[aerohead]

I've re-visited Figures 4.63- 4.65 in Hucho's 2nd-Edition, and something he said stood out that I thought I'd share:
The context is Buchheim et al.' aero development of the Audi 100-III.
The team worked through wind tunnel investigations of shape optimization for the backlight, boot height, and boot length.
Each component was tweaked until they reached 'saturation', after which, no further improvements were derived from additional change to an angle, height, or length.
Hucho wrote to the effect that, ' The possibilities offered by this measure have not been fully exploited due to the reduction of the view to the rear as well as the appearance of the vehicle from the rear.'
My interpretation of the comment is that, what ended up going into production was short of the aerodynamic 'ultimate', and the decision not to explore lower drag had to do with rearward vision limitation, aesthetics, or constraints governed by designer/ stylist.
We know, for instance, because Buchheim has told us elsewhere that, the 1st-gen Golf / Rabbit, could have had Cd 0.34, instead of Cd 0.42 in 1975, however, VW was under contractual agreement with designer, Giorgetto Giugiaro, to construct the Golf essentially as he had 'drawn' it.
FIAT, in their SAE Paper 860212, also explored Cd as a function of rear slope angle and configuration. Their lowest drag occurred with a notchback with such a 'flattened' backlight angle and high boot that, one would wonder how any driver could ever see out through the rearview mirror. Again, a situation where considerations, other than aerodynamics, dictates the final design.
As we move more and more towards 'synthetic' vision, these issues may play a lesser role in a new vehicle's design specification.