Adaptive Training - More differentiated survey response for each workout "failure"

Dear TR Team,

I’ve finally made it into the AT beta a few days ago (Yay!) and I’m already loving it :white_heart:

While the post-workout survey is a great and necessary tool to adapt future workouts, I believe that - especially when workout failures are automatically detected by the TR software - the survey responses could be a little more differentiated.

To illustrate, I’ve taken a snapshot from a pre-AT workout. Here, the first “failure” could’ve been a power outage (I’m living in a low-resource setting), the second failure was most likely where I was actually baked and for the third failure I don’t remember, but maybe I lost my bottle or had to get the door or something else :wink:

Due to the vast reasons for “workout failure” it would be great if users could 1. select the displayed failures (or the range, e.g. from 48:40 to 49:05) and 2. only then select the failure reasons from the given survey responses for each failure individually. Thereby, I believe that AT would get an even more differentiated view on the performance of each user.

Please let me know if I made myself clear enough, otherwise I’m happy to further elaborate. Looking forward to seeing what you guys come up with - please keep up the magnificent work! :muscle:t5:

based on your HR dropping instantly and climbing up again, the first marked point (but also the other 2) is definately a break :wink:
But as you mentioned… not all breaks are because your cooked, could very well be the doorbell, kids, toilet, etc.

In the end the specific reason doesn’t really matter as long as you can tell TR that it was not due to workout intensity :slight_smile:
I think it will be ot much hassle to specify which points are fails and wich are not. The only thing matters is that the workout intensity is OK or too high.

What I was missing was the option to give a correct reason for quiting a workout. I had to cut an easy recovery ride of 75 minutes short on 65 minutes, but I could choose a fitting answer:

I selected equipment, because that’s the answer that indicates no fitness related reason…


I can confirm! This is the biggest takeaway.
Will pass this along to the team that visibility/isolation of what event/pause triggers a ‘struggle’ could be cool, although not sure when that would make it onto the roadmap. Cheers!


There should also be “time” as an option, which I’ve used a couple of times when work interrupts a workout.


Great point. I was suddenly called out (work) during workout and had to cut it short. So it wasn’t a failure due to intensity fatigue motivation etc. I still marked it as “failed” though as I have not completed it but I’m concerned that it may muddle things up in a greater scope…


Really!? i imagine the reason selected would be helpful to feed into the boosted model.

In reality, the adjustment made for being stressed from work or having poor sleep the night prior will be pretty similar. Being sick or injured will currently yield a similar adaptation result, but in the future we’d like more results with variable responses depending upon severity. Just not there yet while AT is still in beta.

Interesting. Thanks for the tidbit. I imagine adding/removing selections isn’t an easy copy and paste as it effects other downstream processes.

I could swear Time was one of the options a while ago. @IvyAudrain Was that cut?

I think it was added between when I wrote that and now

‘Time’ is an available survey option, yes!

1 Like