-
this would really be an intriguing possibility. i have been thinking about writing some software which attempts to make power meters a bit more comparable and easier to do historical reviews of your data and that'd be invaluable. you could maybe smooth the offset over time to adjust data, model l/r imbalance and apply that to single sided power meters, use tests of pms to estimate variance of pms by manufacturer, etc. plus you could estimate uncertainty for all of this so you could see if your data were significantly changing
No idea, but there's a boat load of stuff stored in a .fit file that isn't documented. If it's a .tcx file then I think the answer is no.