I have been using TR off and on to help with structured training but I still don’t understand how I can go and do a three day Stage race 3 hours a day and it doesn’t seem to impact on progression levels.Same for long group or solo rides in fact the new AI seems to think doing these rides will reduce future FTP improvements. What am I missing this seems counterintuitive as I should be getting fitter and faster by doing longer and harder efforts so long as my recovery is good?
Not TR workouts don’t get progression levels as you stated. They do count into your training and AI counts those workouts. It just doesn’t rate them. I actually think it is fine this way and ignore the levels for endurance ![]()
The way I figure it, AI creates what it thinks is the optimal plan and uses that to predict FTP. If you deviate from that plan, going either harder or easier, what you’re doing is less than optimal so predicted FTP will go down. That’s how it seems to be working for me.
I wonder if it’s not interpreting outdoor rides well because I don’t have a power meter so it would have to go off gps data eg meters climbed speed and heart rate to gauge the intensity. Is this possibly the issue?
Progression Levels should just be understood as a representation of the fitness delivered by recent structured workouts.
We don’t use Progression Levels to guide TrainerRoad AI’s prescriptions, so it really doesn’t affect anything adversely. Your unstructured rides are analyzed by TR AI, but Progression Levels are not assigned from them.
Getting accurate numerical levels from unstructured rides is a uniquely complex problem. We’ve aimed to solve it in quite a few ways, but we try to keep the main thing the main thing here at TR. At some point while developing TR AI we realized it could still analyze them and adjust your training appropriately, so we stopped dedicating resources to getting numerical rankings for each zone based on unstructured rides so we could put all of our resources into having TR AI give you the best training possible.
We may do this in the future, but it isn’t a priority for now since it doesn’t harm training prescriptions.
FWIW, I can’t remember the last time I looked at my Progression Levels, hehe. I just let TR AI give me the right workout and it feels great!
Given that PLs don’t influence AI prescriptions and exclude unstructured rides, should cyclists still consider them a meaningful metric? Or are they likely to be deprecated or replaced by a more holistic AI-based progression model in the future?
IDK how the new AI works for unstructured rides without a PM. I use PMs and understand that the new AI utilizes that and HR to evaluate unstructured rides and adapt the ongoing plan appropriately. That what it seems to be doing in my case.
They’re meaningful for many athletes as a visual high water mark for recent workouts, and as a motivator to keep pushing forward and seeing the number increase. Like Jonathan, I don’t look at mine too often now that TR AI takes care of anaylzing and progressing my workouts. I do use them from time to time as a quick reference point to see where I’m at with my training in each zone, and to get a quick idea of how the difficulty of one workout is likely to compare to others.