That is your single experience. Both Dan and I had cases of virtual power (well, the Zwift equivalent, based on the same power curves) giving us extremely high power estimates, so 'slightly off and fairly consistent' might apply to your case, but not to ours. I'm well aware of the single sided power measurement disadvantages, I just don't see how, in a case where you have a 40-70 watt discrepancy between virtual power and a single sided power meter, you would look at the power meter as the more likely one to be wrong. It could end up being the case of course, I just think that more often than not it will be the other way around. That's it, anyone can troubleshoot as they please.