Does TR AI assume I'll fail every not recommended workout?

TrainerRoad Software
1

Hi,

I fixed the hard days before my workout yesterday because they all look fine and achievable. Workout yesterday was a bit longer then an hour because outside I have to deal with roads and traffic :wink: Interval part was fairly similar as the prescription, and I rated is hard as AI expected.

Now my fixed workouts change to not recommended because of a 2.7% chance of failure for the first one. Fine, I can ignore that if I feel like I’ll succeed regardless. But that’s not my question.

I think TR AI assumes you’ll fail a non recommended workout (the 2.7% chance) which gives the next workout a 10% fail rate, … until I’ll not succeed a hard workout for weeks causing the FTP prediction to tank to 247W instead of 267W before my latest workout. Am I correct, and can I force it to assume the most likely ‘hard’ outcomes (45%)?

If not, I’ll turn this into a feature request :smiley:

image
image2369×912 233 KB

2

Hey @Mathieuv :slight_smile:

The failure rate will continue to change as you complete workouts and so will the AI FTP Prediction.

With AI FTP Prediction, it’s not just one thing that changes it, it’s the whole; plan adherence, how you’re rating workouts, the actual power data, future workouts which in your case it’s good to remember that harder workouts doesn’t equal to getting faster, it the progressive overload + recovery that dives adaptations.