One of the key things to understand about the relationship between charging infrastructure, range, and customer acceptability is that there are a number of non-linear relationships that lead to the experience tipping between "really bad" and "really good" very suddenly.<p>First, it doesn't matter that most journeys are short and that the overwhelming majority of drivers will make between zero and two really long range trips a year (yes, even in the US). People size their long term capital purchases based on their perception of the limiting case, not the average. Consumer willingness to rent a longer range vehicle for special occasions will also depend on how convenient that experience is. Many Americans drive long-distance in the Thanksgiving to Christmas window which would require enough long range capacity to be reliably available for all those people every year. Again, we have to size for the extremes of the system. Also, car rental may be logistically challenging (and is certainly <i>currently</i> inconvenient) in most of the rural areas where people are most likely to need long distance driving regularly.<p>Second, the amount of time that a driver will need to find an available charger is very much non-linear in the number and density of chargers. We know quite a lot about this dynamic from studies of parking. Once your parking capacity in a given area is filled by more than about 80%, the time taken to find a spot very rapidly increases from nearly zero to many minutes. I'm sure there are theoretical computer science problems which are analogous. (actually would love to hear if anyone knows of any since most of the infrastructure world has historically just looked at this empirically). The same dynamic happens with chargers. Like with parking, better information provision shifts that transition point. With perfect information.<p>Third, the stress induced by charger unavailability is also not linear in likely time required to find a charger.<p>This is particularly an issue for people who do not have off-street parking but do have cars. Admittedly this is much rare in the US but it is about a third of drivers in the UK and I imagine not uncommon in other European countries. That's because people need this as their default charger essentially every day so any inconvenience is substantially multiplied. Our thinking on public on-street charging for these cases is that you're much better off spending money on a lot of slow trickle chargers and absolutely saturating a neighbourhood with them rather than buying fewer 22kW fast chargers. Two reasons for that.<p>First, 22kW is a sort of "Inconvenient Valley" for public charging in domestic areas. It's not so fast that you can stay near your car but big enough that you never have enough of them to make it ok to just leave your car plugged in. That means that effective use requires coming home, plugging in and then realising as you get ready for bed that you should move your car so one of your neighbours can also come outside and use the charger. Faster 43-50kW chargers are much better in this context because they can be "cycled" a few times in the evening before everyone goes to bed. Otherwise you might as well just save the money and get many more slow chargers fed off lamp-posts to do overnight charging.<p>Second, You can reduce charger anxiety much more effectively by having an excess of slow chargers (even if that means that a fully empty battery is not fully charged by morning) than by having a lower probability of finding a higher power charger. That's because the slow charger effectively guarantees that you won't run out. Again, this is non-linear. An EV works just the same all the way down to zero charge but driver anxiety will start ramping at 15% or so.<p>What all of this means is that if charger build-out strategies go right, we will find that the transition from inconvenient to "why did we ever worry about this?" will happen faster than you might think.