Just wondering if the idea is to start the intended workout as scheduled, but decrease the intensity by 5% or maybe just 2% ?
Could be. I’ve actually never decreased the intensity of a workout in all the years I’ve used TR! Maybe I should do that. ![]()
Could be a solution for when the workout is too high a chance of failure, but you like the format so want to do it anyway.
Now that the ML is looking at actual watts, & the achieved WL is decided by that rather than a pass/fail on the level stipulated in the original workout at 100%, it looks like it doesn’t matter if you turn it down, you just get a lower score.
It’s not a technique that I’ve used before a ride, but I have occasionally used it when I’ve completed a few intervals and realised that I’ve bitten off a bit too much.
But if the new AI is only interested in watts then I don’t know why this wouldn’t work? So stick with the schedule but dial it down slightly. Not sure how I’d record that at the end of the ride though.
The advice TR gave me was “rate the work you did, not the work you were assigned”
After playing around with the calendar this morning, one thing I would like would be for the Ai to choose what kind of workout.
It is great that I can add an Ai workout to schedule in, but then I have to pick zone or purpose.
I already have a training plan with a goal, why cant Ai pick the workout for me based on previous training and plan goal?
AFAIK, Train Now no longer exists.
Trainnow is based on upon what it recommends I do now though.
If say, ive deleted a zwift race in 4 days time and want to add a new Ai workout in its place, I still have to make those choices.
Even if I used trainnow and scheduled it 4 days in the future, it still gives me three options albeit with one recommende.
On browser:
Career > Ai Workouts > Trainnow
That was for Pbase, not mrtopher ![]()
Thanks! Weird that you can’t get to it from the calendar.
Integration with Health Data (Apple Health/Google Fit) for Adaptive Training?
Hi everyone,
I’ve been using TrainerRoad for a while now and I’m a huge fan of how Adaptive Training adjusts my workouts based on my performance. However, I was wondering if there are any official plans to integrate Health Data (like sleep quality, HRV, and daily activity levels) directly into the AI.
Currently, if I have a terrible night’s sleep or a very stressful day at work, the system doesn’t “know” it until I potentially struggle with a workout or rate it as “Very Hard”.
-
The Question: Is there a roadmap for syncing with platforms like Apple Health, Oura, or Garmin Health to allow the Red Light/Green Light feature (or the AI in general) to be even more proactive?
-
The Goal: Having the software suggest a recovery ride before I even start a session because my recovery metrics are down.
I’d love to hear from the TR team or if anyone has found a good workaround to “feed” this data into the system manually beyond the post-workout survey.
Thanks!
#feature-request #training
Update the wording about Pauses. To highlight that excessive pauses “MAY” lower your score. So that there aren’t weekly posts about this when the person’s score wasn’t penalized.
Here is the last time I did Cloudripper:
I didn’t get Anaerobic credit but did get V02Max and no bonus credit for Threshold.
Hi TR
Could you use historical training data to show us how we are improving?
I took a couple of data points to try this out, using workouts I have repeated over the last year. Here are the results.
Endurance power +6%
Aerobic efficiency +10%
Aerobic threshold +10%
I’m sure you could get better insights by using more data
Some simple historical performance metrics could help athletes understand their progress and make sense of future predictions
Can we please be able to look up historical heart rate data?
Great idea.
I would love if TR supported survey response/failure rate predictions with custom workouts.
I wish TR would add integration of workouts with MyWoosh.


