🎉🎉🎉 Introducing AI FTP Detection 🎉🎉🎉

I disagree on this point. The absolute power numbers one trains at should take into account what you have planned. FTP input plus PL is what yields that.

If specific work is important in VO2 zone or sustained threshold zone to meet your goals, then the PL there is super relevant and the intensity setting to get into the right PL is important.

For other areas, it maybe isn’t as important.

I’d personally just like for the whole “percent FTP” to go away. Basically, AT does its analysis of history and goals and recommends a plan that sets your absolute power target ranges. In that scenario you can have dynamic workout levels for anything in the library really.

This is effectively what we are getting, but right now it is set in a discrete manner with a combination of FTP input, PL, and workout level. It could be - and I suspect will be - much more fluid as the system gets refined.

@hvvelo I’m going through the same debate. My hypothesis is that for a large portion of users this issue is reasonably self-correcting if users give honest feedback to AT and accept decreases in PL. And it is also most accurate relative to zones you spend most time in - so doing an FTPD and crossing zone may not be as successful as staying in primary zones focused on leading into the detection (though I bet it still gives a pretty darn good starting guess in those cases).

I suspect that the FTPD value returned is highly dependent on the type of training we are doing. For example, I am doing a ton of SS mostly 120 min workouts and basically have been hitting everything without trouble in that PL 7-10 range. The FTPD has been very aggressive with increasing my FTP input. I got two 3% bumps on day 1 and day 14. I apprehensively accepted the second one and went from SS PL 9 to 7.7. Today I did my first workout post FTPD - a 7.8 and it was fine (Golgotha). That’s great - here’s the thing, it would have equated to a level 10 workout at an FTP 5% lower (Wright Peak). But I have room to grow that SS work over the next 4 weeks of my plan.

My hunch is that if I was doing some other type of work where my PLs aren’t so high, FTPD wouldn’t be so aggressive. Could totally be wrong…

All that is to say that we shouldn’t take FTPD at face value at all. Look at the power numbers you are getting for the workouts and how those are progressing. One ought to know themselves well enough to look at that and see if it is realistic or not. If changing power zone focus, there is no shame in either manually dropping FTPD or telling the system it is too much through dropping intensity in the workouts.

Fundamentally the system is looking at - what have you done, what do you want to do, and how do you get an incremental increase between points A and B. And the FTPD is only one factor in tailoring your workouts to that in AT.

2 Likes

Yes, just to clarify, I’m pretty confident in doing either high PL at true FTP or low PL at the TR FTP.

I also realize my question reduces to more of a general training question of which aporoach is more likely to drive up a relative weakness or stagnation in true FTP threshold power.

But it links directly back to this PL positioning topic, and with this new calculated FTP, it’s another way TR is strongly pushing towards lower PLs with more aggressive FTP settings. I’ve only seen very brief comments from TR so far acknowledging there may be both pros and cons for each of the two approaches of high vs low PL - very interested in hearing more of their rationale on this direction.

1 Like

Yes to all of this. We are 100% on the same page I think.

I was laying out my thoughts - not correcting anything you said.

2 Likes

Finally got my chance to try it today after rounding out my rest week. My current ftp was 246 (coming back after many weeks with no bike) was hoping to see a number 260 or higher, and it recommended 260 exactly so I’m good to go with that. If it came back with something below 255 I probably would have ramp tested to try to beat it. Nice feature!

@trainerroad @ambermalika I don’t know if it’s a known issue or not. The scheduled alternative to the ramp test workout after detection is from a breakthrough group. The new FTP was a noticeable bump so it would be a really hard workout on the old FTP but with new one I don’t feel like doing 9.6 sweet spot at all. I’m going to do a CI adaptation (Carbon based organism Intelligence) and choose something from productive.

Hopefully it’s a correct thread :slight_smile:

You are not the first person to mention being given an incorrect adaptation here.

Just to add a data point my Goddard -4 was adapted correctly to Jane Bald which is a level 3.5.

Turns out I really don’t want to do the ramp test so now I’m going to accept an FTP that I think is too high… for science.

So i’ve gone from 313w @ threshold level 5.9 to 321w @ threshold level 4.6.

Next threshold workout is Warlow 4.7 - looks like its going to be hard - wish me luck.

For reference I think my real “tell my friends” ftp is 310w.

1 Like

Hello
I had an FTP test in my program yesterday.
Last test (a month ago) gave 253W, but for some reason TR did not catch this value and my FPT remained at 258W, and I kept it as I did not feel well during the 253W test. So I did last month trainings at 258W FTP.
Yesterday FTP predictor was predicting 256W, but I did the test anyway “for science”, and I hit a record for me at 264W. Let’s see how the next trainings will feel!

Pretty reasonable.

The numbers FTPD yield are similar to running pace calculators from Daniels and McMillan - those yield paces with effectively equivalent assumed fitness, but doesn’t mean you can do it. Have to actually train for it. That is a little weird in context of what FTP is supposed to mean, which is where confusion comes from.

1 Like

My bet is totally reasonable since the PL adjustments are almost always very good. There are certainly some workouts I feel less sure of than others, or days where accumulated fatigue require over-rides on AT suggestions. But I can always find a great alternative.

This is really interesting and an extremely subtle and underrated comment. It continues to point to the fact that TR “FTP” is not threshold power but just a stand-in for workout completion power (as TR staff have said several times on here). Sounds like some should take that to heart….

6 Likes

I have said it before verbatim (possibly in this thread) but so, so, so much virtual ink would be saved on this forum if it was called “Workout Calibration Level” or something instead of FTP. Hundreds, possibly thousands of threads would just evaporate like a Thanos snap.

1 Like

Should fall in threshold power zone. It’s just not necessarily a current Functional Threshold Power. It’s the first word that is problematic and causes confusion and debate.

Functional

Training

Power

:wink:

9 Likes

That’s excellent actually. Lock it in.

3 Likes

Make the acronym FTrP and it can have double meaning.

1 Like

I even considered pushing a bit further with something like:

Foundation(al)
Training
Power

…and other variants are likely possible with enough dictionary work.

2 Likes

When the dust settles and AI FTP / Adaptive Training / etc are all totally baked into the TR ecosystem it will be interesting to see what kind of onboarding experience they create for new users.

I think the onboarding experience will need to disabuse people of some traditional ideas about lactate threshold, etc, and really help build intuition about the interplay between Functional Training Power and Progression Levels.

Those of us who are in the midst of the rollout of these new features are kind of caught between two different paradigms and haven’t really received a holistic training on how they fit together.

1 Like

Didn’t happen after I accepted my new AI-FTP and I reached out to the support team for this (not only for this reason, but also for the fact that no adaptations were suggested after my FTP-bump, which results in an plan with only achievable workouts instead of productive ones during the next couple of weeks).