I was only adding the data point, that even with just minimal braking required, the regen proved itself superior to freewheeling with empirical data. I was not arguing the original statement, in fact, I explicitly agreed "in theory". I guess what I'm getting at is, is there a scenario, that can be performed in the real-world, in realistic conditions, where turning off the bike and freewheeling would actually give you better range? You would have to know, ahead of time, that you will not need the brakes or throttle. It seems that is a small corner case, and one for which I have not seen empirical evidence to support. Again, I agree "in theory" a freewheeling bike should use less energy in a very narrow set of circumstances. I'm just not sure all those criteria will ever be met in the "real world". Basically if we go for a ride, and you turn your regen on/off all you want, and if I am just pacing you in eco mode, I will still probably go further.
Ultimately, I know we are all interested in getting the most efficiency out of our vehicles, and I am just as curious as the rest of you as to what the best possible drive profile and strategy is. My best guess at this point, as supported by all the data I have access to, is no-regen-freewheeling will provide little, if any, perceptible increase in range in any realistic scenario. While effective management of regen can provide up to 6-8% increase in range in an optimum duty cycle. Possibly clutching the motor/drivetrain out of the loop entirely would make such a strategy more effective, but I think we are reaching the point of diminishing returns here.
Just thinking out loud.