delta-Cd/ delta MPG rules of thumb
 

The relationships are derived from:
' Tractive Energy Based Formulae for the Impact of Aerodynamics on Fuel Economy Over the EPA-Driving Schedules,' SAE Paper # 830304, Detroit, Michigan, U.S.A., Dr. Gino Sovran, General Motors Research Laboratory, and SAE Paper # 810184, Detroit, Michigan, 1981, Sovran.

Thanks for the reference. I'd imagine they'd be quite good relationships for cars of that era, ie cars of about 40 years ago.

imagine
 

I hear what you're saying. And thanks to higher fuel prices and CAFE standards, drag coefficients have seen slow, incremental reductions ( 'without risk to sales' as Hucho would say ) over these 40-years.
That said,at the time of publication, R.G.S. White had already delivered a Cd 0.245 'recipe' car in 1969. Dr. Alberto Morelli and Pininfarina had given us a 'production-ready' Cd 0.204 production car by 1978. The same year, Mercedes-Benz demonstrated the C-111 III, with a Cd ranging from 0.237,to Cd 0.178, depending upon the degree of boat-tailing. Volkswagen had gone to Cd 0.15 with their ARVW. The 1980 VW 2000 came in at Cd 0.25, as a producible design.We were on the cusp of lower drag.By 1990, Sergio Pininfarina commented that Cd 0.25 could be achieved with '... shapes to which the public is accustomed today....' In 1981, Ford had already ventured into pickup truck aerodynamics.
Today, some of the CUVs,SUVs,and Pickups would better represent coefficients of 1981.
Today,as back then, you'd want to investigate the change in road horsepower,given a particular drag reduction,and calculate the delta-mpg, based upon the single variable,at a fixed BSFC or BSFC-e.