Two weeks into the 8week high-volume polarized plan now. Recognize that AT is not supposed to be working with POL yet, but left it on to see what it recommends.
So far, AT seems to be recommending appropriate changes for VO2 and threshold workouts. I have approved all recommendations that have come up so far as they make sense based on my performance and how I answered the surveys. AT seems to work well for these specific workouts.
AT does not work as well for suggesting changes for endurance workouts. It is maybe better than what it sounds like it was a few months ago based on reading old comments (ex: some mention it would start subbing in SS workouts). However, it still often makes seemingly non-nonsensical recommendations. For example, doing a 1.5h z2 workout, rating it “easy,” and then having the recommendation come up to switch upcoming 1.5h Z2 workouts to easier variations. It also struggles a bit with it’s recommendations maintaining a logical progression in weekly total time of Z2 work, and in keeping appropriate progression of the weekly long-ride, though I wonder if these are related to a smaller library of Z2 rides for it to draw from.
As it stands, I think I trust AT to make appropriate changes to the VO2 and threshold workouts in the polarized plans, but ignore everything it says for the endurance workouts.
Once I finish my off season weight lifting plan and Traditional Base I MV I plan to try the 6-week POL Build plan MV. For the endurance rides how easy should they be? I was thinking most should be just below VT1 / LT1 (in the three zone model) but I tried doing some of that during TB and found I still needed an easier ride now and then. For me, VT1 (or LT 1) is right around 73% FTP.
Interesting observation. I’m currently doing the 6 week Polarized and it HAS been upping my Endurance levels after most rides. Same as you, I’m marking them Easy, except for the 2+ hour ones, which I’m marking as Moderate.
I wonder if the difference is the 6 vs 8 week, our starting levels, something different?
Part of the problem is that the surveys really need anchoring statements for their Likert scales, but that is another issue…
I thought some of it may be starting levels at first, as my PL for endurance at the beginning was 1 (according to TR), and AT using some sort of Bayesian method that put too much weight on the prior probability based on the PL, and not enough weight on survey responses. However, then I started to get downgrades in my rides despite rating them as easy, after these same rides had already been upgraded based on rating other rides as easy… so who know.
I’m doing long distance tri base (planned with plan builder) and I’d echo your thoughts about endurance adaptations.
Vo2, tempo and sweet spot adaptations seen quite sensible but the endurance suggestions are hard to understand eg subbing 3hr ride for similar level shorter rides.