Good to hear that you are also so delighted with the i3. It really would hard to go back to an ICE with all their costs, inefficiencies, inconveniences, harshness and other disadvantages. I think our efficiency with the i3, while brilliant, is not all it could be because we don’t routinely use eco mode – love the responsive accelerator too much I guess. Nonetheless, in comparison to an ICE it is brilliant.
Regarding the highway driving mentioned, we don't do much of it, but when I have (e.g. ~100km each way Brisbane to Gold Coast every day for a week at 110km/h for large sections) I actually get better efficiency (e.g. 11.8kWh/100km) than around town. The following is my understanding of why but please correct me anyone, if I have this wrong.
While there is so much in the EV literature about the exponential-like decrease of efficiency with speed due to wind resistance, the assumption made is that high speed driving is less efficient. Now this is absolutely true of ICE and BEV, IF you are comparing constant high speed driving with constant low speed driving. i.e. all other variables are kept CONSTANT. In both cases the slower speed is more efficient, by common laws of physics I guess. However, urban driving is always changing and the need to be repetitively accelerating a 1300kg + weight (often from stand still) in urban driving will usually use far more energy per km travelled, than the gains from less wind resistance at an overall slower speed. The point is, as the ICE is so abysmally inefficient in city driving the variation between city and highway speed is very apparent, where as with an EV with very good efficiency in the city there is less comparable difference between city and highway driving. This means that sometimes city driving maybe more efficient than on the highway and sometimes not, as all sorts simple variables (like head wind) can easily tip the scales to one or the other.