes P/F, there is no reason why 8-9 people wouldn't apply ortho since there is no way to know if you are a good applicant until match day. You might argue that sub-Is will decide who is a good applicant, but we all already know that most LORs are positive and paint applicants in positive light.