Since I want Tesla to succeed (as should anyone who is considering buying their cars), let's look at what they need to do.
1) They need to project the image that Supercharging is a viable substitute for gassing up at the X-mart.
2) They need to project the image that a Tesla car is a full-use car, capable of long road trips.
3) They want to reduce the number of cars on the road, and enable ride-sharing.
4) They need to not go bankrupt.
5) They want to produce a car that will last for 500k to 1M miles.
So,I (hypothetical person) buy a Model 3, as an investment, I am going to rent it out on the Tesla network for its entire life. I live in a state that salts roads, and fairly close to a Supercharger. Salty roads means that my investment will die due to corrosion, not mechanical or battery failure. This implies that maximum ROI comes from keeping the car as busy as possible. I paid $52,000 for the car (long range battery and full self driving). 500,000 miles requires 121 GWh (no, really, I had no idea before I did the math). If I am paying for that, it comes to $24,200 (at my state's rate). If Tesla is, it costs (guessing $0.06/kWh) to $7,250. 25% Margin on the purchase price works out to $13,000. So, more than half of that going to pay for the electricity of 'free' supercharging. My costs per mile are either $0.15 versus $0.10 (there is a viable business model). This is why Tesla can't give free supercharging for life (violates #4). (The fact that they are still offering it on the Model S and X indicates, to me, that they don't expect to sell many of those into the network market.)
#2 is addressed by allowing *some* free supercharging. New owners might make a road trip just to 'take advantage' of that 'free' fuel. Once they see how simple and easy Supercharging is, they will pass that on to others thinking about buying a Model 3. Once they exceed the 1000 kWh, it will be just a small charge on their credit card, not worth noticing. I don't see how 400 kWh/year does this better. No one much is going to be keeping careful track of their yearly usage. It has virtually no impact on the Tesla Network cars.
I think Tesla may have hit on the best plan for them (given their aspirations). I do wish they had gotten there with fewer confusing changes in policy.
Thank you kindly.