It could be useful for the city. In Australia, you can buy the used postie bikes (Honda CT110) that get 1l/100km highway, which indicates that when driven with FE in mind they can probably do better than that in a combined cycle.
From what I've seen of motorcycle fairings, I'm not very impressed with their attempts at drag reduction. Consider how large an area the wheels are wrt the whole frontal area. Compare that to a car, where the wheels account for about 20%. I would not be surprised if they account for 50% of the drag.
I'd fair most of the front of the wheel like an aircraft wheel pant, and attempt to do the rear. After that a bubble for the front. In my drawing I should have made the top of the wind shield closer to horizontal, oh well.
[IMG]
[/IMG]
Revised:
The point of that is a good reduction in drag coefficient, while keeping the addition to the side area increase to a minimum (to reduce buffeting from crosswinds). Note that the wheel fairings will hardly hurt in sidewinds because spokes have a huge drag coefficient anyway, a flat plate would hardly be any worse. You could also stretch lycra from the front and low part of the fairing to the back, streamlining a large portion of the frontal area without increasing the side area.
Of course, the other option would be to go the whole hog with basically a modified HPV design, see
here. Better drag coefficient from the front, at the expense of more buffeting from the crosswinds.
With a 105cc motor the fuel used in the glide portion of a pulse and glide would be negligible. But I'm not sure if you'd achieve a better CdA than the vw 1 litre car with those modifications.