Wow, that is a nice car. I would have liked to have found ready access to a small TDI diesel for this conversion, in fact that was my first choice before being lured to the carbureted engine side where modifications to the air intake and spark ignition would have allowed for gaseous ammonia fuel (which of course wasn't delivered as promised). I also had trouble finding a diesel that would fit in the trunk. Anyway, let me try to get to the point of your question.
This VW concept car is actually a series-parallel hybrid (Pruis style). The motor and engine may assist one another, but the car can be propelled by the electric motor alone. It's distinct from a Prius in that it (apparently) may achieve significant range on electric-only propulsion at speeds greater than the Prius' maximum electric-only speed. This ability to move under electric-only power without the engine running at all is what can cause hyperinflated "miles per gallon" efficiencies to be reported for these types of cars. For example, if I were to take the Chevy Volt with its 40 mile electric-only range through ten consecutive runs on the UDDS driving schedule, it would travel 40 miles without a drop of gasoline, for infinite miles per gallon. The old EPA methodology was to use a distance of 100 miles of mixed city and highway driving (40% city, 60% highway if I recall correctly) to evaluate economy. You may remember the 230 mi/gal claim that GM made a while back. To make up for this, the EPA has split the ratings for the Volt into "all electric" equivalent miles per gallon and a gasoline-driven mileage too. The electric to gallon of gasoline equivalent (gge) is about 33.7 kWh/gge. So now the Volt is rated at 93 miles per equivalent miles per gallon while going electric-only, and 37 mi/gal once the engine turns on. Those are mixed city/highway numbers, too. I'd like to see a similar analysis for the VW concept. It looks very slick (low drag) and is extremely light. I'm sure the true efficiency number is well over 100 mpg on the high way at a not-too-excessive speed.
So what I'm saying is that this kind of comparison is definitely not apples-to-apples, and gets very confusing very quickly. Probably the best way to do the fuel economy calculation is to say that, "driving this car for a week puts this much CO2 in the air, and/or costs you this much money." I like the money angle.
Lets say that my next door neighbor and I are coworkers, and we don't carpool. He has a Volt, I have my converted EV1. We each drive exactly 40 miles to and from work five days a week. He never needs to fill up the tank, and the EPA lists his cost per mile at $0.04/mi for a total of $8 per week. The combined city/highway mileage for the EV1 conversion comes out to 60 mpg (I think). If we pay $3.75/gallon here in Alabama, that means if I drive 200 miles a week I'll pay $12.50 for my weekly commute. The Volt wins.
Now let's say we each take a 200 mile leisurely road trip on state routes (mixed city/highway), doing the driving in one day. He gets his first 40 miles with no gas used, then starts using gasoline at 37 mi/gal. He'll pay $16.22 in gas. If I go the whole way at 60 mi/gal, I still pay my $12.50. Things look even better for the converted EV1 if it's an extra long cruise at highway speeds (my mileage at around 72, his at around 38 or so according to what I've seen published for the Volt).
I'm cutting the Volt a break here, because it also uses premium fuel - which is about a quarter dollar more expensive per gallon around here. Anyway, bottom line is that I'm pretty pleased about the numbers as they came out from a 34% (at best) efficient carbureted lawn mower engine!
As for hub motors, I think they could offer a lot of better options for a production car, if they could be made inexpensively. They were definitely not an option for this project, but I'll be watching the field to see what progress is made for their development. I will definitely keep an eye on the Active Wheel.