See my previous post, where I've 'suttly' shifted my argument to efficiency.
Ahh
I'll do some digging around to try and find it, but I have some graphs that plot KW from the battery with wheel horsepower using a vbox. They start out being perfectly parallel but as speed increases, they start to diverge. I surmised that that wind resistance was factoring into the horsepower measured at the wheels declining against the power at the battery. However, when I did a cost down from something like 100 MPH, the calculation was off. It was then that I discovered that the formula Racelogic uses to calculate horsepower had a typo in it which everyone here confirmed when I posted it.
After fixing that, the KW curve from the battery matched the wheel hp (minus a fixed drivetrain loss). I haven't measured it directly, but it's clear from other vbo files folks have sent me of their 1/4 mile runs, that motor efficiency is nearly flat to about 100 MPH and then starts dropping like a rock after 110 MPH.
Tesla really picked the right ratio based on their motor efficiency. Shorter gearing would have sacrificed too much efficiency at freeway+ speeds. Taller gearing would have resulted in more torque at the motor for every corresponding wheel RPM requiring bigger motor bearings/journals, motor shaft, and beefier diff.