It’s a trap 
5 Likes
Until people try it they won’t believe/trust you. The whole push behind AiFTP was to get rid of the stress/fatigue of testing. Ramp tests give suboptimal data, and most other tests also give suboptimal data as well, but a test like this requires time, effort, and pacing. So if people are hesitant to do a 20 minute ramp test - they’re not going to want to do a long form test.
This comes up daily somewhere on this forum - people (myself included at times) place so much emphasis and personal worth as a rider on this number that admitting you are training at an incorrect number is a hard pill to swallow. I have seen an estimate as high as 313 from AiFTP - and my 20 minute power PR (going at FTP RPE - not an all out effort) is 305 - and my 40 minute PR is 297 (4 weeks after this) - nowhere near 313. So as much as I wanted it to be 313 - it was probably 300. After this I only started doing the long form test/validating with RPE - and now I actually train where I am supposed to. Trying to do 12 minutes at 313 crushed me - second interval destroyed me.
If people can separate ego from reality - their training will improve. (My opinion obviously, I don’t care what people actually do in their own training and how they test on a personal level)
4 Likes
I hated the 20 minute test. I found it mentally brutal to pace. This is why I tell people that a 35-45 minute KM test is actually easier than the 20 minute test. It’s a long sweet spot turning into threshold turning into above threshold interval. You quit when you’ve had enough. That last few minutes is not an all out vo2 effort to make up some ground.
It’s puzzling though why people are so keen to fiddle with their FTP 3 watts here, 5 watts there, etc. They want to spend 20-30-50 hours a month training but can’t commit to a 45 minute workout to know their exact abilities?
After a 45min KM test, I feel like I did a really good threshold workout. Even though it goes to failure, I’d rate it as hard because most of it is not hard. Only the last bit is hard and really not that hard - not painful like a 5 minute vo2max interval.
6 Likes
I agree - I hated ramp tests, and just think about the KM test as a long threshold workout that is going to give me good data.
Obviously people can have good or bad days - but it’s still the best test I’ve done (and actually feels accurate when I do the subsequent workouts)
2 Likes
Remember when Chad, et. al. talked about a 4x10 threshold workout being a good proxy for a threshold test?
To me, you don’t know you’re above threshold until you hit that 10-12 minute point of a 20+ minute effort and go… “oh no”. So I just don’t think you can do those shorter intervals and truly develop a feel for threshold. You have to dance around it in longer sessions before you can develop that feel to where you can start a little under it and roll up into it in a solid 2-3x20 session driven largely by RPE. Most people are going to overestimate it by feel at the start, occasionally wildly so.
4 Likes
How about we go back to the source, pp. 3–5:
Coggan defined FTP as measuring the lactate threshold in a field test. He proposed 3 different test protocols to determine FTP, 40k TT (which depending on your fitness could take 50 minutes or 65 minutes), 20k TT and a test involving 20-minute intervals. The latter would morph into the 20-minute test. We can quibble whether we should distinguish between the results of these protocols in practice, i. e. when we are dealing with athletes as opposed to scientific publications. But Coggan did not equate FTP with hour power, he just said that (among his selective group of test subjects) 40k TT lasts about 1 hour, and a 40k TT is one of several test protocols.
Friel similarly writes in his Training Bible:
Unfortunately, I only have the Kindle version, so I cannot give you sensible page numbers.
I still think context matters: Coggan was testing his theories on a very select group of highly trained road cyclists. So doing practice 40k TTs was probably something they did regularly. This is not true for all cyclists. Being good at TTs is a skill that needs to be honed, just have a look at the hour records and hour record attempts, the pacing skills are out of this world.
FTP test protocols attempt to measure/infer power at MLSS, and any test protocol has inherent statistical and systematic errors. But the two are not identical. It makes a difference whether you ask people to do a TTE test for their power at MLSS and their power at FTP (with a specified protocol). The article investigates the latter.
The 20-minute protocol together with the ramp test are the two most popular ways to measure FTP. So the outcome of the study is relevant to athletes doing structured training. The study also seems to provide no evidence that best-60-minute power is closer to power at MLSS than the result of a 20-minute FTP test.
