To borrow from Joel Spolsky, what this shows you is that you can prove anything with a contrived example. More specifically, for any comparison that isn't night

To borrow from Joel Spolsky, what this shows you is that you can prove anything with a contrived example. More specifically, for any comparison that isn't night-and-day, you can almost always find a reasonable set of assumptions that leads to your preferred conclusion. > I don't know if Toyota makes such a vehicle They don't, and actually the Prius sits between the Corolla and the Camry in size and features. The cost premium over a Camry is only about $2k, and the fuel economy improvement is more dramatic. (The Camry has substantially more power, but the Prius's flat torque curve probably reduces the real-world difference). Another problem with this analysis is that on the one hand it ignores resale value, but on the other hand assumes that the owner won't keep the car more than a few years. Maybe Priuses have crap resale value, but the author doesn't state that much less back it up. But the biggest problem is that only businesses buy cars this way. People don't buy sports cars because they get them to work faster, and people don't generally buy SUVs because they have to drive off-road to get to the grocery store. People buy cars (or forego them) to tell themselves a story about what kind of person they are (and there are many comical instances where it tells an entirely different story to others). Business do buy cars this way, and if you look at those that do a lot of city driving in an area with even moderately high gas prices (say, Vancouver, with gas at US$5.15 a US gallon), you'll find that, for instance, 90% of the taxis are Priuses.