This sounds very plausible. This’ll be that “brittle fitness” that people talk about, right? It seems that there’s a distribution - some people can build fitness this way and be fine; others not so much.
Coggan has changed his tune over the years, which makes citing him problematic. It is so easy to get lost in academic debates on what is FTP and how to test it properly.
Moreover, this IMHO forgets the purpose of FTP in the context of structured training: it defines your power zones in structured training, and setting them correctly is particularly important for threshold workouts. Over/unders should really dip below and above your power at MLSS (see what I did there 
) to do what they are designed to do.
And no matter how you arrive at your FTP, you should validate it with (threshold) workouts (“Trust, but verify.”).
When you are fresh, then no. But there are situations where e. g. you are unable to complete workouts because of fatigue. That happens to me when I have had too much life stress and did three interval workouts per week. I’d often struggle with an otherwise doable interval workout. So context matters.
You, sir, are evil! 
I don’t buy DC Rainmakers assertion that Coggan has tried to change the definition of FTP. I’ve listened to Coggan over the last ten years since I got into structured training and he’s been pretty consistent. Yes, Hunter Allen invented the 20minute test with the 5 minute blowout and Coggan’s name is also on that.
The “problem” is that people want exact but FTP is a proxy for MLSS which is not an exact time frame. Coggan has said 40ktt or about an hour. That’s because 40k TTs back then were about an hour. Now, elite level ITTs are shorter with aero time trial bikes and all the gear.
People should go read Determinants of endurance in well-trained cyclists by Coyle and Coggan. They talk a lot about TTE in that paper. The well trained group’s TTE ranged from 50 to 70 minutes and the mean was 60 minutes. Now you know where he came up with “about an hour”.
My original point still stands. If you struggle with 10 minutes at threshold and you clearly couldn’t ride anywhere close to that pace for an hour, then it’s not your threshold. If you can’t maintain the steady state for an extended period of time, it isn’t your threshold.
And the solution is simple, back off a bit on the power until the RPE is correct.
You don’t have to…Coggan has said repeatedly on this very forum that FTP is not “about an hour”. Just search his posts.
I have no need to. I know what FTP is. I just re-listened to that recent Empirical podcast where he talks about FTP and what it means. It’s pretty clear to me. It’s MLSS. You find that wattage where if you go a bit above you are in the severe domain and you time at that wattage is numbered. You back down just below and you found your FTP.
Honestly, all the naysayers should just do a Kolie Moore FTP test. Then you will see the light.
The other thing is I don’t give a crap what Coggan posted on Slowtwitch 20 years ago.
Well, I never said anything about Slowtwitch, so… 
If you choose to ignore what he himself has said on this forum, that is your choice.
It has nothing to do with the topic. If you struggle with a 10 minute FTP interval, you aren’t at FTP. I’m sorry if that is a blow to people’s FTP ego or they love automatic ftp detection or ramp tests or because they hate any test and long intervals.
Maybe people just practice long intervals more at the correct RPE?
This is hugely key for indoor workouts (especially with longer intervals). You generate a significant amount of heat while putting out higher wattages and even if it’s cool without moving air you don’t actually lose that much of it. You sorta form a little microclimate around you where you heat up the air directly around you (and raise the humidity). But with moving air you (1) replace that heated air with cooler air from the rest of the room and (2) introduce convective cooling with is significantly better at cooling you down (think of wind chill in the winter. 0C isn’t too bad but you add even slight amounts of wind and it quickly gets super cold).
IMO, most people who have significantly different power output indoors and out usually it comes down to lack of adequate cooling.
It is literally part of the topic being discussed now and that you are responding to / advocating.
Carry on…
I don’t think there is much to “buy”, just have a look at what Coggan wrote over the years and what he has written here.
