You prove my point: 2% variation in L-R balance. That's a 4% error off the top on total power. That's consistent with what I saw with DCRainmaker's data. If 4% power errors don't matter to Sky, then fine.
No offence to KW but I don't think he is at the level of riding pro for SKY (correct me if I am wrong
). Based on my own highly unscientific observations, this 4% total error due to L/R imbalance with the Stages assumption is what I'd expect for an enthusiastic amateur, having an off day and maybe carrying a niggling injury to one leg. I would have thought this represents a very worst case error and certainly not something you'd see in a fit and healthy Pro. Total errors due to the one side assumption are therefore pretty low. For all we know the measurement error with Stages is half that of SRM or similar and both output the same power at the end of the day with the same overall uncertainty in the final number???
I do agree there is more potential for magnifying errors with Stages but it's unlikely to bother the Pros and if SKY adopt it then you'd have to assume they've given it some thought. You'd also think that if Stages are confident in their assumption they could publish some data which would be good to look at (rather than some vaguely similar looking power outputs on their website)