🎉 🎉 🎉 Introducing Adaptive Training! 🎉 🎉 🎉

[enki42]

I realize that, but it gives the upper limit. For example Xert seems to think I should do 2-3 hour workouts when I have no time for that. I’m assuming I have no clue how much I should workout or what days I should but I do now when I’m free. But then I’m pretty sure you answered that you plan to have TR know how much of that free time it should take up.

[Nate_Pearson]

@LucasKun

I’m kind of laughing at reading about FTP again, as personally I’m now so much more interested to see my fine-grained relative fitness, to see the small steps in each category. FTP just became a whole lot less interesting to me, and that feel like a very good thing.

I do understand FTP is still a global gauge, and therefore very very telling. And that a FTP increase/decrease immediately corresponds with a total rescaling of all little bars. But the fine-grained energy system progression will be so much fun/motivating that FTP as a single number becomes less relevant, and I think that will be very good. You could even stabilize your FTP (which tends to happen at some point) and keep playing with the energy systems. The new approach has so much potential for fun and inspiration.

Yes, this is our exact experience too.

I think the old system of waiting 4-6 weeks between ramp tests can lead to a bit of anxiety.

“Am I more fit?”
“Will I perform on ramp test day?”

Now you really know where you stand at any time. And if you bomb your test that’s OK and the system will just keep you going.

You could even score the same result on the ramp test after 4 weeks BUT be doing much harder workouts and still progress in fitness.

[RobertSims]

@Nate_Pearson

It will not. We ignore warm up/cool down.

Does that include if the cool down is extended to do extra riding/endurance (or even extra intervals)?

[enki42]

I hope the complaining half is just the more vocal half. Easy to complain but hard to really think about what is happening. (Like my responses up above aren’t meant to complain as much as trying to suggest my thoughts even if they aren’t good)

[Nate_Pearson]

@tofel

Could someone of the beta group tell me what happens when I shorten the breaks between sets, will it register as pass, super pass or fail?

This would be a pass. If you stopped in the middle of the set for an extended duration it would be a struggle or fail.

[tnordberg]

@Nate_Pearson

I might pee 2x during Baxter and it still says it’s a pass.

That’s because you’re still a long distance triathlete at heart and never leave the bike to do so.

[enki42]

This is why I never touch tri bikes, especially the saddle

[jmdcovas]

I don’t understand all the negativity and anxiety about the rollout of the new features. I think TR is being very transparent about it. This is a completely new system that as a lot of testing to do, so I completely understand they are taking time and being cautious adding people. Specially since most people here just want access to the features and have no interest in really being a beta tester, for which you need to be willing to use buggy and not final software, and report the faults to the developers to improve the product.

[Nate_Pearson]

@commanderobvious

On that same note, and yes I’m trying to ad-hoc AT prior to getting into the Beta. I check TrainNow daily before my planned workout and I’ve noticed that the recommendations are a bit more advanced than what my plan is showing. For example, it shows Spencer+3 instead of Mills for VO2 and Bays instead of Petit+1 for endurance. Am I being overconfident here or is this the AT working to rate the quality of my workouts thusfar?

At is working to rate the quality…but TrainNow is either “Achievable” or “progressive”. Your plan might give you a different workout on purpose.

I just checked your profile, it’s not giving Spencer +3 for you, correct? That looks to be too high for your current levels.

[Nate_Pearson]

@BCrossen

This looks so cool. It’s prescribing you workouts that I’ve literally never seen before, and I find that pretty exciting. I always get motivated when I see a workout I’ve never done, so if my adaptations also share a similar outcome, that will be most welcome.

Internally we also love this. It’s fun to do new workouts and we have so many of them that you could go a couple years and not get the same workout twice.

[dprimm]

@Nate_Pearson

“non-plans” won’t have it, but if there is a big outcry for it we can look into doing that.

Nate,

Figuring out how to include the Non-plans persons will help you get the world faster. Most of the world does not race - and doing a plan requires some sort of end goal. If someone can tell TR “I want to train 3x a week for one hour each, make me stronger/faster” you open to a whole new world.

I personally am not on a plan as I really hit the wall in November and am struggling to even get on the bike (Thank you for the Train Now rollout! that is helping when I get on the bike). Having AT just take me from where I am and help me get faster would be great.

Just something you have probably already thought of and have on your long term plan.

