But there WERE more than two points; surely you know that. Q4 was not the first quarter that Model 3s were produced. Q3 of 2017 is. So that's three points.
To use your example:
T-1: 2 vehicles
T0: 10 vehicles
T1: 40 vehicles
That is exponential. If it was only a linear ramp-up, that'd imply negative 20 Model 3s were produced in Quarter 3 of 2017.
You might think this is being pedantic, or that "exponential" is just a buzzword, but it's not: the point about exponential ramp-up is that if you're off by a factor of, say, 4x in production, that may mean you're only behind a few months. A relatively small time change corresponds to a very large difference in production.
Exponentials don't last forever; eventually a plateau is reached. But as you're bringing automation/production up, you're still in the exponential phase (as the data shows), and so small delays can make a big difference in production without telling you much about how many you'll produce once production is fully up and running.
To use your example:
T-1: 2 vehicles
T0: 10 vehicles
T1: 40 vehicles
That is exponential. If it was only a linear ramp-up, that'd imply negative 20 Model 3s were produced in Quarter 3 of 2017.
You might think this is being pedantic, or that "exponential" is just a buzzword, but it's not: the point about exponential ramp-up is that if you're off by a factor of, say, 4x in production, that may mean you're only behind a few months. A relatively small time change corresponds to a very large difference in production.
Exponentials don't last forever; eventually a plateau is reached. But as you're bringing automation/production up, you're still in the exponential phase (as the data shows), and so small delays can make a big difference in production without telling you much about how many you'll produce once production is fully up and running.