🎉 🎉 🎉 Introducing Adaptive Training! 🎉 🎉 🎉

With regard to heartrate, will the system factor in whether the workout is outside vs inside? I know my HR is higher when it’s 90 degrees outside as opposed to inside with the fan even if the RPE is the same. Or would this metric be too “marginal gainy”?

No swim and runs yet, but we plan on adding that for the stress side.

You will get credit for custom workouts in adaptive training.


Yes, it will still work with “ad-hoc” plans. So plans added without plan builder.


+1 to this question for how FTP is being defined here


So did I understand correctly, and you can do just the first ramp test, and then the system estimates the progression and ftp levels with the workout data. Is that correct?

We’re trying to get polarized out within the next two weeks. So relatively soon.

We are also going to add some more volume levels of plans to fill in between low and mid and mid and high. That’s a bit longer timeline though, but still near term.


There’s some metrics and qualitative data that we hope to use to capture this stuff. There’s definitely more to capture and feed into the system.


@Nate_Pearson Amazing stuff!!! Any plan to incorporate sleep into the ML algorithm? I would love to input the amount of sleep I get every night so it can factor that in to my performance. I find sleep to have the biggest impact on my performance and would thus be helpful to capture


Will you release polarised Tri plans?


If being a horrible business person means making decisions that promote undying customer loyalty, then yes, yes you are! :wink:


Are there other objective ways beyond power that classify if a workout is a “pass” or “struggle?” I did Kaweah last night in the sustained power build and “passed” the workout, but it was like 10RPM below my normal cadence, and checking back on previous rides of Kaweah that seems to be the case. From the survey I would check it as RPE 4 and was interrupted once. Would the ML be able to make a determination about one of my energy zones from that and show something is lacking? Or is the survey supposed to cover that gap?

1 Like

Yes, and we want to run it into our ML system to see if it’s better for YOU!


Our models can detect frequency and urgency of clicking, yes. :upside_down_face:


Yes, but more money makes us go faster and get more done. And I think people would pay as I still think we’re the best value in cycling.

But…now at least I have some friends to drink with when I travel. :beers:


There are many ML features that track this stuff. :zipper_mouth_face:


I’m so excited, I can’t wait for being included in the Beta :grin:

1 Like

This is awesome. I feel like I’ve done a good job of adapting plans to fit my needs/life but can’t wait to see how this goes.

1 Like

Are you going to start bringing in weight data?

1 Like

Perfect! I’ll keep on proving the definition of insanity then :slight_smile:

Btw, Congrats!!!


This is a bit of an aside, but:

Are there any thoughts on opening up group workouts to not share the same workout, etc? One of the things I’ve seen is that since we’re all on plans anyway, it often feels like people are ‘giving up’ their standard workout to do the Group workout. As plans get more and more customized with this, that divergence may grow. Love doing Group Workouts and hope there’s a way for them to continue.

(Sorry if this was addressed in the live video, haven’t watched it yet.)