Originally Posted by
colnago62
I think Coggen in his book talks about using a riders best 20 minute average during a ride to determine ftp. The most important thing is to use the same system consistently. Eventually a rider will figure out what they have to do to get their best numbers.
It's arbitrary because without a semblance of a protocol, Garmin misses many confounding factors like the fact that I wasn't riding to exhaustion. Those 20 minutes were during a climb that happened to take me about that long, but with 2 more hours of riding ahead. Another indication of its arbitrariness: none of my outdoor rides are what I would consider useful for estimating FTP, and Garmin only ever managed to find that one ride to produce an estimate.