I don’t think you can look at the result with the rather significant variability amongst subjects and conclude that 60-minute-best power would work better, there is no evidence for that.
1 Like
Yes, I know what I have written. Why are you quoting it back at me, and why would you trust Friel to get things correct??
There are multiple ways of estimating FTP, but using a shorter effort (e.g., 20 min) and then applying a correction factor is one that I have never endorsed. I also disagree with your assertion or implication that methods for estimating FTP need to adapted for untrained/moderately trained individuals. I say that having often coaxed 30 min of supra -steady state exercise out of such folks in lab studies.
Finally, exactly how long individuals can sustain their maximal steady state power isn’t particularly relevant, at least when it comes to estimating that intensity. What matters is the actual power that they can produce for a sufficiently prolonged duration, i.e., one long enough that the contribution of non-sustainable is minimal. IOW, you’re looking at the question backwards (as did the authors of the study you cited). The appropriate study design would be to fix the duration (or even distance, e.g., ~40 km), and see how close the average power is to maximal metabolic steady state. This in fact has been done, and (not surprisingly, at least if you understand the physiology of performance) the approach has been shown to be both accurate and precise - more so, in fact, that doing multiple tests and then applying a mathematical model.
15 Likes
I don’t know, in my experience, 4x10 minutes at FTP works great to validate FTP test results.
It takes a bit of experience, but the first two intervals determine whether I am in the right ball park. The last two intervals are for fine tuning. I use resistance mode (as I do in all non-Z2 workouts), which lets me vary the power. So I don’t have to stick to numbers if I feel like I need to increase or decrease my power targets a bit. If I basically feel nothing in my legs, I’m too low. If lactate accumulates too quickly, my FTP is set slightly too high.
The nice thing is that with an FTP that is in the ballpark (say, ± 2 %), you can finish the workout and incur not much mental or physical fatigue.
1 Like
In your book “Training and Racing with a Power Meter” together with Hunter Allen “you” have proposed a 20-minute FTP test protocol that carries both of your names, which is widely used in the scientific literature.
I have not said that. I have just said given the evidence we have that 60-minute best power is likely not a good proxy for power at MLSS for untrained or moderately trained athletes.
no offense, but if I listened to people like you or insisted those studies were ‘true for everyone’ then I’d have never hit the ability to ride 50-70 minutes at threshold. At fifty six and a training age of 3+ years if you include spin classes. Even on my spin classes I can show you 40-50 minutes at threshold (by heart rate). /walking away, shaking my head
4 Likes
Hunter is the lead author of the book we co-wrote, not me. Furthermore, anyone who believes that I am responsible for, or have even endorsed, his 20 min test has apparently been living under a rock. (And yes, that criticism applies to all the people doing studies using his test and publishing in the peer-reviewed literature - as I said before, applied sports science isn’t a very competitive field, which means that you don’t actually have to be very good to make a living doing it. Think local short track dirt racing vs. F1.)
As for use of ~1 h (or ~40 km) power as a proxy for maximal metabolic steady state in untrained individuals, the question is really moot, because very few truly untrained individuals would attempt such an effort. As I said, though, I have routinely coaxed 30 min of supra -steady exercise out of untrained participants… given the shallow slope of the intensity -duration relationship in this time frame, even that power would only be a few % more than maximal metabolic steady state. Conversely, it would only require decreasing the intensity a small amount to enable such individuals to go significantly longer. Of course, since endurance training seems to extend time-to-fatigue at “threshold”, this means that when tasked with doing a longer TT, trained individuals will be even closer to their true maximal metabolic steady state.
ETA: As I mentioned before, one of the simplest and most accurate approached to determining maximal metabolic steady state intensity is to just task individuals with generating the highest average power (doing the most work) over a sufficiently long duration. (Of course, no surprise here, since 1) by definition, the best predictor of performance is performance itself, and 2) performance reflects the integrated response to all physiological (and psychological) determinants thereof.)
9 Likes
I’m a scientist, and if you are a co-author of a publication, be it a paper or a book, you are responsible for the content.
“The question” seems quite different, all I wrote is that Coggan’s definition of FTP was a measurement of power at MLSS in a field test, and that it wasn’t one’s one-hour power. I did not dispute that you could use a 40k TT to infer FTP for some athletes, i. e. that it is one of several protocols for measuring one’s FTP.
