It’s mentioned in several threads all around as the future solution for unstructured workouts and more. Still no timeline mentioned for when we may get it yet.
2 Likes
This doesn’t make sense. If we assume ai ftp is right then ignoring increases in ftp isn’t allowing you to get more time in zone as you are getting more time at a lower percent of ftp. (Just because you’re ignoring your ftp going up shouldn’t mean anything)
Though I guess this is when it becomes important to tell the difference between a ok change and an ftp change
I’d be very happy for it to be fully automatic, with the proviso that unstructured outdoor rides are taken into account. Until that point, I’d happily opt for the manual update and just spend a little more time working through the levels, as I’m doing now.
It doesn’t change it every time. It does a series of checks after a ride to see if we need to change FTP.
3 Likes
It think it does make sense, in that whenever your FTP is reset your PLs change, and so the AI will say "Aha! You had worked your way up to 3 × 20 minutes Threshold workouts, but there’s no way you can handle that at this new FTP, so I’ll knock you down to 3 × 12s.
Not what you want if you’re trying to work on your ability to work hard for longer. Out, then up. Out, then up.
5 Likes
Nice work guys.
A quick question (though I think I know the answer): As well as detecting any increases, presumably Auto AIFTP would also detect if you’ve been struggling (based on survey answers, reducing workout intensity, not completing workouts, etc.) and prompt you automatically if you had an FTP decrease?
2 Likes
Exciting times, Nate - thanks for the update!
Few thoughts;
I guess the acid test would be how sensitive in predicting change the model is. Too sensitive and those frequent changes may become demoralising for some users - I could see this particularly for those who may not fully understand the change is designed to ensure optimal training and time in ‘zone’ / PL optimisation (be that a change up or down in FTP).
I think this is sensible if going down the route of smart AI FTP detection
So, if I’m understanding this right (assuming the released feature covers Auto & Manual), the user would have the option to ‘toggle’ between either ‘Manual’ or ‘Auto’; if in ‘Manual’ mode, the Ramp test workouts would still appear - as it does today if using AI FTP Detection - but the user could effectively choose to either a) accept the AI FTP (and get a replaced workout) or b) conduct the ramp test? On the flip side, if in ‘Auto’ mode you’d never see the ramp test, it would always be a workout by default and FTP would just be auto set by AI?
Would the auto detection differ in the user message subject to whether that’s being set based on the start of a plan vs mid plan i.e. as a replacement for the ramp test vs how you’re performing / progressing mid plan?
One final point, it’s not a huge one, but maybe worth consideration;
I’m mainly a runner and thus still need to use Training Peaks to keep track of my CTL, ATL and TSB (for rTSS mainly). Keeping bike TSS inline with TR is straight forward when FTP changes infrequently however with Auto detection I could see this (possibly) becoming a pain if AI changes more often - I’m probably part of a small proportion of the user base that would be impacted by this so I wouldn’t expect this to be high on the priority list but just something to have on the back burner. Ideally, I’d be able to cancel my Training Peaks subscription if TR could get the the point where running TSS is calculated and added to the main calendar page where (think you alluded to this on a recent podcast) TR could have an equivalent of the PMC to track overall ‘fitness’ and ‘fatigue’ across multiple disciplines and (longer term) have that all feed into AT - this for me would make TR my single and only platform for not only training but overall load management.
As always, very much appreciate you and the team reaching out to the user base for feedback to support with the decision making around new features. It really sets TR apart from it’s competitors and I very much value that, so thanks, and keep up the good work.
Cheers, Paul.
2 Likes
Love this idea…but for small changes in FTP, rather not bother… let’s face it - if your FTP doesn’t end in a 5 or 0, your kidding yourself about accuracy! 
I.e. rounding to nearest 5 watts is accurate enough (especially with error % of power meters).
