Miles per gallon answers a question no one asks - you don't say "I have a gallon of gas, how far can I go?", you ask "I am going 40 miles, how much gas will it

Miles per gallon answers a question no one asks - you don't say "I have a gallon of gas, how far can I go?", you ask "I am going 40 miles, how much gas will it take?" MPG comparisons are based on two vehicles using the same amount of gas and seeing which will go farther, which clearly isn't how people think. Gal/100mi are based on cars driving the same distance and seeing which uses less gas, which is more applicable to real habits. There's also an interesting study done by some guys at Duke (summarized at http://bit.ly/mYpFV http://bit.ly/mYpFV) showing how GPM makes it much easier for people to grasp fuel savings than MPG does. Let's say I have two cars, each of which is driven 10K miles per year - not unreasonable for a suburban household. One of my cars gets 35MPG, and one gets 20MPG. I can replace the former with a 50MPG car, or the latter with a 30MPG - for differences of 15 and 10MPG respectively. most people would pick the former. But in that case, the 15MPG increase would only save 85gal, whereas the 10MPG increase would save 160gal. If you flip the numbers, however, and consider the choices as Gallons per 100mi, it becomes much clearer: MPG Gal/100mi Savings (gal/100mi) Savings (MPG) 35 2.85 50 2.00 0.85 gal/100mi 15 20 5.00 30 3.33 1.66 gal/100mi 10 The 20 -> 30MPG switch is empirically a better deal, all else equal (since we don't expect our driving habits to change with the new car). Gal/100mi sure makes it easier to see that, though.