Sure it makes sense for TR, but it is still a shortcoming of their methodology.
For sure, I’m just thinking that including more all-out efforts would have to be part of some pretty radical changes to TR’s whole “thing”, so pending that I think FTP testing is still going to be a relevant part of the picture.
I’m quite similar (322 W FTP, 4.36 W/kg). When I started with TR I was already at about 3.8 W/kg through lots of rides and some random intervals thrown in. We should not forget that people like us are rare, though, and while we are not pro team caliber, we should be cautious when comparing us to others.
What do you base that on? That people without power meters can get very fit? That in the past pros got really fast? I don’t think power meters have been a marginal gain in training. Power allows for two things: it allows you to objectively and easily assess an athlete’s fitness. How did your x-minute power develop over time? What power can you put out on the 3rd or 5th sprint? And it enables a level of precision in training sessions that are impossible without it, because the relation between RPE and power is not fixed. That is why many other sports (such as running) are also trying to find ways to measure power, too. Training without power simply is way less effective, and I don’t know why you seem to take that personally.
To add one more anecdata point: to get to 3.7 W/kg, I did way more riding than I do now with TR (this was before I had kids). I also couldn’t hold power for as long. I’m much more in sync with my body, because I can connect sensations to specific power levels and understand the processes taking place in my body. And I always found heart rate quite useless as a metric, except when I try to stay in Z2.
Yes, but not as far as training with a power meter, especially when you compare fitness gains to training time. For Z2 rides you don’t necessarily need power, yes, but for everything else having power is really useful.
Why is this a shortcoming? Including more all-out efforts is less efficient training, because it reduces the amount of training you can do at a given intensity. Even sessions with all-out sprint efforts, which do not add a lot of extra TSS significantly increase RPE. That’s also why many people hate the ramp test, if done right, you are panting and completely exhausted (for a few minutes). I don’t mind the ramp test as much as I hate sprints in workouts, though.
If you aren’t doing maximal efforts, then how are you gauging results? That’s the whole point of any benchmark test, to evaluate where you are physically and base your training program from. End of the program, you do benchmark testing again to see if any improvements occur.
For people who race often, then you don’t really need to do benchmark testing. There will be ways to estimate threshold without benchmark testing.
If you don’t race, then you have to do some kind of maximal effort to get a sufficient benchmark. And maximal efforts at several durations will provide more info than MAP testing or a 20 minute TT alone will provide.
In my opinion, riding hard and generating regular maximal efforts at various durations is one of the keys to improving and benchmark tests, races, and TTs are all a part of that.
You are conflating assessment with training. TR’s ramp test, which is the default assessment, is an all-out effort. But it is a short, sharp effort where you typically spend < 10 minutes above FTP, so it doesn’t cause a lot of fatigue.
Regular training is deliberately below all-out, for otherwise you couldn’t keep up that kind of intensity over a longer period — you simply didn’t have enough time to recover.
You need to balance that with two things: how much fatigue are you incurring because of the assessment? And what actionable information can you glean from the extra information that you get? TR’s solution with AT is to assess your fitness at different without having to test several different power levels explicitly.
Something like a 4D power test sounds much better than a ramp test that determines MAP, but the question is whether you really gain more actionable information? That is, what do you do with the data points? I’d say the answer is a very clear “That depends.”
No, it isn’t. Why do you think most riders finish and feel like they could’ve done more? It’s a proxy for max aerobic power, but you are confusing that with an all out effort. It’s an all out effort at that particular protocol, but it does not ever give you your maximum power output for any of the tested durations.
And yes, you gain a whole lot of actionable information by routinely testing MMP at different intervals. You gain race-specific information, in fact. If you know the decisive climb in your event is going to take 2 minutes, or you want to win a crit with a last lap flier, it is extremely informative to know how much power you can sustain for those durations and train it accordingly.
In addition, one of the best ways to break plateaus is to go as hard as you can for specific durations that are relevant to your goals, and then recover from those efforts.
The answer, obviously, depends. But I doubt there is any coach out there worth his or her salt that never prescribes perodic maximal efforts to their athletes, at least not to athletes with competitive goals.
I’m not confusing that with anything: all out efforts depend on the type and duration of the effort. A 3-hour all-out ride will be at a much lower power level than a 15-second all-out sprint. A ramp test is all-out by definition, because you are supposed to stop when fully exhausted.
Now you can say that most of us still have a bit left in the tank, as in if I put a gun to your head, you’d be able to hang on longer. That’s true of all all-out efforts, though, because our body wants to protect ourselves from ourselves. These mental barriers are part of the limiters you currently have.
Assessment ≠ training ≠ setting goals.
I’m talking about the utility of something like a 4D power test as a regular assessment. 4D power tests have the most utility for people who don’t know their own power profile. But most of us do, after all, TR and TP have power PR charts and we know from riding where our strengths and weaknesses lie. Once you know your power profile, you don’t need to re-test very often if at all, because this changes very slowly over time. And I don’t think you need to re-test: TR workouts include all-out sprints that range from 10 to 30 seconds. Also for some other efforts I have some climbs that I can use to gauge my current fitness.
