I'm arguing that this assumption is wrong. First, how can a med school possibly know what the future plans are for any student? For many people, subjective things like location and personal life are a lot more important when choosing a specialty or residency compared with choosing a med school. Second, just about anyone who gets into med school somewhere in the US is capable of going on, thriving, and matching at a decent program, although maybe not in derm or something. Very few American grads don't match, and if they do, they probably got some crappy advising and it's not a surprise.
As Blesbok already pointed out, there is a HUGE difference between a "research experience" and a real research background. For example, a lot of med students do research for the summer between their first and second years. It's two months in the lab, they do a poster, there you go. That's a research experience. Some people write up a case report. Another research experience. These aren't exactly PhD dissertation projects.
I agree with you that the personal attributes needed to do well in residency are similar to the characteristics that med school adcomms look for in applicants, but we're not talking about the personal attributes of individual applicants. You said earlier that doing research would make you more competitive for residencies. Again, the purpose of most residencies is to train CLINICIANS, not scientists. The last NMRP data I saw concluded that having a grad degree is not helpful for the match. There are some specialties and specific residency programs that are looking to train academically-oriented physicians, but most want to train full-time clinicians who don't do research. Unless you're applying for residency at a program that wants to train academic physicians, they are going to care a lot more about your clinical grades and LORs than they are about your research projects.