I've known already for a while that native speaker children can mess up W and R. Some kids say wed instead of red or Wichard instead of Richard.
There is also an American cartoon character who messes up W and R. He, Elmer Fudd, pronounces rabbit as wabbit.
When I teach eigonodo, some people mispronounce R as W and I think this is a good kind of error. Using our method, Japanese people are making errors that native speaker children make. This shows how true eigonodo method is.
W and R are pronounced/resonated deeply at the bottom area of the neck. For W, this probably is sufficient. For R, you need to add vibration or some kind of rambling sound at the bottom floor of the neck. This trick worked very well when Jeana taught a deaf American girl. Her Mom said that at the deaf children's program they teach W as a transition sound between U and A. They make you practice U-A transition many times to get at W. Such an approach helps approximate [W], but does not get at the true W sound.
To this day, I don't know why deaf educators don't come to us. I subscribe to a Japanese deaf educators' emailing list (they are specialized in English education for deaf kids). They completely ignore my messages. This is a pity.