estimating FTP is pretty straightforward in my world because:
This. Doing testing twice in every block (TR style) is pretty pointless. I can do FTP test 2-3 times a year max when I feel that something has changed in plus. Basically rest is going by feel and when things become significantly easier I plan the test - usually the change is around 10W, less than that - I do not care, I just simply go harder doing workouts if needed 
“simply go harder” depends on the nature of the workout. Sometimes its adding intervals. Sometimes its extending interval time. And sometimes its pushing power. If you understand the goals of a block and workouts, it is easy to flex up and down based on your fitness and recovery level.
If you started at X watts for 30 min, then built up to X watts for 70 min. What would then happen if you then went back and did the 30 min test? Would it still be the same “X” watts, or will this have increased too?
I think this post gets to the heart of the whole thing: you’ve figured out what works for you. From your posts on the forum, seems like it took some time to get there. I think a lot of the fights over FTP and testing come from people arguing over canned protocols and test frequency. I started TR about a year ago, it’s my first experience w/ structure. It took me a while to realize that the canned programs were not ideal for me. So I changed some things, adapted, if you will.
Thanks for this back and forth. You might have convinced me to work up to that 1 hr TT.
Figuring out how to perform outside intervals by HR (or ‘all-out’ for short intervals), is one of the first things I did after starting to ride a road bike in 2016. Mostly CTS (Strava plans) and the Time Crunched Book, with some Friel tossed in (30 minute test). Then GCN inspired me to try doing hour-of-power which we’ve discussed.
Hope that helps.
FTP is whatever you want it to be. If you are using it as a performance benchmark, then it just needs to be a) consistent and b) relevant to your actual goals.
I’ve come to realise that FTP (in typical ramp test or 20 min protocols) is not actually very important for my own personal goals, which are 7+ hour Gran Fondos. You could have two guys with identical FTP benchmarks who are worlds apart after 5 hours on the road. So I’ve stopped focusing on FTP and started thinking more about the longer game, pushing up my IF over longer distances without necessarily seeing any gain in FTP. Does a ramp test really give you much idea about your true endurance beyond a very basic level? - I doubt it. An hour-of-power on the other hand is probably quite a reasonable indicator. Two hours even better.
Anyway that’s my random input!
It generally becomes an asymptote. As I move from 12 hours to 24 hours to 36 hours to 48 to 60 to 72 hours and beyond; my average speed tends to stabilise no matter how much longer I ride for. Of course over these time periods the riding is not continuous. You will stop to eat, go toilet etc. You will sleep for up to 3-4 hours every 24. I’m sure if you didn’t sleep then there’d be a marked deterioration.
Of course over these extended periods other considerations take over from your basic ability to put a certain power number.
Definitely, most people don’t understand what a 12 hour or 24 hour power curve looks like! 
I’ve seen many of them and they do tend to converge asymptotically but anybody who thinks they converge on Critical Power, LT2, MLSS, will be in for a BIG surprise.
Most people can’t comprehend how much their power deteriorates after 24 hours of riding.
Without scanning back, I’m sure there was discussion about why there is nothing magical about 1 hour. Just in case, Kolie Moore has a good blog post on TrainingPeaks.
Speaking of Kolie, just listened to Empirical Cycling’s “200k: The Best and Worst Workouts” and although it is not a workout, the last of Kolie’s worst list was the ramp test. If you want to only listen to the ramp test discussion, skip ahead to minute 56:
FWIW I agree with everything said.
I had it on in the background but my ears pricked up at their discussion over the gradient of the ramp and I would agree with them there. When I first started with TR my FTP from a 20 mins test was quite high for me and the ramp test was only 5w lower but as the ramp got shallower and shallower the FTP number decreased but my TTs, etc were getting faster so I doubt my FTP was actually decreasing. 
Well, the GXT has been around for, what, about a century now? And I think Kolie stipulated that if you are trying to determine VO2(max/peak) it’s the shiznit. I’ll add that it’s good at finding MAP. We both agree it’s not so hot at finding FTP/Lactate Threshold/Whatever.
However, w/in the scope of TR I’m going to disagree with Kolie. Kolie knows a lot about a lot of things, but he doesn’t know as much about TR ramp test as I do. He might be letting some personal bias cloud his judgement in this area. Everybody should listen to the podcast and see if they think that portion of the conversation sounded completely without bias.