7 Likes
I voted auto only but honestly I may still do a ramp test every now and again. I enjoy the build up and anticipation of testing. (I even enjoy the fear and intrepidation ahead of each ramp test)
Edit: excited to see how auto works though and I have no doubt long term it will help me get faster
3 Likes
I think the better question is if ai ftp your current ability or what your current maximum should be? Current ability means if it’s lower for any reason, it will be lower (fatigue, feeling sick, break from training) so needing more recovery means you’re ftp could be going down. So kind of need the red/green functionality so it can give better context is saying it detects you are fatigued so will decrease your ftp and tell you to rest more. Otherwise you could see it as a reason to push harder and dig yourself deeper
2 Likes
Where are you seeing this? I’ve checked both the web and iOS versions of my calendar, and I’m not seeing this
Ive not looked yet but coloured annotation were a topic of the Android update (mentioned in the release notes) I downloaded this morning.
1 Like
I’d be happy to go for the auto full time, but I know some people on here really prefer having some manual control so I think you’d have a small riot on your hands if it launched full auto only.
1 Like
Luckily for me a while lately has been a week- week and a half so I assume that isn’t the trigger. But even in those times I may have opened the app a few times to manually put in a run or swim.
Would the the auto trigger come from either opening a workout or even possibly doing a workout again and not just opening the app after some time away?
Just thinking someone away for illness or injury that might check in every so often not wanting to see a pop up asking them to look at their FTP change, even if it asks before showing you they’ll still likely know it is likely lower and if it does that a few times before they get back to it that could be discouraging.
Between AT and AI FTP I’ve handled come backs in the last 18 months much much better than I ever have in the past, much easier to mentally accept a drop in power and also know it will feed me the right workouts so very much loving the improvements. But seems important to also handle those come back adjustments just one single time once actually back.
To your last point, the change is being detected after every workout, but, as I understand it, that change won’t be exposed to the athlete until the AI thinks that doing so would be A Good Idea. When that is depends partly upon the plan/phase.
In terms of progression, I guess it depends how you want to progress. Mostly, TR seems to default to Up, then Out. What you’re describing is Out, Then Up.
I can see how this is could be good for steady state efforts around sweetspot or threshold. No idea whether it’s more or less effective than Up, then Out. Probably depends on the athlete and their goals, and so it would be nice to have the choice…
It’s not going to work so well for over/unders, though, and many TR plans are o/u heavy.
All interesting stuff. 
I agree thats not what you want to happen, but this way you’re really working around the fact that the FTP + PL levels isn’t really a great representation of what your body can do as you aren’t updating the model when it has updates. Is it better then what came before and relatively close and able to use workarounds like this to get the workouts to be what you want? Yes. So not saying the current system doesn’t work. Just that it would be better if the numeric model TR has of your body would better reflect the current state of your real body without work arounds. Plus then the numeric model should then be able to predict what you can do. (say, help pace yourself on an outdoor ride without going too easy or too hard)
2 Likes
Yeah. Sorry. Could’ve been clearer. My bad.
Up = Watts
Out = Time duration you can hold those watts for
Obvs increasing either while the other stays the same would drive an increase in PLs.
2 Likes
Yeah, people are using that without definition, so here is what I take for that set of phrases:
-
UP then OUT = Increase FTP (UP) as the primary focus, then Increase Time in Zone (OUT)
-
OUT then UP = Increase Time in Zone (OUT) as the primary focus, then Increase FTP (UP)
6 Likes
Good points. What I’d not factored in (because I don’t ride outside) is the fact that you might have a regular group ride (say) where the duration stays roughly the same, and progress in that scenario is best described as “my time in zone needs to stay about the same, and my fitness improvement will either mean I can ride faster while maintaining the same RPE, or I can ride at the same pace with reduced RPE”
I can totally see how a plan progressed by adjusting intensity rather than duration would be better aligned to those goals.
I’m not saying how the training itself is done should be any different. Just that the model of a person (the data TR uses to say what kind of shape you’re in) should always be as up to date as possible. If updating that model makes the training suggested for you turn out to be wrong in the long term it would be better to fix the model, not tweak how you update the model. In the long term this would allow the ai to better understand you and better tweak how your workouts should be done as the AI would better understand everyone on the system and so could notice more trends.
For example person A has a FTP of 200 and PLs of 5 and person B is the same with the same training history. If the AI ftp wants to increase A to 210 but doesn’t because of their training plan, the AI would think these people are the same but they aren’t