Thank you Bob (and Dom)!
Rather than saying smooth or linear is, or is not best, I try to not think of techniques as being good or bad, but instead judge them on whether or not they produce the results you want for a particular application.
I guess it also comes down to what you are focusing on; the action, or the result. Clearly the act of just turning the spoon is ‘smoother’ than that of tapping your wrist. But the ‘results’ from tapping your wrist are ‘smoother’ (or perhaps that’s more controllable).
Clearly Senna’s throttle technique, while unorthodox, worked well for him throughout his whole driving career, but the question is why. We unfortunately we can never know for sure, and I would not be surprised at all if even Senna did not know how/why it worked. Many times the elite are not consciously privy to their intuitive genius. 
Anyway, I have watched a number of the ‘Senna’s Throttle Technique’ videos, and felt the topic could be looked at from another perspective, so I’ve included my $0.02 on the subject below:
Even early in his career, Senna appeared to be a master of managing rotation and trajectory, so I would not be surprised if the throttle technique was a critical tool that allowed him to do this (instead of something related to big HP turbo motor management).
So, I wonder if Senna’s technique might have had three interconnected objectives:
- Managing the way tire loads built during the corner enter phase, but after the driver input that initiated the direction change.
with the objective of
- Controlling where/when and how quickly the car rotates in yaw around it’s center of mass (e.g. transitions from entering the corner to exiting the corner)
with the objective of
- Orienting the car in the most advantages trajectory, relative to the line being driven, for the exit phase of the turn. Whether the ‘most advantages’ trajectory is slightly oversteering, neutral (tangent to the line), or slightly understeering is, of course, dependent on many factors. However, in general, and especially when under hard acceleration, it seems that a trajectory that is neutral to very slightly understeering, allows the acceleration force to efficiently and coherently accelerate the vehicle by pushing through its center of mass. If the vehicle is oriented in even a slightly oversteering attitude, the acceleration forces appear to split: with most of the force accelerating the vehicle forward, while the remaining forces try to accelerate the car in yaw, which can cause excessive rear slip angles (scrub), the need to roll out of the throttle, or a spin.
This YouTube video is low quality (and has an annoying watermark), but if you look closely, you can see some differences between his technique and the other FF2000 competitors; who he completely spanks BTW.
Some of the more illustrative bits of the video occur at these time stamps:
3:00 – 3:23 – Observe how well balanced his car stays as he makes an outside pass, and again when entering the next turn. Also in that next turn notice how his whole car seems to drift (instead of one end or the other). NOTE: It’s easier to see this stuff if you want at 1/2 speed.
4:00 – 4:03 – Senna takes a much wider entry line than the other front runners. It seems like a ‘round’ entry like that would induce a lot (perhaps too much) yaw force, but if you look at this corner from the other camera angle (before or after this shot), he has the car on rails… perhaps this was an example of an ‘anti-yaw’ throttle technique at work?
4:10 – 4:20 – Again, notice how in the entry phase of the turn his whole car is in a balanced drift, and then when he bleeds off enough speed, he points it to the apex and goes.
7:12 – 7:22 – Notice how carefully he manages his slip angles and rotation.
In general, the other drivers in the video seem to have a segmented corner approach; they drive the front tire in, rotate, and drive the rear tire out. But Senna seems to have a more ‘holistic’ or ‘rounder’ (full corner) approach in that it seems like after the initial change of direction, he seems to induce, and maintain, more balanced loads between his outside front and rear tires.
When entering a corner, after the initial direction change, there is an increasingly intense wave of energy flowing longitudinally to the front of the car. If Senna is, in fact, using his throttle technique in this video, then I wonder if he is really using the throttle to send small ‘counter-waves’ that could both influence the momentum of the longitudinal wave’s intensity buildup (which I presume would also have an advantageous impact on the yaw rate ramp up), and limit the load transfer to the outside front tire, which would effectively leave more load on the outside rear tire. Both of these would seem to make it easier to ‘check’ initial yaw momentum, and maintain a trajectory that is relatively tangent to the line being driven throughout the turn. So perhaps he was a biological active suspension system. 
Also, I would not be surprised at all if the frequency or tempo of the throttle applications was directly related to the anticipated longitudinal and yaw forces for the turn. For example, a higher frequency for a tight turn with heavy braking, which would induce high degrees of load transfer and yaw, and lower frequency for higher radius turns where the forces build more gradually.