It seems you came in late for the discussion. I don’t understand the context. The study I referenced does show that the majority of athletes with some training fall into 40–70 minute range, although only a minority among any of the four groups lasts longer than 60 minutes. Moreover, the averages increase with training, which is consistent with your experience: TTE can be trained and tends to increase with training.
As far as I can tell, this study is perfectly consistent with your experience and you see disagreement where there is none 
Scientific publishing is far - FAR! - different than writing for the lay public. Do you really expect me to have refused to cooperate with Hunter (and for the 3rd edition, Steve McGregor) simply because of a relatively minor disagreement over how to best estimate FTP??
I also don’t understand why you keep referring to me by my last name, rather than simply “you”. Do you not truly believe my identity? Would it help if I named my first publication? 
Finally, if your point is that FTP is not synonymous with 1.000… h power, I don’t know why you bothered to cite a study examining time-to-fatigue ( a HIGHLY variable measurement) at 95% of 20 min power. Instead, you should have just quoted any of the thousands of times that I (the originator of the concept) said that wasn’t true.
The bottom line here is that you seem to think that you are some kind of expert on the topic of FTP, yet I have seen no evidence that you really understand the concept or the underlying physiology.
13 Likes
I know, which is why I think your stance is counter to how I was scientifically raised. In my field the answer is yes, especially if it concerns such a central piece of the puzzle. Since an athlete’s FTP is the anchor point for Coggan’s power zones (and everything that follows from that), proposing a workable protocol for determining FTP would have been one of the central pieces of the book. I’d want that part nailed and based on solid scientific evidence. If this were about a more minor point, it’d be different. But if I were one of the authors, I’d want to nail that one.
I’d also say that this isn’t a book aimed at the lay public, it is a book aimed at putting scientific research into practice. Writing a text for people who are in between experts and lay people is hard, because you have to give up precision and write things that experts could nitpick until they arrive at an interpretation that is scientifically speaking partially false. I have never written a book aimed at such an audience, but I have taught them and written lecture notes for them. Getting the scientific fundamentals right was really important to me.
From what I understand (= not an exercise physiologist or similar), the exchange between the scientific side of sports science and the practitioners is crucial. Sports scientist crossing over to the practitioner side seems common and blurs the line between experts and non-experts. Many theories are further developed while putting them into practice and input from practitioners seems important. Calling those people lay people is not quite right.
Lastly, I would never badmouth co-authors in public, unless something egregious happened (e. g. scientific misconduct).
Indeed, I am not convinced that you are who you claim you are. I haven’t reached a conclusion either way to be honest.
If you are who you say you are, I propose you contact TR staff directly. Perhaps they could give you a custom tag indicating that your identity has been vetted. Just put yourself into the position of forum members and staff: what is the probability that the Andy Coggan joins an internet forum just to argue with random bozos like myself about his subject matter?
1 Like
Doesn’t everyone know about slowtwitch by now? Lol
10 Likes
Isn’t that a thing from the early 2000’s?
What’s the reference for those who don’t know?
27 Likes
I think the duration for validation intervals needs to be flexed based on your anaerobic capacity. Those with a big FRC/W’ probably need 20 minute intervals to confirm whether they’re actually at FTP or above it. This is also why they shouldn’t be allowed to set their FTPs via ramp tests or 2 × 8 minute tests even if they want to. 
3 Likes
Perhaps. 4 x 10 minutes suffices for me, but YMMV and I can see that some need/prefer 2 x 20 minutes at FTP. I have heard from others that they validate their FTP with over/unders. At the end of the day, do what works for you. 
My point is simply that you can make most ramp test protocols work, what matters in my opinion is that you validate it with suitable workouts.
That’s a good point, and I agree, this is probably a scenario where the results might be inaccurate. Ultimately, you gotta know where you are on the Bell curve for your preferred FTP test protocol. And if one doesn’t work or is too inaccurate, try others. Honestly, I am just sticking to ramp tests, because they work for me (validated, of course) and I am too lazy to switch. 
1 Like
FTP is not real! Same for elevation!
Listen to the pros 
7 Likes