Firstly congrats on the FTP increase. Always good news.
I think AT has a built in rule that states that after a change of FTP, whether up or down, adjust levels downwards. That makes sense as either you did well and the new levels will make your workouts harder anyway. If you went down then clearly the workouts need to be easier.
Given what you have been given now I would retain them and do the recommended workouts and see how quickly they jump back up again.
I think this a bit like training plans, you have to trust the process.
Anyway thanks for taking the pain for us not yet on the beta.
Yup. I understand and agree on the general direction of a downward adjustment.
My current issue is the severity of the change.
My Endurance was cut nearly in half and VO2 Max & Threshold far more than that. Everything else cut back to 1.0 from different points.
I just donāt see any rhyme or reason to the amount of change, especially when considering the FTP change was a mere 2% delta.
Nonsense reductions in my eyes, when considering the small change in FTP. If this is intended behavior, it better come with some explanation, because I canāt see any reason for the amount changes applied.
I am considering this, but admittedly hesitant. Pending races and such are leading me to question a progression that is far slower than I think it should be.
Essentially, I donāt want to penalize my actual progression from what may well be bad programming via the beta. I am fine with the process and helping out, but I donāt want to sacrifice my training as a result.
Maybe my priorities are off, but I have actual goal events outside of helping TR, and I am seriously concerned by what I see right now, and the limited info I have on whether this is āright or wrongā.
Yeah, but before getting into AT, I spent the last two years modifying the default plans to suit my actual abilities. I changed the work and recovery week layout to allow me to complete the prescribed workouts the majority of the time. A necessary change since at 45yo when I started that experiment (and 47 now), the default plans force me to failure without my mods.
I got a good taste of how positive AT can be with my first 3-4 weeks using it a while ago. These did make sense and I followed them with good results. But recent changes seem to have put the process steps backwards and the new changes and suggested adaptations no longer make sense. So itās hard to follow āa leaderā that I now think is lost and aimless.
Not sure if this is normal behaviour or a bug. My VO2 level was 1.1 as I had been doing SSB and my efforts over threshold had been on Zwift or outside.
Despite this my first VO2 workout set in my plan was an 8.1 and the system even said āNot recommendedā!
So I failed the workout today but completed a good proportion of it at full power or between 5-10% lower. Despite this the system has given me no increase in level, I assume because I failed. So I am still at a 1.1 level despite completing half of a 8.1 level workout!
Welcome to beta testing an applicationā¦thereās a reason it isnāt released in Production - you and the other people in the beta are dealing with these things so they can resolve them before an actual releaseā¦
Didnāt say I had a problem with it, just that until it starts making sense then I wont be trusting it.
Practically speaking Iād argue a ramp test bump from 261 to 266 is basically within the noise floor of power meter accuracy and day-to-day body fluctuations.
And even if you did have a minor bump, it was such a minor bump that basically reseting most of the progressions is clearly a bug.
Totally agreed on the potential for variance. Trainer accuracy for one, my room was cooler this time (which makes me perform better in general), and who knows what else. So even if I had a wash with no FTP change, the magnitude of the change I got in PLās seems just plain wrong.
When it comes to levels after a FTP change it should arguably be relatively linear in my opinion.
For instance, if youāre doing 250w on sweetspot intervals before an FTP test then thereās no real reason you should be suddenly doing significantly less than 250w on sweetspot intervals after an FTP test, assuming youāve gained in FTP (even slightly)ā¦ and even if youād lost FTP, thereās an argument that you were doing those intervals OK before anyway so no need for big changes.
Now I donāt know what levels that aligns with for something like a 5% FTP increase, but I very much doubt it would take a 5 down to 1.
With that in mind, itās looking quite likely Iām going to skip the FTP test next week. Levels are working OK for me at the moment (VO2max issue aside), so I donāt really want it to mess them up.
Yikes. I went from anxiously waiting to get into Beta to wanting no part of it until it is a smooth flawless running machine. Like you, after years Iāve finally reworked the TR plans to suit my needs and abilities and Iāve never been fitter consistently hitting all time PRs this season. Iād be afraid to throw a wrench into the system with a dysfunctional AT. Iāll check back in with AT at the end of this race season.
I was getting frustrated that Iām still not receiving email with the AT inclusion but looking at some of the above comments Iām happy to continue with just plan builder!
I should have clarified that I meant ātip of the icebergā in terms of frustration for users. I think the TR team has been particularly transparent regarding expectations, and their judgement with respect to widespread dissemination has been spot-on. I canāt wait for AT, but Iām more than willing to wait for a polished product.
This is one of the aspects of ML that is challenging in all areas - we basically have no visibility as to why a specific decision is made. This holds true of training adaptations, as well as for when a self-driving car does something stupid. There is no algorithm to debug, just a black box. You can make a new black box with different training data, and make sure it handles the known weird cases, but thatās about it.
Exactly, Just have faith train and see the results if they become better from the time we train without AI.
There has been very few weeks to know buy I believe it is great.
To add some balance to this, the same attributes that make it hard to debug also make it very powerful - ML can find and match patterns that we otherwise wouldnāt know existed, and can make (potentially valid) associations between things that we donāt have a mechanistic or algorithmic understanding of. A few comparisons have been made with Xert, which is interesting as well, but from my understanding is algorithm based, rather than ML. This means that Xert will be easier to debug when it produces strange results, but it will be harder to do new things, as people need to understand the relationships and code them into an algorithm as opposed to having ML look at all the data and make the associations itself.
This is a vast oversimplification of course, but I think is a reasonable description of the tradeoffs between ML and algorithmic methods.