As I said, I know stuff about TR ramp test that Kolie doesn’t. It’s hard to deny that I’ve been a critic of the ramp test because I’ve always felt bad for those riders who had FTP that was less than 75% of MAP. Well, guess what? TR made big changes that really helped address the ramp test’s disservice to those riders.
How do I know? Well, here’s where I know stuff that Kolie doesn’t: I tracked TR compliance as Adaptive Training ramped up. As far as I can tell compliance is WAY UP since AT really rolled out. Maybe TR will confirm, I don’t know, but I say compliance has really, really improved with the introduction of AT. So all those riders that were doing Kaweah at a VO2Max intensity a few days after they took a ramp test are being brough back into the fold with some basic adaptations.
So is TR Ramp Test, stand alone, a great FTP test? No. It has some flaws. Is TR Ramp Test in conjunction with AT a great FTP test? IT IS A LOT BETTER! After all the criticism I’ve given TR about the ramp test I owe them some kudos when they make improvements and I think AT has been a first order improvement in this regard.
Is AT really helping peoples “FTP” estimate using the ramp test though? I’m one of the lucky ones, who when does a ramp test, and immediately after gets Kaweah (edit: meant Lamarck!) it feels about right.
What AT does now, based on my understanding, it it just gives you more achievable workouts to start, but still uses the same potentially over (or under) estimated FTP. It almost seems like AT just fools you for longer, and you hopefully get stronger and catch up. But what if you stall at your progression because you are still using an FTP that is too high.
I guess what I’m saying, is AT needs to also learn what your actual FTP is instead of just giving easier or harder workouts based on a potentially poorly estimated ftp.
Nope.
Thanks for clarifying. Because what you wrote before the s not true. Not worth parsing and rewriting, anyone can adapt around a questionable estimate. That’s not the point.
Do you mean Lamarck? The FTP double checker?
When you say Kaweah feels about right, what does that mean?
I’m not picking holes though I know it might look that way, I’m just clarifying my understanding in case it can improve how I do things.
Sorry, I meant Lamarck!
If you’re doing 4x10 at “FTP”. This shouldn’t actually be super hard, but also not easy.
Power or work rate in Kj’s/second at MLSS=FTP which can be anywhere between 30-70 minutes (TTE). Finding FTP is used to set training zones. Training zones are used to ensures proper physiologic adaptation to raise power and increase TTE.
The TR ramp, the Coggan/Allen 20 minute protocol and even an hour attempt don’t necessarily give you power at MLSS. They might, but, there is no way to know. Does it matter? IDK and I’ll let you decide. I do think as a race oriented rider Kolie’s progressions are better method to estimate FTP simple because they attempt to or cross that TTE inflection point. Still an estimate but, a closer estimate IME/O.
To be clear, I do think increasing TR ramp power and/or 20 minute power is going in the right direction, but I feel for race craft performance, they could be a little narrow. A little short sighted perhaps. AT makes compliance more which could support how poorly the ramp test estimates FTP. I mean what’s the point of better compliance by itself other than TR marketing compliance rates have increased? That seems troubling to me.
Certainly, listening to people talk about FTP on group rides and even races; reading about people post here and other places; it’s clear to me too many are only concerned about what power they can do for a certain test as if that is all important. It’s as if it defines them as a cyclist. I suppose if that makes you feel good that’s cool. More power to you. If all you do is ride in your basement that’s probably a great metric. Events outside that are super dynamic due to climbs, wind, other riders, weather, etc…you better hope your FTP was close. If not, you just spent a lot of time training incorrectly (suboptimal) to get dropped.
Agreed. Just trying to illustrate the quality of work done rather than just worrying about a higher number drives performance.
Yes, same feeling here. I’ve heard the podcast talk about things like removing long rides from the weekend to increase compliance. Is that really a good thing? Maybe for some. 
One of the TR product manager job postings in 2019 included a question that started out “Say you were given the goal of reducing churn rate of users in their first 30 days of signing up by 20%. …" Well the first thing is to NOT assign some of the workouts my coach gives me because there are BORING 
despite being effective.