The example you give of a decisive 2-minute climb in a race is something that concerns goal setting, not training. And this is needed as an input for TR. Do you want to accentuate your strengths or mitigate your weaknesses?
You need to set goals on your own, this is needed as an input for TR. It provides quite a few training plans depending on the types of efforts you want to be good at. And if you are training for an A-race with specific demands (like a 2-minute climb), then you train for that mostly in the specialty phase. You have to come up with a strategy for that race or season in advance, and this has to be decided before you pick a training plan. This is then something you need to input into TR. So if the most important thing for you is short power, go for a crit plan that includes short power build.
During training you can assess your progress by having a look at your FTP and your VO2max progression level. When using AT there is no reason to test this for the purpose of adjusting your training at this point in time. You seem to claim it would. But this is something that the TR team has tried in the past (probably several times) and it did not work for them.
We are talking past each other. If you’re happy with TR, that’s great! TR is a great training value. Best of luck to you!
I don‘t take this personal and I think we agree to 98% but just are splitting hairs (well, maybe the latter is just me 
)
Power absolutely is great to monitor progress the only aspect where we differ is that I think you don‘t need to pin point exact numbers because you hit the right (endurance-tempo-threshold-sustainable/repeatable 3minutes) zone fine enough by feel.
What made me think though was the comment by @ havetodigthatout : Power might be essential to even develop that feel in a few years rather then half of a lifetime.
So yes, maybe I was wrong after all and the close observation of rpe/hr/power for 3 years made me feel like I could have done it without power just because I can now (on some rides)
Our whole discussion is a little off topic in this thread, so sorry for everyone else, but I might have learned a thing.
Individual response to training. It’s not worth debating, I hold an opposing opinion based on my own results.
TrainerRoad has always tried to do things “at scale” and the short-term goal of AT in my opinion is to adjust workouts to rider fitness/experience. The workouts themselves are based on classic Coggan levels and scaled to FTP. But the number of intervals you should be doing is influenced by your current fitness, experience and ability to recover, among other things.
TR solved the individualized levels for VO2max problem by creating a larger library of workouts with a wider range of % FTP, and the fitness/experience problem by having a wider range of # of vo2 intervals.
Vo2 a weakness? Less intervals and at a lower % FTP. Vo2 a strength? More intervals and at a higher % FTP. That’s what AT is doing, helping pick appropriate intervals for those that just want to be told what to do.
Because before AT, the only dial was low vs medium vs high volume, and the intervals were a better fit for stronger/experienced riders resulting in (predictable) issues trying to follow the plans. Some ignored the forum calls to HTFU and were happy doing appropriate work for current fitness and fatigue, even when others would call it a compliance fail.
AT simply tries to institutionalize the reality that one size does not fit all, and there should be no shame in calling 4x3-min vo2 intervals a success if that’s the appropriate stimulus for your level of fitness.
Then they didn’t do it right? That last minute is supposed to be as hard as you can possibly go. E.g. if using a smart trainer you should erg mode death spiral to failure.
That’s a very, very interesting point that hadn’t occurred to me. 
It certainly is compatible with my own experience with power meters, I think I have gotten much more attuned to my body.
I’m confused: I see nothing in your post that is in opposition to mine, I completely agree with what you wrote in your post.
It is an all out effort for that protocol. It is not a mean max power effort and has no practical utility for racing/riding besides estimating FTP.
For me, more all-out efforts is more efficient and bigger gains. And I hate the ramp test and favor sprints in workouts.
Out of curiosity: what do you consider all-out efforts in this context?
For me the only workouts that included all-out efforts for me were ramp tests, workouts with sprints and extreme VO2max workouts (think 7 x 3 minutes at 120–122 % FTP). With most workouts I have plenty of gas left in the tank. These are efforts that push my physical limits. The other workouts tend to be mentally hard, but I know I have something left in the tank.
Sure. I think every athlete has a power zone and workout type that they like or particularly dislike. My point was that I reckon one of the reasons many people dislike the ramp test is that it is an all-out effort, and that all-out efforts tend to increase RPE significantly.
“Mean max power” means nothing unless you state what the duration you are considering is. At least looking at my power PR curve, ramp tests get close to or are responsible for my 3–5 minute power PRs.
And not one of those 3-5 min power PRs from your ramp test represent what you can actually sustain for those durations. That’s exactly the point I am making. A ramp test is an all out effort for that particular protocol; it does not in any way provide a maximal effort for any specific duration whatsoever.
At least in my case that’s not true, the ramp test is a pretty good representation of my VO2max power that I can achieve in 3–5 minute all-out efforts outdoors. I have a small hill where I live, which takes me about 4:30–5:00. My PR is pretty close to the power I make during a ramp test for the same duration. In fact, because the PR for that hill is about 4:30 minutes, so my 5-minute power PRs are invariably due to ramp tests. (Other hills that I ride regularly are a bit longer and later on in the ride, so I tend to be more conservative with my efforts.)
That’s why I was specific about the duration: if I know I need to pace for <3 minutes, I can put out more power, yes. But for 3–5 minutes it seems spot on, about 128–130 % FTP.