I agree with
burnthesheep with respect to accuracy of power meters. I aggregate power results from different bikes, road, track, time trial and tandem into software as I think most racers do that use power on the bike. I do not like the idea of one power meter (or more) being off no matter how consistent it is.
For me, 5 watts of FTP is a lot. In fact 1 watt is a lot. If a PM is already reading 30 watts low, 5 watts is even more difficult to make because it is a bigger percentage of a smaller number.
And for the big engine guys that have 350 watt FTPs, 1 watt is still a lot. If you think it isn’t, try to raise your FTP 1 watt per week. I mean really, what is 1 more watt on a 350 watt basis? In three months, you are up 12 watts. In a year, you are up 52 watts.
Power meters have to be accurate on both a relative and absolute basis to be useful and generate the correct amount of feedback to athletes.