After the workout you get a survey and I have often a new adjustment which you can except or skip. Can there be an explanation why the suggestion is made? It may help understand better what I am doing besides training with the new adjusted plan. Oke I can guess the next is a bit easier or harder. It is nice why TR makes the decisions for me. But some changes are just a little bit easier or more shorter interval instead of longer intervals. Hope I could explain this well to you 
You kind of answered your question. Based on your response to the survey and your workout performance, TR is adjusting your next workout to be harder/easier to ensure you get a workout that meets the planned goal and that the ML assumes you can achieve.
2 Likes
Oke but some explaination should be fine for me as an amateur. So if the suggested changes pop up a explain why is I think helpfull.
There’s a full explanation here
There are a few places to look for context to how Adaptive Training is moving you through your training plan, and one is in your Progression Levels.
- What you’re looking for might already exist to a degree, where you receive insight into what has impacted your changing Progression Level by hovering over each level on your Career page.
You’ll see the most recent workout that is impacting your PL, or in the athlete example below, why it has dropped:
- Another way to gain insight into how AT is moving you through your training is to review how the Workout Levels are changing in your pending adaptations.
This will give you an idea if AT is dialing you back based on your recent performance, or pushing you forward towards more channeling workouts.
- Lastly, reviewing your post-workout surveys can help provide some context. You can dig into some past workouts to see how difficult they were when considering pending adaptations. For example, if you continually found workouts easy, AT would be moving you up in levels more quickly compared to if you were responding in surveys that workouts were ‘very hard’, or if you struggled due to intensity/fatigue/etc.
Do these tools somewhat cover what you’re looking for? Let me know if there are more details you’d like and I’ll see if we can help address it.
@pbase sent over a great resource, and here’s another one with some more info as well: Adaptive Training Overview.
If you have any lingering questions, though, or need a second look at any of your adaptations, don’t hesitate to DM me and I’ll gladly take a look!
4 Likes
Oke here is the workout of today. It is partly in Dutch but I answered today ‘Very Hard’ and after that I have given a more less hard training. On Tuesday from 2.6 to 2.4 which is a little. But the Sunday is from 3.4 to 2.5 which is a bigger difference. So I am curiuos why
I can understand somehow that if you say very hard the suggestion is less harder next time. Hopefully the heartrate will be included because for me HR is the limiting factor. The explaination of @pbase is great. Thank you for the responses. The suggestions are for the better for TR.
1 Like
Another thing what I wonder for the suggested adaption is the AI of TR so smart that in combination with my answer of the survey the adaption is based on the true results of my effort? Sadly my HR is not considered only watts and hopefully the cadance. So if I say very hard TR could say well we think your answer should be hard, because so and so and so. I have a Polar and it suggests my sleep results as good or OK while my feeling says great.
When training is adapted, would it be possible to provide the user some feedback as to reasons? I like to understand the reasons for changes, purely for education and understanding. As an engineer the “trust the algorithm” removes some of the benefit as a user, since I love to learn how my body is adapting and how the training is adapting with it. Even a fairly high level of detail would be awesome, such as;
You didn’t manage to complete the last ?? workouts, so we’re adjusting ???.
You completed the last ?? workouts so we upping the effort ???.
I moved your post under an existing one covering the same topic / feature request.