Shorter tests and ballpark conversion factors were mentioned in the earliest document by Coggan (authored Coggan alone) on that topic I am aware of, from the year 2003 and, of course, in his book with Allen. In a post here, he disavowed the 20-minute test, claiming this was all Allen’s doing and I should know that he hadn’t written that part of the book. (This is not how coauthorship works in science.) He could have said “I have change my mind in light of new data.” He reiterated it here that the proper definition of FTP was best power at 60 minutes. I can link to specific posts if you’d like, but this has been discussed to death.
His behavior on this forum was so bad, he got banned, lots of gaslighting.
I think it is even more basic: Coggan is a scientist, and scientists work differently than practitioners (athletes and coaches). In science, even tests that are “the same” can have different protocols. So the results of a 20-minute test without a hard effort before the test results are not directly comparable to those of a 20-minute test protocol without that bit. How hard should that effort be? (That question was raised by Coggan himself in a post here.) Ditto for different variations of the ramp test.
The results are in any case just estimates of power at MLSS, since an athlete with a TTE of 50 minutes at MLSS power will get a lower result with a 60-minute test (compared to power at MLSS) than one who can manage 70 minutes.
Still, in science you need a single protocol to have consistent data. Without consistency, you cannot compare data.
The needs of practitioners are very different: they need to know the best available estimate of their or their athlete’s power at MLSS to set zones correctly. They are ok with applying correction factors that are specific to the individual.
That’s a post-hoc justification. Like you wrote in your post, the “hour power” characterization came from the fact that “original” FTP test was a 40k TT, and the duration is implied by the length and typical speeds. Coggan observed that average power on a 40k TT was close to power at MLSS.
Yes, but tit for tat about what coggan said here, there or on slowtwitch or 5 or 10 years ago is pointless.
If you want to do the search and find where Coggan materially changed his definition of what FTP is on this forum and post the link, please do. I don’t have time to sift through every Cog post to look for what you are talking about.
I already posted a quote from the cog where he says it’s 40-70m.
You should quote what you are actually talking about if you want to make this argument. I look at that document and the first thing I find on the subject is:
probably the easiest and most direct way of estimating a rider’s functional threshold power is therefore to simply measure their average power during a ~40 km (50-70 min) TT. This highly pragmatic approach is justified by laboratory research showing that the power a cyclist can generate for 60 min correlates very highly with, but is slightly greater than, their power at LT
This is exactly what Coggan says every time I’ve heard him speak on the subject.
And that consistent with what he has said everywhere and what is in his book. 50 to 70 minutes is from the Determinants paper. I’m not going to jump all over him if posts on some forum 40-70. It’s still ‘about an hour’.
It’s you guys want an exact minute and want him to stick with it where TTE at MLSS/FTP is variable.
You are confusing me for someone else, I wholly agree with Coggans definition of FTP.
My above post was mostly poking fun at you saying Coggan definition of FTP is not that it is MLSS and then going on to say that FTP is MLSS
50 min, 60, 70, 1000, MLSS, doesn’t really matter. If one struggles with 8-10 min at threshold they’re probably actually doing vo2. But if one is going to continue to use the TR ecosystem without doing (longer) testing, then the rep from TR has spoken above.
If you can’t do the Threshold workouts that AI prescribes, I guess think about trying to pedal harder. Answer the workout survey at the end honestly and adaptive training will hopefully fix your threshold progression levels.
It seems like I clearly misread BCMs post. In any case, I’m done with the tit for tat debate with some of the others. IMO, coggan has been consistent. And if you struggle with a 10 minute interval, your power is too high IMO.
I think much of the confusion comes from a few of his published articles that had contributions from other authors but without a distinction between what parts were written by him vs others.
100%. I would even say anyone struggling to do 30m of tiz at ftp (assuming it’s not just 1x30) has an overly optimistic FTP.
Unpopular Opinion:
Ditch FTP. Race more.
Most aren’t going to test to find their real FTP and then it changes anyway. If AIFTP doesn’t work for you train off RPE and your current PR’s like they do in distance running.
This is mentioned in the same paragraph you quoted from, taken from p. 4, bottom: