Why look to ten year trends? That graph has the extreme case of buying ten year old cars and keeping them for five years - meaning with that strategy you are at

Why look to ten year trends? That graph has the extreme case of buying ten year old cars and keeping them for five years - meaning with that strategy you are at times driving a car which is fifteen years behind the curve of fuel efficiency development. Not to mention the effect that 15 years of engine wear has on the efficiency of the specific car. To get specific: a 2017 Ford F-150 (to stick with the absurdly parochial example used here) gets 19-26mpg; a 2002 F-150 got 17-21 - that may not look like much but is a 12-25% improvement, not a completely insignificant difference. A 1987 f150 got 10-12mpg, so it's clearly a continuing trend, too.