[Nate_Pearson]

@mattonabike

@Nate_Pearson A few of us asked about this about a million posts up in this thread and I think it got lost. Are there any medium term plans to add prospective data into the featureset such as ‘on a diet’? I could see some of your menstrual cycle data being something like this as well. It is one thing to tell the model a workout went badly because you were 750cal per day restricted for 2 weeks (or a woman on her cycle at x point, or pregnant, or in a house that just had a new baby) and it is another thing to tell the model that before a workout and let it make better recomendations. The tech related to this feels like its in the same wheelhouse of work you already have in place.

Yes. IE it would be good to know if you’re on Keto or a high carb diet.

We’d just need to build a system for diet with a start/end date for those who want to do this.

This is on the long list of things we want to add.

I suspect that keto diet people will have problems with threshold/sweet spot compared to high carb, but it will be interesting to prove/disprove that.

[Nate_Pearson]

@PWeber

Great - I do Phoenix variations regularly.

Will Tempo workouts ever be recommended in TrainNow under Endurance, if you have maxed out the Endurance Progressions for a timeframe or as a Progressive recommendation?

They aren’t in TrainNow at the moment and we don’t have plans to put them in to keep it simple.

[Nate_Pearson]

@mattonabike

One more for you @Nate_Pearson. Can you share what non TR data skynet currently consumes without giving away the secret sauce. I dont generally feed garmin with weight data today, but if the model can use it its trivial to do so. If i wear a garmin watch to bed, does the machine see that data? Calorie data makes it from myfitnesspal into garmin if you want, will it get used? I know its a dangerous place to give away to much, so just looking to know what data many of us could easily make visible to TR that would actually be consumed.

You’ll see the data that we bring it because you’ll need to authorize it for third party APIs. We won’t share specifically how we build features on that data, though.

[Nate_Pearson]

@Jkauffman

Nate, apologize if this has been asked.
I frequently do workouts when I am “on call” for work. As a result during a recovery period I will have to get off the bike for just a minute or two. So frequently a 4 min break will turn into a 5 min recovery ( i wont take the brake if the recovery is supposed to be less then a minute or two) . Also, when i backpeddle for a few moments during a recovery to catch my breath my trainer interprests that i have paused. Will these interruptions cause all of these workouts to be “failed”?? If so, i’m worried that my fitness will be underestimated.

It most cases it will not be failed. We trained the model on cases like these (short pauses). I got to the bathroom a lot during rides .

[Nate_Pearson]

@Craigball

Hi Nate
Wow A personal reply

I know you are thinking this but you are about to change training forever …

I use the type of software in the UK for pricing retail insurance and we look for predictive life time value to price both new business and renewals

I was a pro triathlete (once apon a time many moons ago) and now am a sociable cyclist looking for a balance from being CFO of a three thousand employee business

Anyway . Would love to have an offline with you on some of our key learnings as I really think you are creating a game changer

If i don’t here back the very very best of luck and continue to be an excellent CEO

Thanks! We’re actually hiring a CFO now but I believe you are above our level. We’re at about 90-100 employees depending on how you count open reqs.

[Nate_Pearson]

@trijdc

I am not in the beta yet, but I have been tracking all of these measurements via Oura Ring and HRVTraining for a while. Is there a place that I can record these in TR or provide it through some other means so that the data science team can start to collect this data? I know how important collecting the source data is, and I would like to help in any way I can.

Not yet, but keep using your products. We hope to pull in historic data from APIs. This helps us accelerate training the model.

[Nate_Pearson]

@bobw

Yesterday I had Kaiser+2 planned but subbed it for Kaiser-1, how would AT deal with that? I.e. still doing a workout in the same energy zone as that planned but either easier, as in this case, or harder?

It depends on what else you’ve done recently and what’s coming up next week. These two are pretty close, so your next workout next week might be a 7.5 instead of a 8.1.

But in the big scheme of things is most likely won’t matter much.

[Nate_Pearson]

@RobertSims

Does that include if the cool down is extended to do extra riding/endurance (or even extra intervals)?

We don’t use warm up/cool down to take into account how we categorize a workout.

We do take warm up/cool down into account in FTP prediction.

So you could extend your workout and kinda mess around for an hour and it wouldn’t make your threshold workout a “fail”.

But we would count that fitness towards FTP prediction.

[Nate_Pearson]

@enki42

I hope the complaining half is just the more vocal half. Easy to complain but hard to really think about what is happening. (Like my responses up above aren’t meant to complain as much as trying to suggest my thoughts even if they aren’t good)

We love the feedback! I just wish I could do it all at once. Please keep it up.

I’ll try to be open and honest about how we view features and priorities as a company. That of course won’t make everyone happy.